基于叶片综合特征的阔叶树机器识别研究
|
摘要
为了有效解决传统植物机器识别对象过宽泛、分类特征较单一且正确识别率较低的问题,提出一种专门针对阔叶树的机器识别方法。该方法将阔叶树叶片的形状特征与其纹理特征相结合,构成一个针对阔叶树叶片图像的综合特征向量,以期更为便捷、快速、高效地对阔叶植物进行计算机自动分类识别。首先提取叶片样本的综合特征信息,然后以概率神经网络(Probabilistic Neural Networks PNN)为分类器对所得到的特征信息进行训练,训练好的网络可以用来识别阔叶的类别,从而确定相应阔叶树的种类。本研究有效提取了含有9个分量的阔叶树叶片综合特征向量,通过对PNN分类器的训练,实现了30种阔叶树的快速机器识别,平均正确识别率达98.3%。比较测试表明:若去掉叶片纹理特征,单以其形状特征作为识别依据,平均识别率仅为93.7%。实验证明针对阔叶树叶片的综合特征识别方法,有效弥补了传统单特征识别方法的不足,使得识别效率得到了较大的提高。
本文首次提出基于叶片综合特征的阔叶树机器识别概念,且就此做了比较系统的研究,其中主要的研究工作及结论如下:
(1)本文以阔叶树叶片为具体对象,将其形状特征与纹理特征相结合,挖掘并构建出阔叶综合特征信息,并就此综合特征信息的提取做了较深入的研究。文章结合植物学相关理论对叶脉的脉序类型做了一定的研究,且首次将脉序类型的标定值作为叶片的一个纹理特征应用于阔叶树的机器识别中。
(2)本文利用已有方法对阔叶树叶片的叶脉图像进行了提取,并在此前提下提出一种改进的差分盒维数算法,且利用此算法对所提取出的叶脉图像进行分形维数的计算,测试结果表明:用此算法求取的叶脉图像分形维数有效地克服了传统分形维数求取方法的不足之处,使得纹理图像的分形维数具有良好的动态范围,结果的准确性得到了较大的提高。
(3)首次将概率神经网络(PNN)应用于专门针对阔叶树的机器识别中,且提出一种基于PCA和优化平滑因子的双管道改进PNN分类器模型。
(4)应用Microsoft VC++ 6.0和Matlab 6.5,设计开发了一个针对阔叶树的计算机识别系统,该系统可以对阔叶树叶片图像进行全面而合理的预处理,对预处理后的叶片图像可以进行综合特征的准确提取,最后利用所提取到的综合特征信息完成PNN分类器的训练及识别工作,平均正确识别率达98.3%。
The study is to effectively solve the problem that the objects of traditional plant identification were too broad and the classification features of it were usually not synthetic and the recognition rate was always slightly low. This paper gives one recognition approach, in which the shape features and the texture features of the leaves of broad-leaved trees combine, composing a synthetic feature vector of broad leaves and hoping to realize the computer automatic classification towards broad-leaved plants more convenient, rapidly and efficient. Firstly, the synthetic features of leaf were extracted; Secondly the values of synthetic features of leaf, which had been extracted, were inputted into a classifier, which is Probabilistic Neural Network(sPNN), to be trained; Finally, the PNN trained well could work to classify the broad leaves and the corresponding broad-leaved trees. The synthetic feature vector of broad leaf, which includes 9 elements, was extracted efficiently. By training the classifier PNN, the rapid recognition for thirty kinds of broad-leaved trees was realized and the average correct recognition rate reached 98.3%. Comparison tests demonstrated that if the shape features of broad leaf solely worked as the recognition features without the texture features, the average correct recognition rate just reached 93.7%. The synthetic feature recognition method against broad-leaved trees has effectively made up the drawbacks of traditional recognition method towards unitary feature, strongly advancing the recognition rate.
This paper has demonstrated the notion for the first time that the machine recognition for broad-leaved trees based on leaves’synthetic characteristics, and has made a systematic research about this notion. The following points are the major studys and conclusions of this paper:
(1) This paper made the shape feature and the texture feature of the leaves of broad-leaved trees combine, mining and constructing synthetic feature information of broad leaves, and has done a deep research on the extraction of this synthetic feature information. What’s more, based on corresponding botany theory, the neuration types of vein images of broad leaves have been done some research, and, innovatively, the label values of neuration types were applied to the broad-leaved trees recognition system as a texture feature of broad leaves.
