食品安全突发事件跨媒体信息的语义分析与分类研究
|
摘要
目前,我国食品安全事件频发,互联网上相关信息迅速增多,其数据形式呈现跨媒体的特性。图像作为该类数据的主要成分,由于其包含语义的丰富性,成为大众获取相关信息的重要来源。虽然传统的语义分析、分类技术在图像的语义理解上取得了一定进展,但对如何更加有效地进行图像量化表达以及图像特征选择,仍然有很多问题亟待解决。在食品安全突发事件语义背景中,相关跨媒体信息拥有特征空间异构、语义空间密切关联的性质,如何利用具备多模态表现形式的跨媒体信息,挖掘不同模态数据问的潜在语义关联,以及在不同模态数据间进行知识迁移,从而帮助进行图像的语义理解,这对语义分析与分类任务提出了新的挑战。针对图像表达及特征选择、潜在语义关联关系挖掘、异构数据源间的知识迁移等问题,本文从图像的量化表达、语义标注和语义分类等三个方面,对食品安全突发事件跨媒体信息的语义分析与分类技术进行了研究。论文的主要贡献和创新点如下：
(1)本文提出了一种距离优化的特征压缩降维算法(DO-LLE),解决了由于特征点分布不均衡造成的近邻点参数难以确定和降维效果不稳定的问题。针对视觉词汇本的大小难以确定的问题,本文提出了一种自适应特征聚类算法(AC),该算法根据最大最小距离规则和DBI指标对特征集合进行迭代计算,自适应的生成最优聚类中心。将DO-LLE算法和AC算法在LPS数据集(Lazebnik Schmid and Ponce, LPS)上通过图像分类实验进行验证。DO-LLE算法的平均分类准确率为81.9%,比传统的LLE算法高出3.3%。AC算法可以自适应的确定特征聚类数(聚类数为298)其平均分类准确率为82%,比人工设定聚类数为100、200、300、400的平均分类准确率依次高出10.7%、5.4%、0.3%、1.1%。实验结果表明,DO-LLE算法和AC算法可以有效地提高图像语义分类准确率。
(2)为了解决对图像进行有效量化表达的问题,本文提出了一种基于视觉词袋模型的健壮的图像表达方法(BOVW-RIR)。为了增强图像底层特征的表征力,将图像加速稳健特征和多分辨率直方图矩特征进行了融合,生成了一种具有图像局部信息和结构信息的底层复合特征。依次对特征进行压缩降维和聚类,按照特征聚类结果构建视觉词汇本,并用该视觉词汇本对图像进行表达。BOVW-RIR方法在OT数据集(Oliva and Torralba, OT)、FP数据集(Fei-fei and Perona, FP)(?)口LPS数据集上依次取得了89.1%,83.9%,82%的平均分类准确率,在20类食品安全图像数据集上取得了70%的平均分类准确率。将BOVW-RIR方法与目前三种著名的图像表达方法(Fei-fei, Bosch, Lazebnik)进行了实验比较,BOVW-RIR方法在OT、FP和LPS数据集上的分类准确率比三种比较方法中的最优者分别高出了4.4%,6.6%和7.3%,这表明BOVW-RIR方法能更加有效地对图像进行量化表达。
(3)本文提出了一种基于潜在语义主题加权融合的图像语义标注模型(LSTWF-ISA),该模型对训练数据的语义标注词和图像视觉词汇进行了潜在语义主题建模,得到了文本模态数据和视觉模态数据的潜在语义主题分布。利用计算视觉词汇分布的信息熵所得出的权重参数,对文本模态数据和视觉模态数据的潜在语义主题分布进行加权融合,得到融合语义主题分布,利用融合语义主题分布构建了LSTWF-ISA图像语义标注模型。本文提出的LSTWF-ISA模型在Core15K数据集的49个最优标注词子集上的平均F度量值为0.71,在常用的263个标注词子集上的平均F度量值为0.22,在20类食品安全事件的图像标注数据集上的平均F度量值为0.36。将LSTWF-ISA模型与四种著名的图像语义标注模型(TM,CMRM,CRM, PLSA-WORDS)进行实验比较,并对结果进行符号秩检验。LSTWF-ISA模型在Core15K数据集的49个最优标注词子集和常用的263个标注词子集j二的平均F度量值依次比四种比较方法中的最优者提高了11%和29%,这表明LSTWF-ISA模型能够通过利用文本模态和视觉模态的潜在语义关联提升图像语义标注效果。
(4)本文给出了文本-图像共现数据的形式化定义,描述了基于事件约束的文本-图像伴随文档的特性。提出了一一种文本-图像特征映射算法(TIFM),基于文本-图像共现数据从文本特征空间向图像特征空间进行特征映射。TIFM算法在包含20个类别的食品安全数据集上所计算出的图像特征分布与基准图像特征分布的平均欧氏距离为0.024、平均余弦相似度为0.84、平均K-L分歧值为0.075,这表明TIFM算法能够有效地将文本特征分布映射为图像特征分布。
(5)为了解决利用文本数据帮助进行图像语义分类的问题,本文提出了一种基于特征迁移的图像语义分类模型(FT-ISC)。针对文本特征的海量性和稀疏性导致不易计算的问题,本文提出了一种基于信息增益的文本语义特征选择方法(IGTSFS)。该方法对文本数据进行潜在语义主题建模,计算出各潜在语义主题的信息增益,得出显著文本语义主题特征。本文提出的FT-ISC模型在食品安全数据集上取得了76%的平均分类准确率,在“小龙虾”食品安全语义类别上取得了86%的分类准确率。将FT-ISC模型在食品安全数据集上与贝叶斯分类模型和标记查询分类模型进行比较,实验结果表明,FT-ISC模型在分类准确率上比贝叶斯分类模型和标记查询分类模型依次高出8%和5%。
(6)本文提出了一种基于特征加权的图像语义分类方法(FW-ISC),该方法通过将筛选式的图像特征加权机制(FFW)与支持向量机结合,实现了图像语义分类。在特征加权过程中,针对特征分布不均衡以及特征与语义类别之间的紧密度难以度量的问题,提出一种基于条件互信息的特征差异性度量方法(CMIFDM),通过迭代计算得出某个语义类别下特征的权重值。通过设计特征加权核函数,将FFW特征加权机制与支持向量机结合,在分类器训练过程中可以利用特征加权的结果。FW-ISC方法在LPS数据集上取得了87%的平均分类准确率,在食品安全图像数据集上取得了75%的平均分类准确率。将FW-ISC方法与传统的支持向量机分类器进行了比较,并对实验结果进行了符号秩检验。实验结果表明,FW-ISC方法在LPS数据集和食品安全图像数据集上的分类准确率比传统的支持向量机分类方法依次高出5%和8%,这表明FW-ISC方法可以有效地提高图像语义分类准确率。
At present, China's food safety emergencies happen frequently, which results in a rapid growth of the relevant information on the Internet. The form of these data has the cross-media characteristics. Images, as the main part of such data, become an important source of the relevant information for the public due to their richness of semantics. Although the traditional semantic analysis and classification techniques have achieved certain progress on semantic understanding of images, there are still many problems to be solved in the aspects of the quantitative representation and the feature selection of images. The cross-media information with the semantic background of food safety emergencies has heterogeneous feature spaces and closely related semantic spaces. How to use these cross-media information with a multi-modal form to mine the latent semantic relationships, as well as to transfer knowledge between different modal data, proposes new challenges to the semantic analysis and classification tasks. Thereby, it could help the understanding of the semantics contained in the images. Consider the problems of quantitative representation and feature selection of images, mining the latent semantic relationships and transferring knowledge between heterogeneous data sources, this dissertation studies on the semantic analysis and classification techniques for the cross-media information of the food safety emergencies from three aspects such as image quantitative representation, semantic annotation and semantic classification. The main contributions and innovations of this dissertation are as follows:
(1)This dissertation proposes a Distance Optimization based Locally Linear Embedding (DO-LLE) feature dimension reduction algorithm, which could solve the difficulty in determining the neighbor point parameters and instability of the dimension reduction results. Regarding the difficulty in determining the size of the visual vocabulary, it proposes a feature Adaptive Clustering algorithm (AC). The AC algorithm could adaptively generate the optimized cluster centers through the iterative computation on the features set according to the max-min distance rule and the Davies-Bouldin Index (DBI). The DO-LLE algorithm and the AC algorithm are verified via the image classification experiment on the Lazebnik Schmid and Ponce (LPS) image classification dataset. The DO-LLE algorithm gains an average classification accuracy result of81.9%which is3.3%more than the traditional LLE algorithm. The AC algorithm could adaptively determine the number of feature clusters (the number is298), and its average classification accuracy is82%, which gains an average classification accuracy promotion of10.7%,5.4%,0.3%and1.1%compared to the man-made cluster numbers of100,200,300and400, respectively. The experimental results show that the DO-LLE algorithm and AC algorithm could effectively enhance the image semantic classification performance.
(2) In order to solve the effective quantitative representation problem of images, it proposes a Bag-Of-Visual-Words model based Robust Image Representation approach (BOVW-RIR). In order to enhance the characterization of the low-level image features, it fuses the Speed Up Robust Feature (SURF) and the Multiresolution Histogram Moment feature (MRHM) to generate a low-level compound feature, which contains the local information and the structural information of each image. It successively does feature dimension reduction and feature clustering. It uses the feature clustering results to construct the visual vocabulary and represent each image using the visual vocabulary. The BOVW-RIR approach achieves the average classification accuracy results of89.1%,83.9%and82%on the Oliva and Torralba (OT), the Fei-fei and Perona (FP), and the LPS image classification datasets, respectively. It also achieves an average classification accuracy result of70%on the food safety image dataset. The BOVW-RIR approach is experimentally compared with three famous image representation approaches (Fei-fei, Bosch, Lazebnik) on the OT, FP and LPS datasets. The BOVW-RIR approach gains a classification accuracy promotion of4.4%,6.6%and7.3%on the OT, FP and LPS datasets compared to the best of the three comparison approaches respectively, which shows that the BOVW-RIR approach could more effectively make quantitative representation of each image.
(3) This dissertation proposes a Latent Semantic Topic Weighted Fusion based Image Semantic Annotation model (LSTWF-ISA). This method models the latent semantic topics for the semantic keywords and the image visual words of the training data, in order to get the latent topic distributions of the textual modal data and visual modal data. It utilizes the entropy of the visual words distribution to calculate the weighting parameter, and fuses each latent topic distribution of the textual modal data and visual modal data through this weighting parameter. It generates the fusion latent semantic topic distribution, and establishes the LSTWF-ISA model based on this distribution. The LSTWF-ISA model achieves the average F-measure results of0.71and0.22on the49best words subset and the most commonly used263words subset of the Core15K dataset, respectively. It also gets the average F-measure result of0.36on the image annotation dataset oriented for the20categories of the food safety emergencies. The LSTWF-ISA model is experimentally compared with four well-known image semantic annotation models (TM, CMRM, CRM, and PLSA-WORDS). The signed-rank test is used to check the comparison results. The LSTWF-ISA model gains an average F-measure promotion of11%and29%on the49best words subset and the most common used263words subset of the Core15K dataset compared to the best of the three comparison approaches respectively, which shows that the LSTWF-ISA model could enhance the annotation performance by using the latent semantic relationship between the textual modality and visual modality.
(4) This dissertation gives out the formal definition of the text-image co-occurrence data, which describes the characteristics of the documents containing both texts and images within the event constraint. A Text-Image Feature Mapping (TIFM) algorithm is proposed, which conducts the feature mapping from the text feature space to the image feature space based on the text-image co-occurrence data. The proposed TIFM algorithm gains an average Euclidean distance of0.024, an average cosine similarity of0.84and an average K-L divergence value of0.075between the approximated image feature distribution and the ground truth image feature distribution on a text and image dataset oriented for20categories of food safety emergencies, which shows that the TIFM algorithm could effectively map the text feature distribution into the image feature distribution.
