基于空间特征的多平面支持向量机地形分类
|
摘要
近年来,室外自主移动机器人在野外环境下的有着十分重要的应用,比如在野外救援和月球探测等方面。而室外复杂环境下的地形识别研究是面向移动机器人环境感知和识别的一个重要挑战。针对在室外复杂环境下的光照干扰和遮挡等因素,论文提出了一种基于金字塔化SIFT特征(SIFT Spatial Pyramid Matching,SSPM)与最小二乘相关支持向量机(Least Squares Twin Support Vector Machines,LSTSVM)相结合的地形识别方法。相较于传统的词袋式特征表示,加入了局部和空间信息特征,增强了特征对图像的表现能力,进一步提高了识别率,大大减小了训练时间。再利LSTSVM在组合得到的新特征集上学习,最后在得到的分类器上验证算法的可靠性。
In recent years,autonomous mobile outdoor robots has a very important application in the field environment,such as rescuing and lunar exploration,etc. Terrain recognition research in complex outdoor environment is a big challenge in the field of mobile robot's environment awareness and identification. Considering the factors in the complex outdoor environment,such as lighting disturbance and block,this paper puts forward a kind of terrain recognition method based on Pyramid SIFT features(SIFT Spatial Pyramid Matching,SSPM)associated with the Least Squares Support Vector machine(Least Squares Twin Support Vector,LSTSVM). Compared with the traditional word bag features,the method joins in the space information and local features,which enhances the characteristics of image expression,improves the accuracy and reduces training time. In the end,the LSTSVM is used to study in the new features and analyze the algorithm reliability on the classifier.
In recent years,autonomous mobile outdoor robots has a very important application in the field environment,such as rescuing and lunar exploration,etc. Terrain recognition research in complex outdoor environment is a big challenge in the field of mobile robot's environment awareness and identification. Considering the factors in the complex outdoor environment,such as lighting disturbance and block,this paper puts forward a kind of terrain recognition method based on Pyramid SIFT features(SIFT Spatial Pyramid Matching,SSPM)associated with the Least Squares Support Vector machine(Least Squares Twin Support Vector,LSTSVM). Compared with the traditional word bag features,the method joins in the space information and local features,which enhances the characteristics of image expression,improves the accuracy and reduces training time. In the end,the LSTSVM is used to study in the new features and analyze the algorithm reliability on the classifier.
引文
[1]Wu H,Liu B Z,Su W H,Zhang W C,Sun J G.Bag of words for visual terrain classification:a comprehensive study[J].Journal of Image And Graphics,2016,21(10):1276-1288.
[2]S.Lazebnik.C.Schmid.J.Ponce.Beyond Bags of Features:Spatial Pyramid Matching for Recognizing Natural Scene Categories[J].IEEE Computer Society Conference on Computer Vision and Pattern Recognition,2006,2:2169-2178.
[3]朱旭锋,马彩文,刘波.采用改进词袋模型的空中目标自动分类[J].红外与激光工程,2012,41(5):1384-1388.ZHU Xufeng,MA Caiwen,LIU Bo.Aerial target automatic classification based on improving bag of words model[J].Infrared and Laser Engineering,2012,41(5):1384-1388.
[4]赵春晖,王莹,Masahide KANEKO一种基于词袋模型的图像优化分类方法[J].电子与信息学报,2012,34(09):2064-2070.ZHAO Chunhui,WANG Ying,Masahide KANEKO.An Optimized Method for Image Classification Based on Bag of Words Model[J].Journal of Electronics&Information Technology,2012,34(09):2064-2070.
[5]高常鑫,桑农.整合局部特征和滤波器特征的空间金字塔匹配模型[J].电子学报,2011,39(09):2034-2038.GAO Changxin,SANG Nong.Unifying Local Features and Filterbank Features in the Spatial Pyramid Matching Model[J].Acta Electronica Sinica,2011,39(09):2064-2070.
[6]Jayadeva,Khemchandani R,Chandra S.Twin Support Vector Machines for pattern classification[J].IEEE Trans Pattern Anal Mach Intell,2007,29(5):905-910.
[7]Arun Kumar M,Gopal M.Least squares twin support vector machines for pattern classification[J].Expert Systems with Applications,2009,36(4):7535-7543.
[8]L.Fei-Fei and P.Perona.A Bayesian hierarchical model for learning natural scene categories[J].CVPR,2005:524-531.
[9]Arun Kumar M,Gopal M.Least squares twin support vector machines for pattern classification[J].Expert Systems with Applications,2009,36(4):7535-7543.
[10]业巧林,赵春霞,陈小波.基于正则化技术的对支持向量机特征选择算法[J].计算机研究与发展,2011,48(6):1029-1037.YE Qiaolin,ZHAO Chunxia,CHEN Xiaobo.A Feature Selection Method for TWSVM via a Regularization Technique[J].Journal of Computer Research and Development,2011,48(6):1029-1037.
[11]杨绪兵,陈松灿.基于原型超平面的多类最接近支持向量机[J].计算机研究与发展,2006,43(10):1700-1705.YANG Xubing,CHEN Songcan.Proximal Support Vector Machine Based on Prototypal Multiclassfication Hyperplanes[J].Journal of Computer Research and Development,2006,43(10):1029-1037.
[12]杨绪兵,陈松灿,杨益民.局部化的广义特征值最接近支持向量机[J].计算机学报,2007,30(8):1227-1234.YANG Xubing,CHEN Songcan,YANG Yimin.Localized Proximal Support Vector Machine via Generalized Eigenvalues[J].Chinese Journal of Computers,2007,30(8):1227-1234.