(2) The vein image of broad leaf has been extracted applying existed method, after which, an improved difference box dimension algorithm were proposed to compute the fractal dimension of vein image. The test result demonstrated that the new algorithm could effectively overcome the drawbacks of the traditional fractal dimension computing method, making the fractal dimension of texture image have a good dynamic range, and the accuracy of results has been better enhanced.
(3) Probabilistic Neural Networks (PNN) was first time applied to the broad-leaved trees recognition, what’s more, this paper designed one double-pipes improved PNN classifier model based on PCA and optimization smoothing factor.
(4) The broad-leaved trees recognition system has been designed and developed with Microsoft VC++ 6.0 and Matlab 6.5. This system can give comprehensive and reasonable pre-processing to broad leaf images and get accurate extraction of synthetic features of pre-processed leaf image and finally accomplish the training and identification task of PNN classifier by using the information of synthetic features. The average correct recognition rate reached 98.3%.
本文首次提出基于叶片综合特征的阔叶树机器识别概念,且就此做了比较系统的研究,其中主要的研究工作及结论如下:
(1)本文以阔叶树叶片为具体对象,将其形状特征与纹理特征相结合,挖掘并构建出阔叶综合特征信息,并就此综合特征信息的提取做了较深入的研究。文章结合植物学相关理论对叶脉的脉序类型做了一定的研究,且首次将脉序类型的标定值作为叶片的一个纹理特征应用于阔叶树的机器识别中。
(2)本文利用已有方法对阔叶树叶片的叶脉图像进行了提取,并在此前提下提出一种改进的差分盒维数算法,且利用此算法对所提取出的叶脉图像进行分形维数的计算,测试结果表明:用此算法求取的叶脉图像分形维数有效地克服了传统分形维数求取方法的不足之处,使得纹理图像的分形维数具有良好的动态范围,结果的准确性得到了较大的提高。
(3)首次将概率神经网络(PNN)应用于专门针对阔叶树的机器识别中,且提出一种基于PCA和优化平滑因子的双管道改进PNN分类器模型。
(4)应用Microsoft VC++ 6.0和Matlab 6.5,设计开发了一个针对阔叶树的计算机识别系统,该系统可以对阔叶树叶片图像进行全面而合理的预处理,对预处理后的叶片图像可以进行综合特征的准确提取,最后利用所提取到的综合特征信息完成PNN分类器的训练及识别工作,平均正确识别率达98.3%。
The study is to effectively solve the problem that the objects of traditional plant identification were too broad and the classification features of it were usually not synthetic and the recognition rate was always slightly low. This paper gives one recognition approach, in which the shape features and the texture features of the leaves of broad-leaved trees combine, composing a synthetic feature vector of broad leaves and hoping to realize the computer automatic classification towards broad-leaved plants more convenient, rapidly and efficient. Firstly, the synthetic features of leaf were extracted; Secondly the values of synthetic features of leaf, which had been extracted, were inputted into a classifier, which is Probabilistic Neural Network(sPNN), to be trained; Finally, the PNN trained well could work to classify the broad leaves and the corresponding broad-leaved trees. The synthetic feature vector of broad leaf, which includes 9 elements, was extracted efficiently. By training the classifier PNN, the rapid recognition for thirty kinds of broad-leaved trees was realized and the average correct recognition rate reached 98.3%. Comparison tests demonstrated that if the shape features of broad leaf solely worked as the recognition features without the texture features, the average correct recognition rate just reached 93.7%. The synthetic feature recognition method against broad-leaved trees has effectively made up the drawbacks of traditional recognition method towards unitary feature, strongly advancing the recognition rate.
This paper has demonstrated the notion for the first time that the machine recognition for broad-leaved trees based on leaves’synthetic characteristics, and has made a systematic research about this notion. The following points are the major studys and conclusions of this paper:
(1) This paper made the shape feature and the texture feature of the leaves of broad-leaved trees combine, mining and constructing synthetic feature information of broad leaves, and has done a deep research on the extraction of this synthetic feature information. What’s more, based on corresponding botany theory, the neuration types of vein images of broad leaves have been done some research, and, innovatively, the label values of neuration types were applied to the broad-leaved trees recognition system as a texture feature of broad leaves.