(5) In order to solve the problem of using text data to aid the image semantic classification, this dissertation proposes a Feature Transferring based Image Semantic Classification (FT-ISC) approach. Regarding the calculating difficulty of the text features caused by their massiveness and sparseness, it proposes an Information Gain based Text Semantic Feature Selection (IGTSFS) approach, which figures out the effective text topic features by computing the information gain of each latent semantic topic extracted from the text data. The proposed FT-ISC model achieves an average classification accuracy result of76%over the whole food safety dataset and a classification accuracy result of86%on the "Crayfish" food safety semantic category. The FT-ISC model is compared to the Bayesian classification model and the Label Query classification model on the food safety dataset. The FT-ISC model gains a classification accuracy promotion of8%and5%compared to the Bayesian classification model and the Label Query classification model, respectively.
(6) This dissertation proposes a Feature Weighting based Image Semantic Classification (FW-ISC) approach. This approach combines the Filter-based Feature Weighting mechanism (FFW) with the support vector machine to achieve image semantic classification. To solve the problem of the uneven distribution of features and the difficulty in measuring the closeness of the features with semantic categories in the feature weighting procedure, it proposes a Conditional Mutual Information based Feature Diversity Measuring approach (CMIFDM). It calculates the weight values of the features of each semantic category through iterative computation. It designs the feature weighting kernel function and combines the FFW feature weighting mechanism with SVM. The feature weighted results are used in the training procedure of this classifier. The FW-ISC approach achieves an average classification accuracy of87%and75%on the LPS image dataset and the food safety image dataset, respectively. The FW-ISC approach is compared with the traditional support vector machine classifier. The signed-rank test is used to check the experimental results. The experimental results show that the FW-ISC method gains a classification accuracy promotion of5%and8%compared to the traditional support vector machine classifier on the LPS dataset and food safety dataset respectively, which indicates that the FW-ISC method could effectively enhance the image semantic classification performance.
(1)本文提出了一种距离优化的特征压缩降维算法(DO-LLE),解决了由于特征点分布不均衡造成的近邻点参数难以确定和降维效果不稳定的问题。针对视觉词汇本的大小难以确定的问题,本文提出了一种自适应特征聚类算法(AC),该算法根据最大最小距离规则和DBI指标对特征集合进行迭代计算,自适应的生成最优聚类中心。将DO-LLE算法和AC算法在LPS数据集(Lazebnik Schmid and Ponce, LPS)上通过图像分类实验进行验证。DO-LLE算法的平均分类准确率为81.9%,比传统的LLE算法高出3.3%。AC算法可以自适应的确定特征聚类数(聚类数为298)其平均分类准确率为82%,比人工设定聚类数为100、200、300、400的平均分类准确率依次高出10.7%、5.4%、0.3%、1.1%。实验结果表明,DO-LLE算法和AC算法可以有效地提高图像语义分类准确率。
(2)为了解决对图像进行有效量化表达的问题,本文提出了一种基于视觉词袋模型的健壮的图像表达方法(BOVW-RIR)。为了增强图像底层特征的表征力,将图像加速稳健特征和多分辨率直方图矩特征进行了融合,生成了一种具有图像局部信息和结构信息的底层复合特征。依次对特征进行压缩降维和聚类,按照特征聚类结果构建视觉词汇本,并用该视觉词汇本对图像进行表达。BOVW-RIR方法在OT数据集(Oliva and Torralba, OT)、FP数据集(Fei-fei and Perona, FP)(?)口LPS数据集上依次取得了89.1%,83.9%,82%的平均分类准确率,在20类食品安全图像数据集上取得了70%的平均分类准确率。将BOVW-RIR方法与目前三种著名的图像表达方法(Fei-fei, Bosch, Lazebnik)进行了实验比较,BOVW-RIR方法在OT、FP和LPS数据集上的分类准确率比三种比较方法中的最优者分别高出了4.4%,6.6%和7.3%,这表明BOVW-RIR方法能更加有效地对图像进行量化表达。
(3)本文提出了一种基于潜在语义主题加权融合的图像语义标注模型(LSTWF-ISA),该模型对训练数据的语义标注词和图像视觉词汇进行了潜在语义主题建模,得到了文本模态数据和视觉模态数据的潜在语义主题分布。利用计算视觉词汇分布的信息熵所得出的权重参数,对文本模态数据和视觉模态数据的潜在语义主题分布进行加权融合,得到融合语义主题分布,利用融合语义主题分布构建了LSTWF-ISA图像语义标注模型。本文提出的LSTWF-ISA模型在Core15K数据集的49个最优标注词子集上的平均F度量值为0.71,在常用的263个标注词子集上的平均F度量值为0.22,在20类食品安全事件的图像标注数据集上的平均F度量值为0.36。将LSTWF-ISA模型与四种著名的图像语义标注模型(TM,CMRM,CRM, PLSA-WORDS)进行实验比较,并对结果进行符号秩检验。LSTWF-ISA模型在Core15K数据集的49个最优标注词子集和常用的263个标注词子集j二的平均F度量值依次比四种比较方法中的最优者提高了11%和29%,这表明LSTWF-ISA模型能够通过利用文本模态和视觉模态的潜在语义关联提升图像语义标注效果。
(4)本文给出了文本-图像共现数据的形式化定义,描述了基于事件约束的文本-图像伴随文档的特性。提出了一一种文本-图像特征映射算法(TIFM),基于文本-图像共现数据从文本特征空间向图像特征空间进行特征映射。TIFM算法在包含20个类别的食品安全数据集上所计算出的图像特征分布与基准图像特征分布的平均欧氏距离为0.024、平均余弦相似度为0.84、平均K-L分歧值为0.075,这表明TIFM算法能够有效地将文本特征分布映射为图像特征分布。
(5)为了解决利用文本数据帮助进行图像语义分类的问题,本文提出了一种基于特征迁移的图像语义分类模型(FT-ISC)。针对文本特征的海量性和稀疏性导致不易计算的问题,本文提出了一种基于信息增益的文本语义特征选择方法(IGTSFS)。该方法对文本数据进行潜在语义主题建模,计算出各潜在语义主题的信息增益,得出显著文本语义主题特征。本文提出的FT-ISC模型在食品安全数据集上取得了76%的平均分类准确率,在“小龙虾”食品安全语义类别上取得了86%的分类准确率。将FT-ISC模型在食品安全数据集上与贝叶斯分类模型和标记查询分类模型进行比较,实验结果表明,FT-ISC模型在分类准确率上比贝叶斯分类模型和标记查询分类模型依次高出8%和5%。
(6)本文提出了一种基于特征加权的图像语义分类方法(FW-ISC),该方法通过将筛选式的图像特征加权机制(FFW)与支持向量机结合,实现了图像语义分类。在特征加权过程中,针对特征分布不均衡以及特征与语义类别之间的紧密度难以度量的问题,提出一种基于条件互信息的特征差异性度量方法(CMIFDM),通过迭代计算得出某个语义类别下特征的权重值。通过设计特征加权核函数,将FFW特征加权机制与支持向量机结合,在分类器训练过程中可以利用特征加权的结果。FW-ISC方法在LPS数据集上取得了87%的平均分类准确率,在食品安全图像数据集上取得了75%的平均分类准确率。将FW-ISC方法与传统的支持向量机分类器进行了比较,并对实验结果进行了符号秩检验。实验结果表明,FW-ISC方法在LPS数据集和食品安全图像数据集上的分类准确率比传统的支持向量机分类方法依次高出5%和8%,这表明FW-ISC方法可以有效地提高图像语义分类准确率。
At present, China's food safety emergencies happen frequently, which results in a rapid growth of the relevant information on the Internet. The form of these data has the cross-media characteristics. Images, as the main part of such data, become an important source of the relevant information for the public due to their richness of semantics. Although the traditional semantic analysis and classification techniques have achieved certain progress on semantic understanding of images, there are still many problems to be solved in the aspects of the quantitative representation and the feature selection of images. The cross-media information with the semantic background of food safety emergencies has heterogeneous feature spaces and closely related semantic spaces. How to use these cross-media information with a multi-modal form to mine the latent semantic relationships, as well as to transfer knowledge between different modal data, proposes new challenges to the semantic analysis and classification tasks. Thereby, it could help the understanding of the semantics contained in the images. Consider the problems of quantitative representation and feature selection of images, mining the latent semantic relationships and transferring knowledge between heterogeneous data sources, this dissertation studies on the semantic analysis and classification techniques for the cross-media information of the food safety emergencies from three aspects such as image quantitative representation, semantic annotation and semantic classification. The main contributions and innovations of this dissertation are as follows:
(1)This dissertation proposes a Distance Optimization based Locally Linear Embedding (DO-LLE) feature dimension reduction algorithm, which could solve the difficulty in determining the neighbor point parameters and instability of the dimension reduction results. Regarding the difficulty in determining the size of the visual vocabulary, it proposes a feature Adaptive Clustering algorithm (AC). The AC algorithm could adaptively generate the optimized cluster centers through the iterative computation on the features set according to the max-min distance rule and the Davies-Bouldin Index (DBI). The DO-LLE algorithm and the AC algorithm are verified via the image classification experiment on the Lazebnik Schmid and Ponce (LPS) image classification dataset. The DO-LLE algorithm gains an average classification accuracy result of81.9%which is3.3%more than the traditional LLE algorithm. The AC algorithm could adaptively determine the number of feature clusters (the number is298), and its average classification accuracy is82%, which gains an average classification accuracy promotion of10.7%,5.4%,0.3%and1.1%compared to the man-made cluster numbers of100,200,300and400, respectively. The experimental results show that the DO-LLE algorithm and AC algorithm could effectively enhance the image semantic classification performance.
(2) In order to solve the effective quantitative representation problem of images, it proposes a Bag-Of-Visual-Words model based Robust Image Representation approach (BOVW-RIR). In order to enhance the characterization of the low-level image features, it fuses the Speed Up Robust Feature (SURF) and the Multiresolution Histogram Moment feature (MRHM) to generate a low-level compound feature, which contains the local information and the structural information of each image. It successively does feature dimension reduction and feature clustering. It uses the feature clustering results to construct the visual vocabulary and represent each image using the visual vocabulary. The BOVW-RIR approach achieves the average classification accuracy results of89.1%,83.9%and82%on the Oliva and Torralba (OT), the Fei-fei and Perona (FP), and the LPS image classification datasets, respectively. It also achieves an average classification accuracy result of70%on the food safety image dataset. The BOVW-RIR approach is experimentally compared with three famous image representation approaches (Fei-fei, Bosch, Lazebnik) on the OT, FP and LPS datasets. The BOVW-RIR approach gains a classification accuracy promotion of4.4%,6.6%and7.3%on the OT, FP and LPS datasets compared to the best of the three comparison approaches respectively, which shows that the BOVW-RIR approach could more effectively make quantitative representation of each image.
(3) This dissertation proposes a Latent Semantic Topic Weighted Fusion based Image Semantic Annotation model (LSTWF-ISA). This method models the latent semantic topics for the semantic keywords and the image visual words of the training data, in order to get the latent topic distributions of the textual modal data and visual modal data. It utilizes the entropy of the visual words distribution to calculate the weighting parameter, and fuses each latent topic distribution of the textual modal data and visual modal data through this weighting parameter. It generates the fusion latent semantic topic distribution, and establishes the LSTWF-ISA model based on this distribution. The LSTWF-ISA model achieves the average F-measure results of0.71and0.22on the49best words subset and the most commonly used263words subset of the Core15K dataset, respectively. It also gets the average F-measure result of0.36on the image annotation dataset oriented for the20categories of the food safety emergencies. The LSTWF-ISA model is experimentally compared with four well-known image semantic annotation models (TM, CMRM, CRM, and PLSA-WORDS). The signed-rank test is used to check the comparison results. The LSTWF-ISA model gains an average F-measure promotion of11%and29%on the49best words subset and the most common used263words subset of the Core15K dataset compared to the best of the three comparison approaches respectively, which shows that the LSTWF-ISA model could enhance the annotation performance by using the latent semantic relationship between the textual modality and visual modality.