[13]杨绪兵,潘志松,陈松灿.半监督型广义特征值最接近支持向量机[J].模式识别与人工智能,2009,22(3):349-353.YANG Xubing,PAN Zhisong,Chen Songca.Semi-Supervised Proximal Support Vector Machine Via Generalized Eigenvalues[J].Pattern Recognition and Artificial Intelligence,2009,22(3):349-353.
[14]李新德,刘苗苗,徐叶帆,等.一种基于2D和3D SIFT特征级融合的一般物体识别算法[J].电子学报,2005,43(11):2277-2283.LI Xinde,LIU Miaomiao,XU Yefan,et al.A Recognition Algorithm of Generic Object Based on Feature-Level Fusion of 2D and 3D SIFT Descriptors[J].Acta Electronica Sinica,2005,43(11):2277-2283.
[15]Chang C C,Lin C J.LIBSVM:A library for support vector machines[J].Acm Transactions on Intelligent Systems&Technology,2011,2(3):389-396.
[16]Shao Y H,Deng N Y,Chen W J,et al.Improved Generalized Eigenvalue Proximal Support Vector Machine[J].Signal Processing Letters IEEE,2013,20(3):213-216.
[2]S.Lazebnik.C.Schmid.J.Ponce.Beyond Bags of Features:Spatial Pyramid Matching for Recognizing Natural Scene Categories[J].IEEE Computer Society Conference on Computer Vision and Pattern Recognition,2006,2:2169-2178.
[3]朱旭锋,马彩文,刘波.采用改进词袋模型的空中目标自动分类[J].红外与激光工程,2012,41(5):1384-1388.ZHU Xufeng,MA Caiwen,LIU Bo.Aerial target automatic classification based on improving bag of words model[J].Infrared and Laser Engineering,2012,41(5):1384-1388.
[4]赵春晖,王莹,Masahide KANEKO一种基于词袋模型的图像优化分类方法[J].电子与信息学报,2012,34(09):2064-2070.ZHAO Chunhui,WANG Ying,Masahide KANEKO.An Optimized Method for Image Classification Based on Bag of Words Model[J].Journal of Electronics&Information Technology,2012,34(09):2064-2070.
[5]高常鑫,桑农.整合局部特征和滤波器特征的空间金字塔匹配模型[J].电子学报,2011,39(09):2034-2038.GAO Changxin,SANG Nong.Unifying Local Features and Filterbank Features in the Spatial Pyramid Matching Model[J].Acta Electronica Sinica,2011,39(09):2064-2070.
[6]Jayadeva,Khemchandani R,Chandra S.Twin Support Vector Machines for pattern classification[J].IEEE Trans Pattern Anal Mach Intell,2007,29(5):905-910.
[7]Arun Kumar M,Gopal M.Least squares twin support vector machines for pattern classification[J].Expert Systems with Applications,2009,36(4):7535-7543.
[8]L.Fei-Fei and P.Perona.A Bayesian hierarchical model for learning natural scene categories[J].CVPR,2005:524-531.
[9]Arun Kumar M,Gopal M.Least squares twin support vector machines for pattern classification[J].Expert Systems with Applications,2009,36(4):7535-7543.
[10]业巧林,赵春霞,陈小波.基于正则化技术的对支持向量机特征选择算法[J].计算机研究与发展,2011,48(6):1029-1037.YE Qiaolin,ZHAO Chunxia,CHEN Xiaobo.A Feature Selection Method for TWSVM via a Regularization Technique[J].Journal of Computer Research and Development,2011,48(6):1029-1037.
[11]杨绪兵,陈松灿.基于原型超平面的多类最接近支持向量机[J].计算机研究与发展,2006,43(10):1700-1705.YANG Xubing,CHEN Songcan.Proximal Support Vector Machine Based on Prototypal Multiclassfication Hyperplanes[J].Journal of Computer Research and Development,2006,43(10):1029-1037.
[12]杨绪兵,陈松灿,杨益民.局部化的广义特征值最接近支持向量机[J].计算机学报,2007,30(8):1227-1234.YANG Xubing,CHEN Songcan,YANG Yimin.Localized Proximal Support Vector Machine via Generalized Eigenvalues[J].Chinese Journal of Computers,2007,30(8):1227-1234.
[13]杨绪兵,潘志松,陈松灿.半监督型广义特征值最接近支持向量机[J].模式识别与人工智能,2009,22(3):349-353.YANG Xubing,PAN Zhisong,Chen Songca.Semi-Supervised Proximal Support Vector Machine Via Generalized Eigenvalues[J].Pattern Recognition and Artificial Intelligence,2009,22(3):349-353.
[14]李新德,刘苗苗,徐叶帆,等.一种基于2D和3D SIFT特征级融合的一般物体识别算法[J].电子学报,2005,43(11):2277-2283.LI Xinde,LIU Miaomiao,XU Yefan,et al.A Recognition Algorithm of Generic Object Based on Feature-Level Fusion of 2D and 3D SIFT Descriptors[J].Acta Electronica Sinica,2005,43(11):2277-2283.
[15]Chang C C,Lin C J.LIBSVM:A library for support vector machines[J].Acm Transactions on Intelligent Systems&Technology,2011,2(3):389-396.
[16]Shao Y H,Deng N Y,Chen W J,et al.Improved Generalized Eigenvalue Proximal Support Vector Machine[J].Signal Processing Letters IEEE,2013,20(3):213-216.