(2) The vein image of broad leaf has been extracted applying existed method, after which, an improved difference box dimension algorithm were proposed to compute the fractal dimension of vein image. The test result demonstrated that the new algorithm could effectively overcome the drawbacks of the traditional fractal dimension computing method, making the fractal dimension of texture image have a good dynamic range, and the accuracy of results has been better enhanced.
(3) Probabilistic Neural Networks (PNN) was first time applied to the broad-leaved trees recognition, what’s more, this paper designed one double-pipes improved PNN classifier model based on PCA and optimization smoothing factor.
(4) The broad-leaved trees recognition system has been designed and developed with Microsoft VC++ 6.0 and Matlab 6.5. This system can give comprehensive and reasonable pre-processing to broad leaf images and get accurate extraction of synthetic features of pre-processed leaf image and finally accomplish the training and identification task of PNN classifier by using the information of synthetic features. The average correct recognition rate reached 98.3%.
引文
[1] Knapp AK, Fay PA, Blair JM, et al. Rainfall Variability, Carbon Cycling, and Plant Species Diversity in a Mesic Grassland[J]. Science, 2002, 5601(298): 2202-2205.
[2] Ahmad IS, Reid JF. Evaluation of Colour Representations for Maize Images[J]. Journal of Agricultural Engineering Research, 1996, 3(63): 185-195.
[3] Seginer I, McClendon RW. Methods for optimal control of the greenhouse environment[J]. Transactions of the ASAE, 1992, 4(35): 1299-1307.
[4] Zhang N, Chaisattapagon C. Effective criteria for weed identification in wheat fields using machine vision[J]. Trans. ASAE, 1995, 38(3): 965-974.
[5] Meyer GE. Textural imaging and discriminant analysis for distinguishing weeds for spot spraying[J]. Transactions of the ASAE, 1998, 41(4): 1189-1197.
[6] Ray TS. Landmark eigenshape analysis: homologous contours: leaf shape in Syngonium[J]. American Journal of Botany, 1992, 1(79): 69-76.
[7] Guyer DE, Miles GE, Gaultney LD, et al. Application of machine vision to shape analysis in leaf and plant identification[J]. Transactions of the ASAE, 1992, 36(1): 163-171.
[8] Yonekawa S, Sakai N, Kitani O. Identification of idealized leaf types using simple dimensionless shape factors by image analysis[J]. Transactions of the ASAE, 1996, 39(4): 1525-1533.
[9] Rui Y, She AC, Huang TS. Modified Fourier descriptors for shape representation-a practical approach[A]. In First International Workshop on Image Databases and MultiMedia Search, Amsterdam, the Netherlands, 1996: 403-413.
[10] Abbasi S, Mokhtarian F, Kittler J. Reliable classification of chrysanthemum leaves through curvature scale space[J]. ICSSTCV, 1997, 97: 284-295.
[11] Mokhtarian F, Abbasi S. Matching shapes with self-intersection: application to leaf classification[J]. IEEE Trans. on Image Processing, 2004, 13(5): 653-661.
[12] Cholhong I, Nishida H. Recognizing plant species by normalized leaf shapes[J]. Vision Interface, 1999: 397-404.
[13] Neto JC, Meyer GE, Jones DD, et al. Plant species identification using Elliptic Fourier leaf shape analysis[J]. Computers and Electronics in Agriculture, 2006, 2(50): 121-134.
[14] Bruno OM, Plotze RO, Falvo M, et al. Fractal dimension applied to plant identification[J]. Inform. Sci. (2008).
[15] Wang ZY, Chi ZR, Feng DG. Fuzzy integral for leaf image retrieval[J]. Proc. Fuzzy Systems, 2002, (1): 372-377.
[16] Wang ZY, Chi ZR, Feng DG. Shape based leaf image retrieval[J]. Image Signal Process, 2003, 150(1): 34-43.
[17] Fu H, Chi ZR, Feng DG, et al. Machine learning techniques for ontology-based leaf classification[J]. IEEE Conference on Control, Automation, Robotics and Vision Conference, 2004, (1): 6-9, 68-686.
[18] Ye YH, Chen Chun, Chi ZR, et al. A computerized plant species recognition system[J]. Proceedings of 2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, 2004:723-726, 20-22.
[19] Qi HN, Yang JG. Sawtooth Feature Extraction of Leaf Edge Based on Support Vector Machine[C]. IEEE, 2003.