(4) This dissertation gives out the formal definition of the text-image co-occurrence data, which describes the characteristics of the documents containing both texts and images within the event constraint. A Text-Image Feature Mapping (TIFM) algorithm is proposed, which conducts the feature mapping from the text feature space to the image feature space based on the text-image co-occurrence data. The proposed TIFM algorithm gains an average Euclidean distance of0.024, an average cosine similarity of0.84and an average K-L divergence value of0.075between the approximated image feature distribution and the ground truth image feature distribution on a text and image dataset oriented for20categories of food safety emergencies, which shows that the TIFM algorithm could effectively map the text feature distribution into the image feature distribution.
(5) In order to solve the problem of using text data to aid the image semantic classification, this dissertation proposes a Feature Transferring based Image Semantic Classification (FT-ISC) approach. Regarding the calculating difficulty of the text features caused by their massiveness and sparseness, it proposes an Information Gain based Text Semantic Feature Selection (IGTSFS) approach, which figures out the effective text topic features by computing the information gain of each latent semantic topic extracted from the text data. The proposed FT-ISC model achieves an average classification accuracy result of76%over the whole food safety dataset and a classification accuracy result of86%on the "Crayfish" food safety semantic category. The FT-ISC model is compared to the Bayesian classification model and the Label Query classification model on the food safety dataset. The FT-ISC model gains a classification accuracy promotion of8%and5%compared to the Bayesian classification model and the Label Query classification model, respectively.
(6) This dissertation proposes a Feature Weighting based Image Semantic Classification (FW-ISC) approach. This approach combines the Filter-based Feature Weighting mechanism (FFW) with the support vector machine to achieve image semantic classification. To solve the problem of the uneven distribution of features and the difficulty in measuring the closeness of the features with semantic categories in the feature weighting procedure, it proposes a Conditional Mutual Information based Feature Diversity Measuring approach (CMIFDM). It calculates the weight values of the features of each semantic category through iterative computation. It designs the feature weighting kernel function and combines the FFW feature weighting mechanism with SVM. The feature weighted results are used in the training procedure of this classifier. The FW-ISC approach achieves an average classification accuracy of87%and75%on the LPS image dataset and the food safety image dataset, respectively. The FW-ISC approach is compared with the traditional support vector machine classifier. The signed-rank test is used to check the experimental results. The experimental results show that the FW-ISC method gains a classification accuracy promotion of5%and8%compared to the traditional support vector machine classifier on the LPS dataset and food safety dataset respectively, which indicates that the FW-ISC method could effectively enhance the image semantic classification performance.
引文
[1]http://www.cnnic.net.cn/hlwfzyj/hlwxzbg/.
[2]谢毓湘,栾悉道,吴玲达.多媒体数据语义鸿沟问题分析[J].武汉理工大学学报,2011,33(6):859-863.
[3]张鸿.基于相关性挖掘的跨媒体检索研究[D].杭州：浙江大学,2007.
[4]徐昊.基于跨媒体信息和高效图像编码算法的图像搜索关键技术[D].合肥：中国科学技术大学,2011.
[5]Jitao Sang. User-Aware Image Tag Refinement via Ternary Semantic Analysis [J], IEEE Transactions on Multimedia,2012,14(3):883-895.
[6]Lin Wanxia, Lu Tong, and Su Feng. A Novel Multi-modal Integration and Propagation Model for Cross-Media Information Retrieval [M]. Lecture Notes in Computer Science (LNCS), Springer,2012:740-749.
[7]Xiaohua Zhai, Yuxin Peng, and Jianguo Xiao. Effective Heterogeneous Similarity Measure with Nearest Neighbors for Cross-Media Retrieval [M]. Lecture Notes in Computer Science (LNCS), Springer,2012:312-322.
[8]Fei Wu, Yahong Han, and Yueting Zhuang. Multiple Hypergraph Clustering of Web Images by Mining Word2Image Correlations [J]. Journal of Computer Science and Technology,2010,25(4):750-760.
[9]Yi Yang, Fei Wu, Feiping Nie, et al. Web and Personal Image Annotation by Mining Label Correlation with Relaxed Visual Graph Embedding [J]. IEEE Transactions on Image Processing,2012,21(3):1339-1351.
[10]J. C. Caicedo, J. BenAbdallah, F. A. Gonzdlez, et al. Multimodal Representation, Indexing, Automated Annotation and Retrieval of Image Collections via Non-negative Matrix Factorization [J]. Neurocomputing,2012,76(1):50-60.
[11]Yusuke Yamamoto, Katsumi Tanaka. ImageAlert:Credibility Analysis of Text-image Pairs on the Web [A]. In:Proceedings of the 2011 ACM Symposium on Applied Computing [C], TaiChung, Taiwan.2011:1724-1731.
[12]Dhruv Kumar Mahajan, Malcolm Slaney. Image Classification Using the Web Graph [A]. In:Proceedings of the International Conference on Multimedia [C], Firenze, Italy.2010:991-994.
[13]Stefan Romberg, Rainer Lienhart, and Eva Horster. Multimodal Image Retrieval [J], International Journal of Multimedia Information Retrieval,2012,1(1): 31-44.
[14]李志欣,施智平,刘曦等.建模连续视觉特征的图像语义标注方法[J].计算机辅助设计与图形学学报,2010,22(8)：1412-1420.
[15]E. Moxley, M. Tao, and B. S. Manjunath. Video Annotation through Search and Graph Reinforcement Mining [J]. IEEE Transactions on Multimedia,2010, 12(3):184-193.
[16]Yueting Zhuang, Yahong Han, Fei Wu, et al. Stable Multi-label Boosting for Image Annotation with Structural Feature Selection [J]. Science in China Series F:Information Sciences,2011,54(12):2508-2521.
[17]Nikhil Rasiwasia, Jose Costa Pereira, Emanuele Coviello, et al. A New Approach to Cross-Modal Multimedia Retrieval [A]. In:Proceedings of the International Conference on Multimedia [C], Firenze, Italy.2010:251-260.
[18]Hong Zhang, Xiaoli Liu. Boosting Multimodal Semantic Understanding by Local Similarity Adaptation and Global Correlation Propagation [M]. Lecture Notes in Computer Science (LNCS), Springer,2010:148-158.
[19]Chung-Hong Lee, Shih-Hao Wang. An Information Fusion Approach to Integrate Image Annotation and Text Mining Methods for Geographic Knowledge Discovery [J]. Expert Systems with Applications,2012, Available online, doi:http://dx.doi.org/10.1016/j.bbr.2011.03.
[20]M. Guillaumin, J. Verbeek, and C. Schmid. Multimodal Semi-supervised Learning for Image Classification [A]. In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], San Francisco, USA. 2010:902-909.
[21]Yanwei Pang, Zhao Ma, and Yuan Yuan. Multimodal Learning for Multi-label Image Classification [A]. In:Proceedings of IEEE International Conference on Image Processing (ICIP) [C], Brussels.Belgium.2011:1797-1800.
[22]Ningning Liu, Emmanuel Dellandrea, Liming Chen, et al. Multimodal Recognition of Visual Concepts Using Histograms of Textual Concepts and Selective Weighted Late Fusion Scheme [J], Computer Vision and Image Understanding,2013,117(5):493-512.
[23]Jun Yu, Dacheng Tao, and Meng Wang. Adaptive Hypergraph Learning and Its Application in Image Classification [J]. IEEE Transactions on Image Processing, 2012,21(7):3262-3272.
[24]Fuzhen Zhuang, Ping Luo, Hui Xiong, et al. Cross-Domain Learning from Multiple Sources:A Consensus Regularization Perspective [J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(12):1664-1678.
[25]Yin Zhu, Yuqiang Chen, Zhongqi Lu, et al. Heterogeneous Transfer Learning for Image Classification [A]. In:Proceedings of 25th AAAI Conference on Artificial Intelligence [C], San Francisco, CA, USA.2011:1304-1308.
[26]EvanWei Xiang, Sinno Jialin Pan, Weike Pan, et al. Source Selection Free Transfer Learning [A]. In:Proceedings of the 22nd International Joint Conference on Artificial Intelligence [C], Barcelona, Catalonia, Spain.2011: 2355-2360.
[27]T. Tommasi, F. Orabona, and B. Caputo. Safety in numbers:Learning Categories from Few Examples with Multi Model Knowledge Transfer [A]. In: Proceedings of 2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], San Francisco, CA, USA.2010:3081-3088.
[28]Y. P. Zhou, J. Hu, and J. P. Du. Cross-media Topic Analysis and Information Retrieval [A]. In:Proceedings of the 9th International Conference on Fuzzy Systems and Knowledge Discovery [C], Chengdu, China.2012:1211-1215.
[29]江悦.场景图像内容表述和分类研究[D].长沙：国防科学技术大学,2010.
[30]J. C. Van Gemert, C. J. Veenman, A. W. M. Smeulders, et al. Visual Word Ambiguity [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010,32(7):1271-1283.
[31]M. Sun, H. V. hamme. Image Pattern Discovery by Using the Spatial Closeness of Visual Code Words [A]. In:Proceedings of the 18th IEEE International Conference on Image Processing (ICIP) [C], Brussels, Belgium.2011:205-208.
[32]胡正平,戎怡.基于EILBP视觉描述子结合PLSA的场景分类算法[J].光电工程,2010,37(11)：128-134.
[33]J. D. Deng. Feature Analysis for Object and Scene Categorization [M]. Studies in Computational Intelligence, Springer,2011:225-244.
[34]Yuanqing Lin, Fengjun Lv, Shenghuo Zhu, et al. Large-scale Image Classification:Fast Feature Extraction and SVM Training [A]. In:Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], Colorado, Springs, USA.2011:1689-1696.
[35]Guanghua Gu, Yao Zhao, Zhenfeng Zhu, et al. Scene Classification based on Spatial Pyramid Representation by Superpixel Lattices and Contextual Visual Features [J]. Optical Engineering,2012,51(1):017201:1-8.
[36]Jianzhao Qin, Nelson H. C. Yung. Feature Fusion within Local Region Using Localized Maximum Margin Learning for Scene Categorization [J]. Pattern Recognition,2012,45(4):1671-1683.
[37]Yue Jiang, Jin Chen, and Runsheng Wang. Fusing Local and Global Information for Scene Classification [J]. Optical Engineering,2010,49(4):047001:1-9.
[38]江悦,王润生.基于多特征潜在主题模型的场景图像分类[J].信号处理,2010,26(4):530-544.
[39]陈诗国,张道强.半监督降维方法的实验比较[J].软件学报,2011,22(1):28-43.
[40]Zhenyue Zhang, Jing Wang, and Hongyuan Zha. Adaptive Manifold Learning [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2012, 34(2):253-265.
[41]C. Hongping, K. Mikolajczyk, and J. Matas. Learning Linear Discriminant Projections for Dimensionality Reduction of Image Descriptors [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2010, 33(2):338-352.
[42]Wen Ge, Xu Hongzhe, and Zheng Weibin. Multi-kernel PCA based High-dimensional Images Feature Reduction [A]. In:Proceedings of 2011 International Conference on Electric Information and Control Engineering (ICEICE) [C], Wuhan, China.2011:5966-5969.
[43]N. Naikal, A. Y. Yang, and S. S. Sastry. Informative Feature Selection for Object Recognition via Sparse PCA [A]. In:Proceedings of 2011 IEEE International Conference on Computer Vision (ICCV) [C], Barcelona, Spain.2011:818-825.
[44]M. Brown, H. Gang, and S. Winder. Discriminative Learning of Local Image Descriptors [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(1):43-57.
[45]Ruiping Wang, Shiguang Shan, Xilin Chen, et al. Maximal Linear Embedding for Dimensionality Reduction [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(9):1776-1792.
[46]杨丹,李博,赵红.鲁棒视觉词汇本的自适应构造与自然场景分类应用[J].电子与信息学报,2010,32(9):2139-2144.
[47]Yu Kong, Xiaoqin Zhang, Weiming Hu, et al. Adaptive Learning Codebook for Action Recognition [J]. Pattern Recognition Letters,2011,32(8):1178-1186.
[48]Zhiguo Yang, Yuxin Peng, and Jianguo Xiao. Visual Vocabulary Optimization with Spatial Context for Image Annotation and Classification [M]. Lecture Notes in Computer Science (LNCS), Springer,2012:89-102.
[49]Xie Wenjie, Xu De, Tang Yingjun, et al. Mutual Information Based Codebooks Construction for Natural Scene Categorization [J]. Chinese Journal of Electronics,2011,20(3):419-424.
[50]Wang Zhe, Liu Guizhong, and Qian Xueming. An Approach to the Compact and Efficient Visual Codebook Based on the SIFT Descriptor [M]. Lecture Notes in Computer Science (LNCS), Springer,2010:461-469.
[51]Lei Ji, Zheng Qin, Kai Chen, et al. Visual Recognition Using Density Adaptive Clustering [A], In:Proceedings of the 5th FTRA International Conference on Multimedia and Ubiquitous Engineering [C], Loutraki, Greece.2011:67-72.
[52]Zeng Chen, Jin Hou, Dengsheng Zhang, et al. An Annotation Rule Extraction Algorithm for Image Retrieval [J]. Pattern Recognition Letters,2012,33(10): 1257-1268.
[53]Ceyhun Burak Akgul, Daniel L. Rubin, Sandy Napel, et al. Content-Based Image Retrieval in Radiology:Current Status and Future Directions [J]. Journal of Digital Imaging,2011,24(2):208-222.
[54]Dengsheng Zhang, M. Monirul Islam, and Guojun Lu. A Review on Automatic Image Annotation Techniques [J]. Pattern Recognition,2012,45(1):346-362.
[55]J. Tang, Z. Zha, D. Tao, et al. Semantic-Gap-Oriented Active Learning for Multi-Label Image Annotation [J]. IEEE Transactions on Image Processing, 2012,21(4):2354-2360.
[56]王科平.自动图像标注的关键技术研究[D].北京：北京邮电大学,2011.
[57]G. Carneiro, A. B. Chan, P. J. Moreno, et al. Supervised Learning of Semantic Classes for Image Annotation and Retrieval [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(3):394-410.
[58]路晶,马少平.使用基于多例学习的启发式SVM算法的图像自动标注[J].计算机研究与发展,2609,46(5)：864-871.
[59]Lei Yinjie, Wong Wilson, Liu Wei, et al. An HMM-SVM-Based Automatic Image Annotation Approach [M]. Lecture Notes in Computer Science (LNCS), Springer,2011:115-126.
[60]Feng Songhe, Xu De. Transductive Multi-Instance Multi-Label Learning Algorithm with Application to Automatic Image Annotation [J]. Expert Systems with Applications,2010,37(1):661-670.
[61]Jianping Fan, Yi Shen, Chunlei Yang, et al. Structured Max-Margin Learning for Inter-Related Classifier Training and Multilabel Image Annotation [J]. IEEE Transactions on Image Processing,2011,20(3):837-854.
[62]P. Duygulu, K. Barnard, J. F. Q de Freitas, et al. Object Recognition as Machine Translation:Learning a Lexicon for a Fixed Image Vocabulary [M]. Lecture Notes in Computer Science (LNCS), Springer,2006:349-354.
[63]V. Lavrenko, R. Manmatha, and J. Jeon. A Model for Learning the Semantics of Pictures [A]. In:Proceedings of the 18th Annual Conference on Neural Information Processing Systems (NIPS) [C]. Vancouver, Canada.2004: 553-560.
[64]F. Monay, D. Gatica-Perez. Modeling Semantic Aspects for Cross-media Image Indexing [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007,29(10):1802-1817.
[65]Anne-Marie Tousch, Stephane Herbin, and Jean-Yves Audibert. Semantic Hierarchies for Image Annotation:A Survey [J]. Pattern Recognition,2012, 45(1):333-345.
[66]Xing Xu, Kyushu, A. Shimada, et al. Correlated Topic Model for Image Annotation [A]. In:Proceedings of the 19th Korea-Japan Joint Workshop on Frontiers of Computer Vision [C], Incheon, Korea.2013:201-208.
[67]Ning Zhou, Yi Shen, Jinye Peng, et al. Leveraging Auxiliary Text Terms for Automatic Image Annotation [A]. In:Proceedings of the 20th international conference on World Wide Web [C], Hyderabad, India.2011:175-176.
[68]Zhiwu Lu, H. S. Ip. Horace. Automatic Image Annotation Based on Generalized Relevance Models [J]. Journal of Signal Processing Systems,2011,65(1): 23-33.
[69]Bing-Kun Bao, Teng Li, and Shuicheng Yan. Hidden-Concept Driven Multilabel Image Annotation and Label Ranking [J]. IEEE Transactions on Multimedia, 2012,14(1):199-210.
[70]柯逍,李绍滋,曹冬林.基于判别模型与生成模型的层叠图像自动标注[J].模式识别与人工智能,2011,24(3)：305-313.
[71]M. Dixit, N. Rasiwasia, and N. Vasconcelos. Adapted Gaussian Models for Image Classification [A]. In:Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], Colorado, USA.2011: 937-943.
[72]Zhang Lei, Ma Jun. Image Annotation by Incorporating Word Correlations into Multi-class SVM [J]. Soft Computing-A Fusion of Foundations, Methodologies and Applications,2011,15(5):917-927.
[73]W. Hua, H. Heng, and C. Ding. Image Annotation Using Bi-relational Graph of Images and Semantic Labels [A]. In:Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], Colorado, USA.2011: 793-800.
[74]Jinhui Tang, Haojie Li, Guojun Qi, et al. Image Annotation by Graph-Based Inference With Integrated Multiple/Single Instance Representations [J]. IEEE Transactions on Multimedia,2010,12(2):131-141.
[75]Yuanlong Shao, Yuan Zhou, and Yuan Zhou. Variational Inference with Graph Regularization for Image Annotation [J]. ACM Transactions on Intelligent Systems and Technology,2011,2(2):1-21.
[76]Michael I. Jordan, Zoubin Ghahramani, and S. Tommi. An Introduction to Variational Methods for Gramphical Models [J]. Machine Learning,1999, 37(1999):183-233.
[77]郭跃华,朱月萍.概率论与数理统计[M],北京：高等教育出版社,2011:201-253.
[78]王桂平,王衍,任嘉辰.图论算法理论、实现及应用[M].北京：北京大学出版社,2011：2-10.
[79]Lei Zhang, Qiang Ji. Image Segmentation with a Unified Graphical Model [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(8): 1406-1425.
[80]David Newman, Jey Han Lau, Karl Grieser, et al. Automatic Evaluation of Topic Coherence [A]. In:Proceedings of the 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics [C], Los Angeles, USA.2010:100-108.
[81]Mark Steyvers, Tom Griffths. Latent Semantic Analysis:A Road to Meaning [M]. Laurence Erlbaum,2007:427-442.
[82]陶长琪.决策理论与方法[M].北京：中国人民大学出版社,2010:34-54.
[83]Hanna Wallach, Iain Murray, Ruslan Salakhutdinov, et al. Evaluation Methods for Topic Models [A]. In:Proceedings of the 26th International Conference on Machine Learning [C], Montreal, Canada.2009:1105-1112.
[84]Hofmann Thomas. Probabilistic Latent Semantic Analysis [A]. In:Proceedings of the 15th International Conference on Uncertainity in Artificial Intelligence [C], Stockholm, Sweden.1999:289-296.
[85]徐戈,王厚.自然语言处理中主题模型的发展[J].计算机学报,2011,34(8):1423-1436.
[86]周志华,杨强.机器学习及其应用[M].北京：清华大学出版社,2011:127-165.
[87]Gregor Heinrich. A Generic Approach to Topic Models [M]. Lecture Notes in Computer Science (LNCS), Springer,2009:517-532.
[88]Hanna Wallach, David Mimno, and Andrew McCallum. Rethinking LDA:Why Priors Matter [A]. In:Proceedings of the 23rd Annual Conference on Neural Information Processing Systems (NIPS) [C], Vancouver, Canada.2009: 1973-1981.
[89]D. Blei. Probabilistic Models of Text and Images [D]. California:University of California at Berkeley,2004.
[90]Christopher M. Bishop. Pattern Recognition and Machine Learning [M]. Springer,2006:276-461.
[91]Ian Porteous, David Newman, Alexander Ihler, et al. Fast Collapsed Gibbs Sampling for Latent Dirichlet Allocation [A]. In:Proceedings of the 14th ACM International Conference on Knowledge Discovery and Data Mining [C], New York, USA.2008:569-577.
[92]曹娟,张勇东,李锦涛等.一种基于密度的自适应最优LDA模型选择方法[J].计算机学报,2008,31(10)：1780-1787.
[93]T. M. Hospedales, J. Li, S. G. Gong, et al. Identifying Rare and Subtle Behaviors:A Weakly Supervised Joint Topic Model [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(12):2451-2464.
[94]Heping Li, Fangyuan Wang, and Shuwu Zhang. Global and Local Features based Topic Model for Scene Recognition [A]. In:Proceedings of 2011 IEEE International Conference on Systems, Man, and Cybernetics (SMC) [C], Anchorage, USA.2011:532-537.
[95]Garg Vikram, Chaudhury Santanu, and Gopal Madan. Scene Categorization Using Topic Model Based Hierarchical Conditional Random Fields [A]. In: Proceedings of the 4th International Conference on Pattern Recognition and Machine Intelligence [C], Moscow, Russia.2011:206-212.
[96]Jun Zhu, Ni Lao, Ning Chen, et al. Conditional Topical Coding:An Efficient Topic Model Conditioned on Rich Features [A]. In:Proceedings of the 17th ACM International Conference on Knowledge Discovery and Data Mining [C], San Diego, USA.2011:475-483.
[97]Lu Yue, Mei Qiaozhu, and Zhai ChengXiang. Investigating Task Performance of Probabilistic Topic Models:An Empirical Study of PLSA and LDA [J]. Information Retrieval.2011,14(2):178-203.
[98]Shang Lifeng, Chan Kwokping. A Temporal Latent Topic Model for Facial Expression Recognition [M]. Lecture Notes in Computer Science (LNCS), Springer,2011:51-63.
[99]Wan Kong-Wah, Tan Ah-Hwee, Lim Joo-Hwee, et al. Topic Based Query Suggestions for Video Search [M]. Lecture Notes in Computer Science (LNCS), Springer,2012:288-299.
[100]Tao Mei, Li Lusong, Hua Xiansheng, et al. ImageSense:Towards Contextual Image Advertising [J]. ACM Transactions on Multimedia Computing, Communications and Applications,2012,8(1):98-116.
[101]I. H. Witten, E. Frank. Data Mining:Practical Machine Learning Tools and Techniques (Third Edition) [M]. Morgan Kaufmann,2011:23-56.
[102]Matthew E. Taylor, Peter Stone. Transfer Learning for Reinforcement Learning Domains:A Survey [J]. Journal of Machine Learning Research,2009,10(2009): 1633-1685.
[103]J. Blitzer, R, McDonald, and F. Pereira. Domain Adaptation with Structural Correspondence Learning [A]. In:Proceedings of International Conference on Empirical Methods in Natural Language Processing [C], Sydney, Australia. 2006:120-128.
[104]M. Rohrbach, M. Stark, G. Szarvas, et al. What Helps Where and Why? Semantic Relatedness for Knowledge Transfer [A]. In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], San Francisco, USA.2010:910-917.
[105]Moez Ben Haj Hmida, Yahya Slimani. Meta-learning in Grid-based Data Mining Systems [J]. International Journal of Communication Networks and Distributed Systems,2010,5(3):214-228.
[106]Loeff Nicolas. A New Framework for Semi-supervised Multitask Learning [D]. Illinois:The University of Illinois at Urbana-Champaign,2009.
[107]D. Zhou, C. J. C. Burges. Spectral Clustering and Transductive Learning with Multiple Views [A]. In:Proceedings of the 24th International Conference on Machine Learning [C], Oregon, USA.2007:1159-1166.