[20] Cai YD, Liu XJ, Chou KC. Artificial neural network for predicting protein subcellular location[J]. Computers and Chemistry, 2002, 26: 179-182.
[21] Chaudhuri BB, Sarkar N. Texture segmentation using fractal dimension[J]. IEEE Trans. on Pattern Analysis and Machine Intelligence, 1995, 17(1): 72-77.
[22] Mandelbrot BB. The fractal geometry of nature[M]. San Francisco: Freeman, 1982.
[23] Fernandez E, Jelinek HF. Use of fractal theory in neuroscience: methods, advantages, and potential problems[J]. METHODS, 2001, 24: 309-321.
[24] Philippe PO, Ricarard CD. Performance evaluation for four classes of texture features[J]. Pattern Recognition, 1992, 25(8): 819-833.
[25] Dubuisson MP. Efficacy of fractal features in segmenting images of natural textures[J]. Pattern Recognition Letters, 1994, 15(4): 419-430.
[26] Kasparis T. Texture description using fractal and energy features[J]. Computers & Electrical Engineering, 1995, 21(1): 21-31.
[27] Chaudhuri BB, Sarkar N. Texture segmentation using fractal dimension [J]. IEEE Trans. on Pattern Analysis and Machine Intelligence, 1995, 17(1): 72-77.
[28] Gomm JB, Yu DL. Selecting radial basis function network centers with recursive orthogonal least squares training[J]. IEEE Transactions on Networks, 2000, 11(2): 306-314.
[29] Specht DF. Probabilistic neural networks for classification, mapping, orassociative memory[A]. IEEE International Conference on Neural Networks[C]. San Diego, CA, USA: Lockheed Palo Alto Res. Lab., CA, 1988: 525-532.
[30] Labonte G. On the efficiency of OLS reduced probabilistic neural networks for aircraft-flare discrimination[C]. Proceedings of the International Joint Conference on Neural Networks. New York. IEEE Press, 2003: 2306-2311.
[31] Hui C, Mats N, Sirkka L, et al. Evaluation of PCA methods with improved fault isolation capabilities on a paper machine simulator[J]. Chemometrics and Intelligent Laboratory Systems, 2008.
[32] Fatma L, Kemal P, Sadlk K, et al. Medical diagnosis of atherosclerosis from Carotid Artery Doppler Signals using principal component analysis (PCA), k-NN based weighting pre-processing and Artificial Immune Recognition System (AIRS)[J]. Journal of Biomedical Informatics, 2008, 1(41): 15-23.
[33] Zacharia P, Menti A, Zacharias T. Genetic algorithm-based optimal design of shunt compensators in the presence of harmonics[J]. Electric Power Systems Research, 2008, 4(78): 728-735.
[34] Buyck B. Taxonomists are an endangered species in Europe[J]. Nature, 1999, 401: 321.
[35] Soh LK, Tsatsoulis C. Texture analysis of SAR sea ice imagery using gray level co-occurrence matrices[J]. IEEE Trans. on Geoscience and Remote Sensing, 1999, 2(37): 780-795.
[36] Gibson D, Gaydecki RA. Definition and application of a Fourier domain texture measure: Applications to histological image segmentation[J]. Computers in Biology and Medicine, 1995, 6(25): 551-557.
[37] Diaz G, Cappai C, Setzu MD, et al. Nuclear pattern recognition by two-parameter texture analysis[J]. Computer Methods and Programs in Biomedicine, 1996, 1(49): 1-9.
[38] Tao B, Dickinson BW. Texture Recognition and Image Retrieval Using Gradient Indexing[J]. Journal of Visual Communication and Image Representation, 2000, 3(11): 327-342.
[39] McNitt-Gray MF, Wyckoff N, Sayre JW, et al. The effects of co-occurrence matrix based texture parameters on the classification of solitary pulmonary nodules imaged on computed tomography[J]. Computerized Medical Imaging and Graphics, 1999, 6(23): 339-348.
[40] Latif-Amet A, Ertuzun A, Ergil A. An efficient method for texture defect detection: sub-band domain co-occurrence matrices[J]. Image and Vision Computing, 2000, 6-7(18): 543-553.
[41] Gadelmawla ES. A vision system for surface roughness characterization using the gray level co-occurrence matrix[J]. NDT and E International, 2004, 7(37): 577-588.