[108]W. Dai, G. Xue, Q. Yang, et al. Transferring Naive Bayes Classifiers for Text Classification [A]. In:Proceedings of 22nd National Conference on Artificial Intelligence [C], Vancouver, Canada.2007:540-545.
[109]W. Dai, Y. Chen, G. Xue, et al. Translated Learning:Transfer Learning across Different Feature Spaces [A]. In:Proceedings of the 23rd Annual Conference on Neural Information Processing Systems (NIPS) [C], Vancouver, Canada.2008: 353-360.
[110]Pan Jialin. Feature-based Transfer Learning with Real-world Application [D]. Hong Kong:The Hong Kong University of Science and Technology,2010.
[111]Y. Yi, G. Doretto. Boosting for Transfer Learning with Multiple Sources [A]. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], San Francisco, USA.2010:1855-1862.
[112]R. Raina, A. Battle, H. Lee, et al. Self-taught Learning:Transfer Learning from Unlabeled Data [A]. In:Proceedings of the 24th International Conference on Machine Learning [C], Oregon, USA.2007:759-766.
[113]W. Dai, Q. Yang, G. R. Xue, et al. Self-taught Clustering [A]. In:Proceedings of the 25th Annual International Conference on Machine Learning [C], Helsinki, Finland.2008:200-207.
[114]W. Dai, Y. Chen, G R. Xue, et al. Translated Learning:Transfer Learning across Different Feature Spaces [A]. In:Proceedings of the 23 rd Annual Conference on Neural Information Processing Systems (NIPS) [C], Whistler, Canada.2009: 353-360.
[115]Selen Uguroglu, Jaime Carbonell. Feature Selection for Transfer Learning [M]. Lecture Notes in Computer Science (LNCS), Springer,2011:430-442.
[116]Fernando Fernandez, Javier Garcia, and Manuela Veloso. Probabilistic Policy Reuse for Inter-task Transfer Learning [J]. Robotics and Autonomous Systems, 2010,58(7):866-871.
[117]W. Dai, Q. Yang, G. Xue, et al. Boosting for Transfer Learning [A]. In: Proceedings of the 24th International Conference on Machine Learning [C], Corvallis, Oregon, USA.2007:193-200.
[118]A. Quattoni, M. Collins, and T. Darrell. Transfer Learning for Image Classification with Sparse Prototype Representations [A]. In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], Anchorage, Alaska, USA.2008:1-8.
[119]Si Si, Dacheng Tao, and Bo Geng. Bregman Divergence-Based Regularization for Transfer Subspace Learning [J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(7):929-942.
[120]Erheng Zhong, Wei Fan, Qiang Yang, et al. Cross Validation Framework to Choose amongst Models and Datasets for Transfer Learning [M], Lecture Notes in Computer Science (LNCS), Springer,2010:547-562.
[121]M. Rosenstein, Z. Marx, and L. P. Kaelbling. To Transfer or Not to Transfer [A]. In:Proceedings of International Conference on Neural Information Processing Systems (NIPS) [C], Vancouver, Canada.2005:1-4.
[122]Bin Cao, Sinno Jialin Pan, Yu Zhang, et al. Adaptive Transfer Learning [A]. In: Proceedings of the 24th AAAI Conference on Artificial Intelligence [C], Atlanta, USA.2010:407-412.
[123]Ke Zhou, Xue Gui-Rong, and Qiang Yang. Learning with Positive and Unlabeled Examples Using Topic-sensitive PLSA [J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(1):46-58.
[124]Akinori Fujino, Naonori Ueda, and Masaaki Nagata. A Robust Semi-supervised Classification Method for Transfer Learning [A]. In:Proceedings of International Conference on Information and Knowledge Management [C], Toronto, Ontario, Canada.2010:379-388.
[125]L. Feifei, P. Perona. A Bayesian Hierarchical Model for Learning Natural Scene Categories [A]. In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition [C], San Diego, USA.2005:524-531.
[126]S. Lazebnik, C. Schmid, and J. Ponce. Beyond Bags of Features:Spatial Pyramid Matching for Recognizing Natural Scene Categories [A]. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition [C], New York, USA,2006:2169-2178.
[127]A. Bosch, A. Zisserman, and X. Munoz. Scene Classification Using A Hybrid Generative/Discriminative Approach [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(4):712-727.
[128]N. Rasiwasia, N. Vasconcelos. Scene Classification with Low-dimensional Semantic Spaces and Weak Supervision [A]. In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition [C], Alaska, USA. 2008:1-6.
[129]D. Lowe. Distinctive Image Features from Scale-Invariant Key-points [J]. International Journal of Computer Vision,2004,60(2):91-110.
[130]Herbert Bay, Andreas Eelaars, and Luc Van Gool. Speed-Up Robust Features (SURF) [J]. Computer Vision and Image Understanding,2008,110(3):346-359.
[131]David Gossow, Peter Decker, and Dietrich Paulus. An Evaluation of Open Source SURF Implementations [M]. Lecture Notes in Computer Science (LNCS), Springer,201.1:169-179.
[132]Brown M., Gang H., and Winder S. Discriminative Learning of Local Image Descriptors [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(1):43-57.
[133]Yue Jiang, Runshen Wang, and Peng Zhang. Texture Description based on Multiresolution Moments of Image Histograms [J]. Optical Engineering,2008, 47(3):37005:1-9.
[134]S. T. Roweis, L. K. Saul. Nonlinear Dimensionality Reduction by Locally Linear Embedding [J]. Science,2000,290(5500):2323-2326.
[135]Bingyu Sun, Xiaoming Zhang, Jiuyong Li, et al. Feature Fusion Using Locally Linear Embedding for Classification [J]. IEEE Transactions on Neural Networks,2010,21(1):163-168.
[136]Shiqing Zhang. Enhanced Supervised Locally Linear Embedding [J]. Pattern Recognition Letters,2009,30(13):1208-1218.
[137]J. Yugang, N. ChongWah, J. Yang. Towards Optimal Bag-of-features for Object Categorization and Semantic Video Retrieval [A]. In:Proceedings of ACM International Conference on Image and Video Retrieval [C], New York, USA. 2007:494-501.
[138]Zhe Wang, Guizhong Liu, Xueming Qian, et al. An Approach to the Compact and Efficient Visual Codebook Based on SIFT Descriptor [M]. Lecture Notes in Computer Science (LNCS), Springer,2010:461-469.
[139]Darui Li, Linjun Yang, Xiansheng Hua, et al. Large-scale Robust Visual Codebook Construction [A]. In:Proceedings of International Conference on Multimedia [C], Firenze, Italy.2010:1183-1186.
[140]Luo Juan, Oubong Gwun. A Comparison of SIFT, PCA-SIFT and SURF [J]. International Journal of Image Processing,2010,3(4):143-152.
[141]江悦,王润生.基于多特征潜在主题模型的场景图像分类[J].信号处理,2010,26(4):530-544.
[142]N. Bolshakova, F. Azuaje. Cluster Validation Techniques for Genome Expression Data [J]. Signal Processing,2003,83(4):825-833.
[143]Michael Steinbach, George Karypis, and Vipin Kumar. A Comparison of Document Clustering Techniques [A]. In:Proceedings of the 6th ACM International Conference on Knowledge Discovery and Data Mining [C], Boston, USA.2000:1-20.
[144]R. F. S. Filho, A. J. M. Traina, T. Cactano, et al. Similarity Search without Tears: The OMNI Family of All-Purpose Access Methods [A]. In:Proceedings of the 17th International Conference on Data Engineering [C], Heidelberg, Germany. 2001:623-630.
[145]Slobodan Petrovic, Gonzalo Alvarez, Agustin Orfila, et al. Labelling Clusters in an Intrusion Detection System Using a Combination of Clustering Evaluation Techniques [A]. In:Proceedings of the 39th Hawaii International Conference on System Sciences, Hawaii, USA.2006:129-140.
[146]Chih-Chung Chang, Chih-Jen Lin. LIBSVM:A Library for Support Vector Machines,2001. Software available at http://www.csie.ntu.edu.tw/-cjlin/libsvm.
[147]A. Oliva, A. Torralba. Modeling the Shape of the Scene:A Holistic Representation of the Spatial Envelope [J]. International Journal of Computer Vision,2001,42(3):145-175.
[148]D. M. Blei, Andrew Y. Ng, and Michael I. Jordan. Latent Dirichlet Allocation [J]. Journal of Machine Learning Research,2003,3(2003):993-1022.
[149]D. M. Blei. Probabilistic Topic Models [J]. Communications of the ACM,2012, 55(4):77-84.
[150]A. Azevedo-Filho, R. Shachter. Laplace's Method Approximations for Probabilistic Inference in Belief Networks with Continuous Variables [A]. In: Proceedings of the 10th International Conference on Uncertainty in Artificial Intelligence [C], San Francisco, USA.1994:28-36.
[151]Christopher M. Bishop. Pattern Recognition and Machine Learning [M]. Springer,2006:276-461.
[152]Bolstad, M. William. Understanding Computational Bayesian Statistics [M]. John Wiley,2010:89-110.
[153]George Casella, Edward I. George. Explaining the Gibbs sampler [J]. The American Statistician,1992,46(3):167-174.
[154]P. Duygulu, K. Barnard, J. F. G. de Freitas, et al. Object Recognition as Machine Translation:Learning a Lexicon for a Fixed Image Vocabulary [M]. Lecture Notes in Computer Science (LNCS), Springer,2006:349-354.
[155]S. Feng, R. Manmatha, and V. Lavrenko. Multiple Bernoulli Relevance Models for Image and Video Annotation [A]. In:Proceedings of 2004 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], Washington D. C., USA.2004:1002-1009.
[156]V. Lavrenko, R. Manmatha, and J. Jeon. A Model for Learning the Semantics of Pictures [A]. In:Proceedings of the 18th Annual Conference on Neural Information Processing Systems (NIPS) [C], Vancouver, Canada.2004: 553-560.
[157]J. Jeon, V. Lavrenko, and R. Manmatha. Automatic Image Annotation and Retrieval Using Cross-media Relevance Models [A]. In:Proceedings of the 26th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval [C], Toronto, Canada.2003:119-126.
[158]A. Blum, T. Mitchell. Combining Labeled and Unlabeled Data with Co-training [A]. In:Proceedings of the 11th Annual Conference on Computational Learning Theory [C], Madison, USA.1998:92-100.
[159]K. Nigam, R. Ghani. Analyzing the Effectiveness and Applicability of Co-training [A]. In:Proceedings of the 9th International Conference on Information and Knowledge Management [C], McLean, USA.2000:86-93.
[160]I. Muslea, S. Minton, and C. A. Knoblock. Active+Semi-supervised Learning = Robust Multi-view Learning [A]. In Proceedings of the 19th International Conference on Machine Learning [C], Sydney, Australia.2002:435-442.
[161]Sinno Jialin Pan, Qiang Yang. A Survey on Transfer Learning [J]. IEEE Transactions on Knowledge and Data Engineering,22(10):1345-1359.
[162]Ning Zhong. Effective Pattern Discovery for Text Mining [J]. IEEE Transactions on Knowledge and Data Engineering,2012,24(1):30-44.
[163]Lewis David. Naive (Bayes) at Forty:The Independence Assumption in Information Retrieval [A]. In:Proceedings of the 10th European Conference on Machine Learning [C]. Chemnitz, Germany.1998:4-15.
[164]Hani M. Samawia, Martin Dunbar and Ding-Geng Chen. Steady-state Ranked Gibbs Sampler [J]. Journal of Statistical Computation and Simulation,2012, 82(8):1223-1238.
[165]禹思敏.信息论、编码及应用[M].西安：西安电子科技大学出版社,2012:34-41.
[166]A. Hansen Glen, R. W. Douglass, and A. Zardecki. Mesh Enhancement [M]. Imperial College Press,2005:404-408.
[167]I. S. Gradshteyn, I. M. Ryzhik. Tables of Integrals, Series, and Products (6th Edition) [M]. California:Academic Press,2000:1114-1125.
[168]P. N. Tan, M. Steinbach, and V. Kumar. Introduction to Data Mining [M]. Massachusetts:Addison-Wesley,2005:23-78.
[169]张燕平,张铃.机器学习理论与算法[M].北京：科学出版社,2012:168-201.
[170]王肃.基于多Agent的突发事件信息智能监测系统研究[D].北京：北京邮电大学,2011.
[171]Yijun Sun. Iterative RELIEF for Feature Weighting:Algorithms, Theories, and Applications [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(6):1035-1051.
[172]J. Lafferty, C. Zhai. Document Language Models, Query Models, and Risk Minimization for Information Retrieval [A]. In Proceedings of the 24th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval [C], Orleans, USA.2001:111-119.
[173]B. N. Jiang, X. Q. Ding, L. T. Ma, et al. A Hybrid Feature Selection Algorithm: Combination of Symmetrical Uncertainty and Genetic Algorithms [A]. In Proceedings of the 2nd International Symposium on Optimization and Systems Biology [C], Shanghai, China,2008:152-157.
[174]E. R. Laura and S. Kifian. Theoretical Comparison between the Gini Index and Information Gain Criteria [J]. Analysis of Mathematics and Artificial Intelligence,2004,41(1):77-93.
[2]谢毓湘,栾悉道,吴玲达.多媒体数据语义鸿沟问题分析[J].武汉理工大学学报,2011,33(6):859-863.
[3]张鸿.基于相关性挖掘的跨媒体检索研究[D].杭州：浙江大学,2007.
[4]徐昊.基于跨媒体信息和高效图像编码算法的图像搜索关键技术[D].合肥：中国科学技术大学,2011.
[5]Jitao Sang. User-Aware Image Tag Refinement via Ternary Semantic Analysis [J], IEEE Transactions on Multimedia,2012,14(3):883-895.
[6]Lin Wanxia, Lu Tong, and Su Feng. A Novel Multi-modal Integration and Propagation Model for Cross-Media Information Retrieval [M]. Lecture Notes in Computer Science (LNCS), Springer,2012:740-749.
[7]Xiaohua Zhai, Yuxin Peng, and Jianguo Xiao. Effective Heterogeneous Similarity Measure with Nearest Neighbors for Cross-Media Retrieval [M]. Lecture Notes in Computer Science (LNCS), Springer,2012:312-322.
[8]Fei Wu, Yahong Han, and Yueting Zhuang. Multiple Hypergraph Clustering of Web Images by Mining Word2Image Correlations [J]. Journal of Computer Science and Technology,2010,25(4):750-760.
[9]Yi Yang, Fei Wu, Feiping Nie, et al. Web and Personal Image Annotation by Mining Label Correlation with Relaxed Visual Graph Embedding [J]. IEEE Transactions on Image Processing,2012,21(3):1339-1351.
[10]J. C. Caicedo, J. BenAbdallah, F. A. Gonzdlez, et al. Multimodal Representation, Indexing, Automated Annotation and Retrieval of Image Collections via Non-negative Matrix Factorization [J]. Neurocomputing,2012,76(1):50-60.
[11]Yusuke Yamamoto, Katsumi Tanaka. ImageAlert:Credibility Analysis of Text-image Pairs on the Web [A]. In:Proceedings of the 2011 ACM Symposium on Applied Computing [C], TaiChung, Taiwan.2011:1724-1731.
[12]Dhruv Kumar Mahajan, Malcolm Slaney. Image Classification Using the Web Graph [A]. In:Proceedings of the International Conference on Multimedia [C], Firenze, Italy.2010:991-994.
[13]Stefan Romberg, Rainer Lienhart, and Eva Horster. Multimodal Image Retrieval [J], International Journal of Multimedia Information Retrieval,2012,1(1): 31-44.
[14]李志欣,施智平,刘曦等.建模连续视觉特征的图像语义标注方法[J].计算机辅助设计与图形学学报,2010,22(8)：1412-1420.
[15]E. Moxley, M. Tao, and B. S. Manjunath. Video Annotation through Search and Graph Reinforcement Mining [J]. IEEE Transactions on Multimedia,2010, 12(3):184-193.
[16]Yueting Zhuang, Yahong Han, Fei Wu, et al. Stable Multi-label Boosting for Image Annotation with Structural Feature Selection [J]. Science in China Series F:Information Sciences,2011,54(12):2508-2521.
[17]Nikhil Rasiwasia, Jose Costa Pereira, Emanuele Coviello, et al. A New Approach to Cross-Modal Multimedia Retrieval [A]. In:Proceedings of the International Conference on Multimedia [C], Firenze, Italy.2010:251-260.
[18]Hong Zhang, Xiaoli Liu. Boosting Multimodal Semantic Understanding by Local Similarity Adaptation and Global Correlation Propagation [M]. Lecture Notes in Computer Science (LNCS), Springer,2010:148-158.
[19]Chung-Hong Lee, Shih-Hao Wang. An Information Fusion Approach to Integrate Image Annotation and Text Mining Methods for Geographic Knowledge Discovery [J]. Expert Systems with Applications,2012, Available online, doi:http://dx.doi.org/10.1016/j.bbr.2011.03.
[20]M. Guillaumin, J. Verbeek, and C. Schmid. Multimodal Semi-supervised Learning for Image Classification [A]. In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], San Francisco, USA. 2010:902-909.
[21]Yanwei Pang, Zhao Ma, and Yuan Yuan. Multimodal Learning for Multi-label Image Classification [A]. In:Proceedings of IEEE International Conference on Image Processing (ICIP) [C], Brussels.Belgium.2011:1797-1800.
[22]Ningning Liu, Emmanuel Dellandrea, Liming Chen, et al. Multimodal Recognition of Visual Concepts Using Histograms of Textual Concepts and Selective Weighted Late Fusion Scheme [J], Computer Vision and Image Understanding,2013,117(5):493-512.
[23]Jun Yu, Dacheng Tao, and Meng Wang. Adaptive Hypergraph Learning and Its Application in Image Classification [J]. IEEE Transactions on Image Processing, 2012,21(7):3262-3272.
[24]Fuzhen Zhuang, Ping Luo, Hui Xiong, et al. Cross-Domain Learning from Multiple Sources:A Consensus Regularization Perspective [J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(12):1664-1678.
[25]Yin Zhu, Yuqiang Chen, Zhongqi Lu, et al. Heterogeneous Transfer Learning for Image Classification [A]. In:Proceedings of 25th AAAI Conference on Artificial Intelligence [C], San Francisco, CA, USA.2011:1304-1308.
[26]EvanWei Xiang, Sinno Jialin Pan, Weike Pan, et al. Source Selection Free Transfer Learning [A]. In:Proceedings of the 22nd International Joint Conference on Artificial Intelligence [C], Barcelona, Catalonia, Spain.2011: 2355-2360.
[27]T. Tommasi, F. Orabona, and B. Caputo. Safety in numbers:Learning Categories from Few Examples with Multi Model Knowledge Transfer [A]. In: Proceedings of 2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], San Francisco, CA, USA.2010:3081-3088.
[28]Y. P. Zhou, J. Hu, and J. P. Du. Cross-media Topic Analysis and Information Retrieval [A]. In:Proceedings of the 9th International Conference on Fuzzy Systems and Knowledge Discovery [C], Chengdu, China.2012:1211-1215.
[29]江悦.场景图像内容表述和分类研究[D].长沙：国防科学技术大学,2010.
[30]J. C. Van Gemert, C. J. Veenman, A. W. M. Smeulders, et al. Visual Word Ambiguity [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010,32(7):1271-1283.
[31]M. Sun, H. V. hamme. Image Pattern Discovery by Using the Spatial Closeness of Visual Code Words [A]. In:Proceedings of the 18th IEEE International Conference on Image Processing (ICIP) [C], Brussels, Belgium.2011:205-208.
[32]胡正平,戎怡.基于EILBP视觉描述子结合PLSA的场景分类算法[J].光电工程,2010,37(11)：128-134.
[33]J. D. Deng. Feature Analysis for Object and Scene Categorization [M]. Studies in Computational Intelligence, Springer,2011:225-244.
[34]Yuanqing Lin, Fengjun Lv, Shenghuo Zhu, et al. Large-scale Image Classification:Fast Feature Extraction and SVM Training [A]. In:Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], Colorado, Springs, USA.2011:1689-1696.
[35]Guanghua Gu, Yao Zhao, Zhenfeng Zhu, et al. Scene Classification based on Spatial Pyramid Representation by Superpixel Lattices and Contextual Visual Features [J]. Optical Engineering,2012,51(1):017201:1-8.
[36]Jianzhao Qin, Nelson H. C. Yung. Feature Fusion within Local Region Using Localized Maximum Margin Learning for Scene Categorization [J]. Pattern Recognition,2012,45(4):1671-1683.
[37]Yue Jiang, Jin Chen, and Runsheng Wang. Fusing Local and Global Information for Scene Classification [J]. Optical Engineering,2010,49(4):047001:1-9.
[38]江悦,王润生.基于多特征潜在主题模型的场景图像分类[J].信号处理,2010,26(4):530-544.
[39]陈诗国,张道强.半监督降维方法的实验比较[J].软件学报,2011,22(1):28-43.
[40]Zhenyue Zhang, Jing Wang, and Hongyuan Zha. Adaptive Manifold Learning [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2012, 34(2):253-265.
[41]C. Hongping, K. Mikolajczyk, and J. Matas. Learning Linear Discriminant Projections for Dimensionality Reduction of Image Descriptors [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2010, 33(2):338-352.
[42]Wen Ge, Xu Hongzhe, and Zheng Weibin. Multi-kernel PCA based High-dimensional Images Feature Reduction [A]. In:Proceedings of 2011 International Conference on Electric Information and Control Engineering (ICEICE) [C], Wuhan, China.2011:5966-5969.
[43]N. Naikal, A. Y. Yang, and S. S. Sastry. Informative Feature Selection for Object Recognition via Sparse PCA [A]. In:Proceedings of 2011 IEEE International Conference on Computer Vision (ICCV) [C], Barcelona, Spain.2011:818-825.
[44]M. Brown, H. Gang, and S. Winder. Discriminative Learning of Local Image Descriptors [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(1):43-57.
[45]Ruiping Wang, Shiguang Shan, Xilin Chen, et al. Maximal Linear Embedding for Dimensionality Reduction [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(9):1776-1792.
[46]杨丹,李博,赵红.鲁棒视觉词汇本的自适应构造与自然场景分类应用[J].电子与信息学报,2010,32(9):2139-2144.
[47]Yu Kong, Xiaoqin Zhang, Weiming Hu, et al. Adaptive Learning Codebook for Action Recognition [J]. Pattern Recognition Letters,2011,32(8):1178-1186.
[48]Zhiguo Yang, Yuxin Peng, and Jianguo Xiao. Visual Vocabulary Optimization with Spatial Context for Image Annotation and Classification [M]. Lecture Notes in Computer Science (LNCS), Springer,2012:89-102.
[49]Xie Wenjie, Xu De, Tang Yingjun, et al. Mutual Information Based Codebooks Construction for Natural Scene Categorization [J]. Chinese Journal of Electronics,2011,20(3):419-424.
[50]Wang Zhe, Liu Guizhong, and Qian Xueming. An Approach to the Compact and Efficient Visual Codebook Based on the SIFT Descriptor [M]. Lecture Notes in Computer Science (LNCS), Springer,2010:461-469.
[51]Lei Ji, Zheng Qin, Kai Chen, et al. Visual Recognition Using Density Adaptive Clustering [A], In:Proceedings of the 5th FTRA International Conference on Multimedia and Ubiquitous Engineering [C], Loutraki, Greece.2011:67-72.
[52]Zeng Chen, Jin Hou, Dengsheng Zhang, et al. An Annotation Rule Extraction Algorithm for Image Retrieval [J]. Pattern Recognition Letters,2012,33(10): 1257-1268.
[53]Ceyhun Burak Akgul, Daniel L. Rubin, Sandy Napel, et al. Content-Based Image Retrieval in Radiology:Current Status and Future Directions [J]. Journal of Digital Imaging,2011,24(2):208-222.