[42] Mandal MK, Aboulnasr T, Panchanathan S. Fast Wavelet Histogram Techniques for Image Indexing[J]. Computer Vision and Image Understanding, 1999, 1-2(75): 99-110.
[43] Mausumi A, Malay KK. An adaptive approach to unsupervised texture segmentation using M-Band wavelet transform[J]. Signal Processing, 2001, 7(81): 1337-1356.
[44] Tang X, Stewart WK. Optical and Sonar Image Classification:Wavelet Packet Transform vs Fourier Transform[J]. Computer Vision and Image Understanding, 2000, 1(79): 25-46.
[45] Angel P, Morris C. Analyzing the Mallat Wavelet Transform to Delineate Contour and Textural Features[J]. Computer Vision and Image Understanding, 2000, 3(80): 267-288.
[46] Liapis S, Sifakis E, Tziritas G. Color and texture segmentation using wavelet frame analysis, deterministic relaxation, and fast marching algorithms[J]. Journal of Visual Communication and Image Representation, 2004, 1(15): 1-26.
[47] 耿 楠, 何东健. 小麦生长信息计算机视觉检测技术研究[J]. 农业工程学报, 2001, 17(1): 136-139.
[48] 中华木材商务网. 阔叶树. http://knology.chinaccm.com/phrase-2006011014310300322.html.
[49] 李 巍, 张春生. 浅谈增加阔叶树种比重在更新造林中的作用[J]. 林业勘查设计, 2002, 3: 47-48.
[50] 李正理, 殷宏章. 植物形态.植物生理[M]. 中国大百科全书出版社, 1988: 174.
[51] 张 彪, 淮虎银, 金银根. 植物分类学实验[M]. 东南大学出版社, 2007: 3-5.
[52] 周世权, 马恩伟. 植物分类学[M]. 北京: 林业出版社, 1995.
[53] (英)V·H·海吾德, 植物分类学(第二版)[M]. 北京: 科学出版社, 1979.
[54] 徐克学.生物数学[M]. 北京: 科学出版社, 1998.
[55] 唐守正, 李 勇. 生物数学模型的统计学基础[M]. 北京: 科学出版社, 2002.
[56] 洪德元, 葛 颂, 张大明等. 植物濒危机制研究的原理和方法[A]. 生物多样性研究进展: 首届全国生物多样性保护与持续利用研讨会论文集.
[57] 孙德敏, 秦琳琳, 王 永, 等. 基于纹理的植物叶片红外图像的模式识别[J]. 模式识别与人工智能, 2004, 17(1): 77-81.
[58] 秦琳琳. 植物叶温的模式识别与营养液中 pH 值的监控研究[D]. 中国科学技术大学硕士学位论文, 2003.
[59] 刘 敏. 基于分形的田间杂草图像分析与识别的研究[D]. 北京工业大学硕士学位论文, 2002.
[60] 张健钦. 计算机视觉技术在杂草识别中的应用研究[D]. 河北农业大学硕士学位论文, 2003.
[61] 于 峰. 计算机图像处理技术在植物氮营养诊断中的应用及其软件开发[D]. 中国农业大学硕士学位论文, 2003.
[62] 傅 星, 卢汉清. 应用计算机进行植物自动分类的初步研究[J], 生态学杂志, 1994, 13(2): 69-71.
[63] 傅 弘, 池哲儒, 常 杰, 等. 基于人工神经网络的叶脉信息提取[J]. 植物学通报, 2004, 21(4): 429-436.
[64] 祁亨年, 寿 韬, 金水虎. 基于叶片特征的计算机辅助植物识别模型[J]. 浙江林学院学报, 2003, 20(3): 281-284.
[65] 祁亨年. 植物外观特征自动获取及计算机辅助植物分类与识别[J]. 浙江林学院学报, 2004, 21(2): 222-227.
[66] 王晓峰, 黄德双, 杜吉祥, 等. 叶片图像特征提取与识别技术的研究[J]. 计算机工程与应用, 2006, 3: 191-192.
[67] 朱 静, 田兴军, 陈 彬, 等. 植物叶型的计算机识别系统[J]. 植物学通报, 2005, 22(5): 599-604.
[68] 张 蕾. 基于叶片特征的计算机自动植物种类识别研究[D]. 东北师范大学硕士学位论文, 2007.