[54]Dengsheng Zhang, M. Monirul Islam, and Guojun Lu. A Review on Automatic Image Annotation Techniques [J]. Pattern Recognition,2012,45(1):346-362.
[55]J. Tang, Z. Zha, D. Tao, et al. Semantic-Gap-Oriented Active Learning for Multi-Label Image Annotation [J]. IEEE Transactions on Image Processing, 2012,21(4):2354-2360.
[56]王科平.自动图像标注的关键技术研究[D].北京：北京邮电大学,2011.
[57]G. Carneiro, A. B. Chan, P. J. Moreno, et al. Supervised Learning of Semantic Classes for Image Annotation and Retrieval [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(3):394-410.
[58]路晶,马少平.使用基于多例学习的启发式SVM算法的图像自动标注[J].计算机研究与发展,2609,46(5)：864-871.
[59]Lei Yinjie, Wong Wilson, Liu Wei, et al. An HMM-SVM-Based Automatic Image Annotation Approach [M]. Lecture Notes in Computer Science (LNCS), Springer,2011:115-126.
[60]Feng Songhe, Xu De. Transductive Multi-Instance Multi-Label Learning Algorithm with Application to Automatic Image Annotation [J]. Expert Systems with Applications,2010,37(1):661-670.
[61]Jianping Fan, Yi Shen, Chunlei Yang, et al. Structured Max-Margin Learning for Inter-Related Classifier Training and Multilabel Image Annotation [J]. IEEE Transactions on Image Processing,2011,20(3):837-854.
[62]P. Duygulu, K. Barnard, J. F. Q de Freitas, et al. Object Recognition as Machine Translation:Learning a Lexicon for a Fixed Image Vocabulary [M]. Lecture Notes in Computer Science (LNCS), Springer,2006:349-354.
[63]V. Lavrenko, R. Manmatha, and J. Jeon. A Model for Learning the Semantics of Pictures [A]. In:Proceedings of the 18th Annual Conference on Neural Information Processing Systems (NIPS) [C]. Vancouver, Canada.2004: 553-560.
[64]F. Monay, D. Gatica-Perez. Modeling Semantic Aspects for Cross-media Image Indexing [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007,29(10):1802-1817.
[65]Anne-Marie Tousch, Stephane Herbin, and Jean-Yves Audibert. Semantic Hierarchies for Image Annotation:A Survey [J]. Pattern Recognition,2012, 45(1):333-345.
[66]Xing Xu, Kyushu, A. Shimada, et al. Correlated Topic Model for Image Annotation [A]. In:Proceedings of the 19th Korea-Japan Joint Workshop on Frontiers of Computer Vision [C], Incheon, Korea.2013:201-208.
[67]Ning Zhou, Yi Shen, Jinye Peng, et al. Leveraging Auxiliary Text Terms for Automatic Image Annotation [A]. In:Proceedings of the 20th international conference on World Wide Web [C], Hyderabad, India.2011:175-176.
[68]Zhiwu Lu, H. S. Ip. Horace. Automatic Image Annotation Based on Generalized Relevance Models [J]. Journal of Signal Processing Systems,2011,65(1): 23-33.
[69]Bing-Kun Bao, Teng Li, and Shuicheng Yan. Hidden-Concept Driven Multilabel Image Annotation and Label Ranking [J]. IEEE Transactions on Multimedia, 2012,14(1):199-210.
[70]柯逍,李绍滋,曹冬林.基于判别模型与生成模型的层叠图像自动标注[J].模式识别与人工智能,2011,24(3)：305-313.
[71]M. Dixit, N. Rasiwasia, and N. Vasconcelos. Adapted Gaussian Models for Image Classification [A]. In:Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], Colorado, USA.2011: 937-943.
[72]Zhang Lei, Ma Jun. Image Annotation by Incorporating Word Correlations into Multi-class SVM [J]. Soft Computing-A Fusion of Foundations, Methodologies and Applications,2011,15(5):917-927.
[73]W. Hua, H. Heng, and C. Ding. Image Annotation Using Bi-relational Graph of Images and Semantic Labels [A]. In:Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], Colorado, USA.2011: 793-800.
[74]Jinhui Tang, Haojie Li, Guojun Qi, et al. Image Annotation by Graph-Based Inference With Integrated Multiple/Single Instance Representations [J]. IEEE Transactions on Multimedia,2010,12(2):131-141.
[75]Yuanlong Shao, Yuan Zhou, and Yuan Zhou. Variational Inference with Graph Regularization for Image Annotation [J]. ACM Transactions on Intelligent Systems and Technology,2011,2(2):1-21.
[76]Michael I. Jordan, Zoubin Ghahramani, and S. Tommi. An Introduction to Variational Methods for Gramphical Models [J]. Machine Learning,1999, 37(1999):183-233.
[77]郭跃华,朱月萍.概率论与数理统计[M],北京：高等教育出版社,2011:201-253.
[78]王桂平,王衍,任嘉辰.图论算法理论、实现及应用[M].北京：北京大学出版社,2011：2-10.
[79]Lei Zhang, Qiang Ji. Image Segmentation with a Unified Graphical Model [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(8): 1406-1425.
[80]David Newman, Jey Han Lau, Karl Grieser, et al. Automatic Evaluation of Topic Coherence [A]. In:Proceedings of the 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics [C], Los Angeles, USA.2010:100-108.
[81]Mark Steyvers, Tom Griffths. Latent Semantic Analysis:A Road to Meaning [M]. Laurence Erlbaum,2007:427-442.
[82]陶长琪.决策理论与方法[M].北京：中国人民大学出版社,2010:34-54.
[83]Hanna Wallach, Iain Murray, Ruslan Salakhutdinov, et al. Evaluation Methods for Topic Models [A]. In:Proceedings of the 26th International Conference on Machine Learning [C], Montreal, Canada.2009:1105-1112.
[84]Hofmann Thomas. Probabilistic Latent Semantic Analysis [A]. In:Proceedings of the 15th International Conference on Uncertainity in Artificial Intelligence [C], Stockholm, Sweden.1999:289-296.
[85]徐戈,王厚.自然语言处理中主题模型的发展[J].计算机学报,2011,34(8):1423-1436.
[86]周志华,杨强.机器学习及其应用[M].北京：清华大学出版社,2011:127-165.
[87]Gregor Heinrich. A Generic Approach to Topic Models [M]. Lecture Notes in Computer Science (LNCS), Springer,2009:517-532.
[88]Hanna Wallach, David Mimno, and Andrew McCallum. Rethinking LDA:Why Priors Matter [A]. In:Proceedings of the 23rd Annual Conference on Neural Information Processing Systems (NIPS) [C], Vancouver, Canada.2009: 1973-1981.
[89]D. Blei. Probabilistic Models of Text and Images [D]. California:University of California at Berkeley,2004.
[90]Christopher M. Bishop. Pattern Recognition and Machine Learning [M]. Springer,2006:276-461.
[91]Ian Porteous, David Newman, Alexander Ihler, et al. Fast Collapsed Gibbs Sampling for Latent Dirichlet Allocation [A]. In:Proceedings of the 14th ACM International Conference on Knowledge Discovery and Data Mining [C], New York, USA.2008:569-577.
[92]曹娟,张勇东,李锦涛等.一种基于密度的自适应最优LDA模型选择方法[J].计算机学报,2008,31(10)：1780-1787.
[93]T. M. Hospedales, J. Li, S. G. Gong, et al. Identifying Rare and Subtle Behaviors:A Weakly Supervised Joint Topic Model [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(12):2451-2464.
[94]Heping Li, Fangyuan Wang, and Shuwu Zhang. Global and Local Features based Topic Model for Scene Recognition [A]. In:Proceedings of 2011 IEEE International Conference on Systems, Man, and Cybernetics (SMC) [C], Anchorage, USA.2011:532-537.
[95]Garg Vikram, Chaudhury Santanu, and Gopal Madan. Scene Categorization Using Topic Model Based Hierarchical Conditional Random Fields [A]. In: Proceedings of the 4th International Conference on Pattern Recognition and Machine Intelligence [C], Moscow, Russia.2011:206-212.
[96]Jun Zhu, Ni Lao, Ning Chen, et al. Conditional Topical Coding:An Efficient Topic Model Conditioned on Rich Features [A]. In:Proceedings of the 17th ACM International Conference on Knowledge Discovery and Data Mining [C], San Diego, USA.2011:475-483.
[97]Lu Yue, Mei Qiaozhu, and Zhai ChengXiang. Investigating Task Performance of Probabilistic Topic Models:An Empirical Study of PLSA and LDA [J]. Information Retrieval.2011,14(2):178-203.
[98]Shang Lifeng, Chan Kwokping. A Temporal Latent Topic Model for Facial Expression Recognition [M]. Lecture Notes in Computer Science (LNCS), Springer,2011:51-63.
[99]Wan Kong-Wah, Tan Ah-Hwee, Lim Joo-Hwee, et al. Topic Based Query Suggestions for Video Search [M]. Lecture Notes in Computer Science (LNCS), Springer,2012:288-299.
[100]Tao Mei, Li Lusong, Hua Xiansheng, et al. ImageSense:Towards Contextual Image Advertising [J]. ACM Transactions on Multimedia Computing, Communications and Applications,2012,8(1):98-116.
[101]I. H. Witten, E. Frank. Data Mining:Practical Machine Learning Tools and Techniques (Third Edition) [M]. Morgan Kaufmann,2011:23-56.
[102]Matthew E. Taylor, Peter Stone. Transfer Learning for Reinforcement Learning Domains:A Survey [J]. Journal of Machine Learning Research,2009,10(2009): 1633-1685.
[103]J. Blitzer, R, McDonald, and F. Pereira. Domain Adaptation with Structural Correspondence Learning [A]. In:Proceedings of International Conference on Empirical Methods in Natural Language Processing [C], Sydney, Australia. 2006:120-128.
[104]M. Rohrbach, M. Stark, G. Szarvas, et al. What Helps Where and Why? Semantic Relatedness for Knowledge Transfer [A]. In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], San Francisco, USA.2010:910-917.
[105]Moez Ben Haj Hmida, Yahya Slimani. Meta-learning in Grid-based Data Mining Systems [J]. International Journal of Communication Networks and Distributed Systems,2010,5(3):214-228.
[106]Loeff Nicolas. A New Framework for Semi-supervised Multitask Learning [D]. Illinois:The University of Illinois at Urbana-Champaign,2009.
[107]D. Zhou, C. J. C. Burges. Spectral Clustering and Transductive Learning with Multiple Views [A]. In:Proceedings of the 24th International Conference on Machine Learning [C], Oregon, USA.2007:1159-1166.
[108]W. Dai, G. Xue, Q. Yang, et al. Transferring Naive Bayes Classifiers for Text Classification [A]. In:Proceedings of 22nd National Conference on Artificial Intelligence [C], Vancouver, Canada.2007:540-545.
[109]W. Dai, Y. Chen, G. Xue, et al. Translated Learning:Transfer Learning across Different Feature Spaces [A]. In:Proceedings of the 23rd Annual Conference on Neural Information Processing Systems (NIPS) [C], Vancouver, Canada.2008: 353-360.
[110]Pan Jialin. Feature-based Transfer Learning with Real-world Application [D]. Hong Kong:The Hong Kong University of Science and Technology,2010.
[111]Y. Yi, G. Doretto. Boosting for Transfer Learning with Multiple Sources [A]. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], San Francisco, USA.2010:1855-1862.
[112]R. Raina, A. Battle, H. Lee, et al. Self-taught Learning:Transfer Learning from Unlabeled Data [A]. In:Proceedings of the 24th International Conference on Machine Learning [C], Oregon, USA.2007:759-766.
[113]W. Dai, Q. Yang, G. R. Xue, et al. Self-taught Clustering [A]. In:Proceedings of the 25th Annual International Conference on Machine Learning [C], Helsinki, Finland.2008:200-207.