[69] 杨顺辽. 弗里曼链码法在图像处理中的应用[J]. 中国水运(理论版). 2006, 1(4): 24
[70] 周培德. 计算几何——算法分析与设计[M]. 北京: 清华大学山版社, 2003.
[71] 何东健, 耿 楠, 张义宽. 数字图象处理[M]. 西安:西安电子科技大学出版社, 2003.
[72] 贾永红. 数字图像处理[M]. 武汉: 武汉大学出版社, 2003.
[73] 彭瑞东, 谢和平, 鞠 杨. 二维数字图像分形维数的计算方法[J]. 中国矿业大学学报, 2004, 33(1): 19-25.
[74] 王印培, 陈 进, 孙晓明, 等. 珠光体球化的分形研究[J]. 理化检验—物理分册, 2003, 39(3): 129-133.
[75] 屈朝霞, 张汉谦. 材料科学中的分形理论应用进展[J]. 宇航材料工艺, 1999(5): 5-9.
[76] 李留仁, 赵艳艳, 李忠兴, 等. 多孔介质微观孔隙结构分形特征及分形维数的意义[J]. 石油大学学报(自然科学版), 2004, 28(3): 105-107.
[77] 何隆华, 赵 宏. 水系的分形维数及其含义[J]. 地理科学, 1996, 16(2): 124-128.
[78] 贺 伟, 钟孚勋. 储层岩石孔隙的分形结构研究和应用[J]. 天然气工业, 2000, 20(2): 67-70.
[79] 孙博文, 张本祥. 上证指数的分形结构及其盒维数测量的研究[J]. 哈尔滨理工大学学报, 2001, 33(6): 25-28.
[80] 曼德尔布洛特. 分形对象—形、机遇和维数[M]. 文志英, 苏虹, 译. 北京:世界图书出版公司, 1999: 34-38.
[81] 孙博玲. 分形维数及其测量方法[J]. 东北林业大学学报, 2004, 32(3): 116-119.
[82] 许 东, 吴 铮. 基于 MATLAB6.x 的系统分析与设计——神经网络[M]. 西安电子科技大学出版社, 2002: 24.
[83] 邢 杰, 萧德云. 基于 PCA 的概率神经网络结构优化[J]. 清华大学学报(自然科学版), 2008, 1(48): 141-144.
[84] 陈 波, 郭壮志. 基于优化平滑因子σ 的概率神经网络的变压器故障诊断方法研究[J]. 现代电力, 2007, 2(24): 46-46.
[85] 李双虎, 张艳秋. 基于 PCA 的指纹识别算法研究[J]. 河北省科学院学报, 2007, 2(24): 1-2.
[86] 杨 静, 毛宗源. 基于 PCA 和神经网络的识别方法研究[J]. 计算机工程与应用, 2007, 43(25): 246.
[87] 施光林, 史维祥. 遗传算法及其研究与应用新进展[J]. 科技导报, 1997, 4: 35-37.
[88] 杨晓鹏, 宗 明. Visual C++实用编程技术[M]. 北京: 中国水利水电出版社, 2002.
[89] 何 强, 何 英. MATLAB 扩展编程[M]. 北京: 清华大学出版社, 2002.
[90] 郭 虹, 薄云飞, 林 冬. VC++与 MATLAB 混合编程技术研究[J]. 计算机工程, 2002, 9(28): 269-271.
[91] 李志明, 曹 杰, 谢求成. VC++与 MATLAB 的混合编程[J]. 遥测遥控, 1999, 6(20): 59-61.
[92] 许雪开. VC++与 MATLAB 混合编程[J]. 机电工程, 2007, 2(24): 26-28.
[93] 谢 芳. CV 与 MATLAB 数学库和图形库的混合编程应用[J]. 2005, 1(24): 38-40.
[94] 新电脑课堂编委会. Visual C++编程篇[M]. 北京: 电子工业大学出版社, 2004: 133-271.
[95] 尹立民, 王兴东. VC++应用编程 150 例[M]. 北京: 电子工业出版社, 2004: 1-135.
[96] 伍 怡. 三维植物器官的几何建模[D]. 中国科学院自动化研究所硕士学位论文, 2001.
[97] 严红平. 基于子结构和水力模型的虚拟植物建模和可视化研究[D]. 中国科学院自动化研究所博士后论文, 2002.