[114]W. Dai, Y. Chen, G R. Xue, et al. Translated Learning:Transfer Learning across Different Feature Spaces [A]. In:Proceedings of the 23 rd Annual Conference on Neural Information Processing Systems (NIPS) [C], Whistler, Canada.2009: 353-360.
[115]Selen Uguroglu, Jaime Carbonell. Feature Selection for Transfer Learning [M]. Lecture Notes in Computer Science (LNCS), Springer,2011:430-442.
[116]Fernando Fernandez, Javier Garcia, and Manuela Veloso. Probabilistic Policy Reuse for Inter-task Transfer Learning [J]. Robotics and Autonomous Systems, 2010,58(7):866-871.
[117]W. Dai, Q. Yang, G. Xue, et al. Boosting for Transfer Learning [A]. In: Proceedings of the 24th International Conference on Machine Learning [C], Corvallis, Oregon, USA.2007:193-200.
[118]A. Quattoni, M. Collins, and T. Darrell. Transfer Learning for Image Classification with Sparse Prototype Representations [A]. In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], Anchorage, Alaska, USA.2008:1-8.
[119]Si Si, Dacheng Tao, and Bo Geng. Bregman Divergence-Based Regularization for Transfer Subspace Learning [J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(7):929-942.
[120]Erheng Zhong, Wei Fan, Qiang Yang, et al. Cross Validation Framework to Choose amongst Models and Datasets for Transfer Learning [M], Lecture Notes in Computer Science (LNCS), Springer,2010:547-562.
[121]M. Rosenstein, Z. Marx, and L. P. Kaelbling. To Transfer or Not to Transfer [A]. In:Proceedings of International Conference on Neural Information Processing Systems (NIPS) [C], Vancouver, Canada.2005:1-4.
[122]Bin Cao, Sinno Jialin Pan, Yu Zhang, et al. Adaptive Transfer Learning [A]. In: Proceedings of the 24th AAAI Conference on Artificial Intelligence [C], Atlanta, USA.2010:407-412.
[123]Ke Zhou, Xue Gui-Rong, and Qiang Yang. Learning with Positive and Unlabeled Examples Using Topic-sensitive PLSA [J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(1):46-58.
[124]Akinori Fujino, Naonori Ueda, and Masaaki Nagata. A Robust Semi-supervised Classification Method for Transfer Learning [A]. In:Proceedings of International Conference on Information and Knowledge Management [C], Toronto, Ontario, Canada.2010:379-388.
[125]L. Feifei, P. Perona. A Bayesian Hierarchical Model for Learning Natural Scene Categories [A]. In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition [C], San Diego, USA.2005:524-531.
[126]S. Lazebnik, C. Schmid, and J. Ponce. Beyond Bags of Features:Spatial Pyramid Matching for Recognizing Natural Scene Categories [A]. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition [C], New York, USA,2006:2169-2178.
[127]A. Bosch, A. Zisserman, and X. Munoz. Scene Classification Using A Hybrid Generative/Discriminative Approach [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2008,30(4):712-727.
[128]N. Rasiwasia, N. Vasconcelos. Scene Classification with Low-dimensional Semantic Spaces and Weak Supervision [A]. In:Proceedings of IEEE Conference on Computer Vision and Pattern Recognition [C], Alaska, USA. 2008:1-6.
[129]D. Lowe. Distinctive Image Features from Scale-Invariant Key-points [J]. International Journal of Computer Vision,2004,60(2):91-110.
[130]Herbert Bay, Andreas Eelaars, and Luc Van Gool. Speed-Up Robust Features (SURF) [J]. Computer Vision and Image Understanding,2008,110(3):346-359.
[131]David Gossow, Peter Decker, and Dietrich Paulus. An Evaluation of Open Source SURF Implementations [M]. Lecture Notes in Computer Science (LNCS), Springer,201.1:169-179.
[132]Brown M., Gang H., and Winder S. Discriminative Learning of Local Image Descriptors [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(1):43-57.
[133]Yue Jiang, Runshen Wang, and Peng Zhang. Texture Description based on Multiresolution Moments of Image Histograms [J]. Optical Engineering,2008, 47(3):37005:1-9.
[134]S. T. Roweis, L. K. Saul. Nonlinear Dimensionality Reduction by Locally Linear Embedding [J]. Science,2000,290(5500):2323-2326.
[135]Bingyu Sun, Xiaoming Zhang, Jiuyong Li, et al. Feature Fusion Using Locally Linear Embedding for Classification [J]. IEEE Transactions on Neural Networks,2010,21(1):163-168.
[136]Shiqing Zhang. Enhanced Supervised Locally Linear Embedding [J]. Pattern Recognition Letters,2009,30(13):1208-1218.
[137]J. Yugang, N. ChongWah, J. Yang. Towards Optimal Bag-of-features for Object Categorization and Semantic Video Retrieval [A]. In:Proceedings of ACM International Conference on Image and Video Retrieval [C], New York, USA. 2007:494-501.
[138]Zhe Wang, Guizhong Liu, Xueming Qian, et al. An Approach to the Compact and Efficient Visual Codebook Based on SIFT Descriptor [M]. Lecture Notes in Computer Science (LNCS), Springer,2010:461-469.
[139]Darui Li, Linjun Yang, Xiansheng Hua, et al. Large-scale Robust Visual Codebook Construction [A]. In:Proceedings of International Conference on Multimedia [C], Firenze, Italy.2010:1183-1186.
[140]Luo Juan, Oubong Gwun. A Comparison of SIFT, PCA-SIFT and SURF [J]. International Journal of Image Processing,2010,3(4):143-152.
[141]江悦,王润生.基于多特征潜在主题模型的场景图像分类[J].信号处理,2010,26(4):530-544.
[142]N. Bolshakova, F. Azuaje. Cluster Validation Techniques for Genome Expression Data [J]. Signal Processing,2003,83(4):825-833.
[143]Michael Steinbach, George Karypis, and Vipin Kumar. A Comparison of Document Clustering Techniques [A]. In:Proceedings of the 6th ACM International Conference on Knowledge Discovery and Data Mining [C], Boston, USA.2000:1-20.
[144]R. F. S. Filho, A. J. M. Traina, T. Cactano, et al. Similarity Search without Tears: The OMNI Family of All-Purpose Access Methods [A]. In:Proceedings of the 17th International Conference on Data Engineering [C], Heidelberg, Germany. 2001:623-630.
[145]Slobodan Petrovic, Gonzalo Alvarez, Agustin Orfila, et al. Labelling Clusters in an Intrusion Detection System Using a Combination of Clustering Evaluation Techniques [A]. In:Proceedings of the 39th Hawaii International Conference on System Sciences, Hawaii, USA.2006:129-140.
[146]Chih-Chung Chang, Chih-Jen Lin. LIBSVM:A Library for Support Vector Machines,2001. Software available at http://www.csie.ntu.edu.tw/-cjlin/libsvm.
[147]A. Oliva, A. Torralba. Modeling the Shape of the Scene:A Holistic Representation of the Spatial Envelope [J]. International Journal of Computer Vision,2001,42(3):145-175.
[148]D. M. Blei, Andrew Y. Ng, and Michael I. Jordan. Latent Dirichlet Allocation [J]. Journal of Machine Learning Research,2003,3(2003):993-1022.
[149]D. M. Blei. Probabilistic Topic Models [J]. Communications of the ACM,2012, 55(4):77-84.
[150]A. Azevedo-Filho, R. Shachter. Laplace's Method Approximations for Probabilistic Inference in Belief Networks with Continuous Variables [A]. In: Proceedings of the 10th International Conference on Uncertainty in Artificial Intelligence [C], San Francisco, USA.1994:28-36.
[151]Christopher M. Bishop. Pattern Recognition and Machine Learning [M]. Springer,2006:276-461.
[152]Bolstad, M. William. Understanding Computational Bayesian Statistics [M]. John Wiley,2010:89-110.
[153]George Casella, Edward I. George. Explaining the Gibbs sampler [J]. The American Statistician,1992,46(3):167-174.
[154]P. Duygulu, K. Barnard, J. F. G. de Freitas, et al. Object Recognition as Machine Translation:Learning a Lexicon for a Fixed Image Vocabulary [M]. Lecture Notes in Computer Science (LNCS), Springer,2006:349-354.
[155]S. Feng, R. Manmatha, and V. Lavrenko. Multiple Bernoulli Relevance Models for Image and Video Annotation [A]. In:Proceedings of 2004 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) [C], Washington D. C., USA.2004:1002-1009.
[156]V. Lavrenko, R. Manmatha, and J. Jeon. A Model for Learning the Semantics of Pictures [A]. In:Proceedings of the 18th Annual Conference on Neural Information Processing Systems (NIPS) [C], Vancouver, Canada.2004: 553-560.
[157]J. Jeon, V. Lavrenko, and R. Manmatha. Automatic Image Annotation and Retrieval Using Cross-media Relevance Models [A]. In:Proceedings of the 26th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval [C], Toronto, Canada.2003:119-126.
[158]A. Blum, T. Mitchell. Combining Labeled and Unlabeled Data with Co-training [A]. In:Proceedings of the 11th Annual Conference on Computational Learning Theory [C], Madison, USA.1998:92-100.
[159]K. Nigam, R. Ghani. Analyzing the Effectiveness and Applicability of Co-training [A]. In:Proceedings of the 9th International Conference on Information and Knowledge Management [C], McLean, USA.2000:86-93.
[160]I. Muslea, S. Minton, and C. A. Knoblock. Active+Semi-supervised Learning = Robust Multi-view Learning [A]. In Proceedings of the 19th International Conference on Machine Learning [C], Sydney, Australia.2002:435-442.
[161]Sinno Jialin Pan, Qiang Yang. A Survey on Transfer Learning [J]. IEEE Transactions on Knowledge and Data Engineering,22(10):1345-1359.
[162]Ning Zhong. Effective Pattern Discovery for Text Mining [J]. IEEE Transactions on Knowledge and Data Engineering,2012,24(1):30-44.
[163]Lewis David. Naive (Bayes) at Forty:The Independence Assumption in Information Retrieval [A]. In:Proceedings of the 10th European Conference on Machine Learning [C]. Chemnitz, Germany.1998:4-15.
[164]Hani M. Samawia, Martin Dunbar and Ding-Geng Chen. Steady-state Ranked Gibbs Sampler [J]. Journal of Statistical Computation and Simulation,2012, 82(8):1223-1238.
[165]禹思敏.信息论、编码及应用[M].西安：西安电子科技大学出版社,2012:34-41.
[166]A. Hansen Glen, R. W. Douglass, and A. Zardecki. Mesh Enhancement [M]. Imperial College Press,2005:404-408.
[167]I. S. Gradshteyn, I. M. Ryzhik. Tables of Integrals, Series, and Products (6th Edition) [M]. California:Academic Press,2000:1114-1125.
[168]P. N. Tan, M. Steinbach, and V. Kumar. Introduction to Data Mining [M]. Massachusetts:Addison-Wesley,2005:23-78.
[169]张燕平,张铃.机器学习理论与算法[M].北京：科学出版社,2012:168-201.
[170]王肃.基于多Agent的突发事件信息智能监测系统研究[D].北京：北京邮电大学,2011.
[171]Yijun Sun. Iterative RELIEF for Feature Weighting:Algorithms, Theories, and Applications [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2007,29(6):1035-1051.
[172]J. Lafferty, C. Zhai. Document Language Models, Query Models, and Risk Minimization for Information Retrieval [A]. In Proceedings of the 24th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval [C], Orleans, USA.2001:111-119.
[173]B. N. Jiang, X. Q. Ding, L. T. Ma, et al. A Hybrid Feature Selection Algorithm: Combination of Symmetrical Uncertainty and Genetic Algorithms [A]. In Proceedings of the 2nd International Symposium on Optimization and Systems Biology [C], Shanghai, China,2008:152-157.
[174]E. R. Laura and S. Kifian. Theoretical Comparison between the Gini Index and Information Gain Criteria [J]. Analysis of Mathematics and Artificial Intelligence,2004,41(1):77-93.