[98] 王 矼. 虚拟植物器官库几何造型软件与相关问题研究[D]. 中国科学院自动化研究所硕士学位论文, 2003.
[99] 赵 星. 忠实于植物学的虚拟植物生长研究[D]. 中国科学技术大学博士学位论文, 2001.
[100] 赵 星. 虚拟植物生长中几个问题的研究[D]. 中国科学院自动化研究所博士后论文, 2003.
[101] 丁维龙. 虚拟植物生长模型及其与智能系统集成研究[D]. 中国科学技术大学博士学位论文, 2004.
[2] Ahmad IS, Reid JF. Evaluation of Colour Representations for Maize Images[J]. Journal of Agricultural Engineering Research, 1996, 3(63): 185-195.
[3] Seginer I, McClendon RW. Methods for optimal control of the greenhouse environment[J]. Transactions of the ASAE, 1992, 4(35): 1299-1307.
[4] Zhang N, Chaisattapagon C. Effective criteria for weed identification in wheat fields using machine vision[J]. Trans. ASAE, 1995, 38(3): 965-974.
[5] Meyer GE. Textural imaging and discriminant analysis for distinguishing weeds for spot spraying[J]. Transactions of the ASAE, 1998, 41(4): 1189-1197.
[6] Ray TS. Landmark eigenshape analysis: homologous contours: leaf shape in Syngonium[J]. American Journal of Botany, 1992, 1(79): 69-76.
[7] Guyer DE, Miles GE, Gaultney LD, et al. Application of machine vision to shape analysis in leaf and plant identification[J]. Transactions of the ASAE, 1992, 36(1): 163-171.
[8] Yonekawa S, Sakai N, Kitani O. Identification of idealized leaf types using simple dimensionless shape factors by image analysis[J]. Transactions of the ASAE, 1996, 39(4): 1525-1533.
[9] Rui Y, She AC, Huang TS. Modified Fourier descriptors for shape representation-a practical approach[A]. In First International Workshop on Image Databases and MultiMedia Search, Amsterdam, the Netherlands, 1996: 403-413.
[10] Abbasi S, Mokhtarian F, Kittler J. Reliable classification of chrysanthemum leaves through curvature scale space[J]. ICSSTCV, 1997, 97: 284-295.
[11] Mokhtarian F, Abbasi S. Matching shapes with self-intersection: application to leaf classification[J]. IEEE Trans. on Image Processing, 2004, 13(5): 653-661.
[12] Cholhong I, Nishida H. Recognizing plant species by normalized leaf shapes[J]. Vision Interface, 1999: 397-404.
[13] Neto JC, Meyer GE, Jones DD, et al. Plant species identification using Elliptic Fourier leaf shape analysis[J]. Computers and Electronics in Agriculture, 2006, 2(50): 121-134.
[14] Bruno OM, Plotze RO, Falvo M, et al. Fractal dimension applied to plant identification[J]. Inform. Sci. (2008).
[15] Wang ZY, Chi ZR, Feng DG. Fuzzy integral for leaf image retrieval[J]. Proc. Fuzzy Systems, 2002, (1): 372-377.
[16] Wang ZY, Chi ZR, Feng DG. Shape based leaf image retrieval[J]. Image Signal Process, 2003, 150(1): 34-43.
[17] Fu H, Chi ZR, Feng DG, et al. Machine learning techniques for ontology-based leaf classification[J]. IEEE Conference on Control, Automation, Robotics and Vision Conference, 2004, (1): 6-9, 68-686.
[18] Ye YH, Chen Chun, Chi ZR, et al. A computerized plant species recognition system[J]. Proceedings of 2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, 2004:723-726, 20-22.
[19] Qi HN, Yang JG. Sawtooth Feature Extraction of Leaf Edge Based on Support Vector Machine[C]. IEEE, 2003.
[20] Cai YD, Liu XJ, Chou KC. Artificial neural network for predicting protein subcellular location[J]. Computers and Chemistry, 2002, 26: 179-182.
[21] Chaudhuri BB, Sarkar N. Texture segmentation using fractal dimension[J]. IEEE Trans. on Pattern Analysis and Machine Intelligence, 1995, 17(1): 72-77.
[22] Mandelbrot BB. The fractal geometry of nature[M]. San Francisco: Freeman, 1982.
[23] Fernandez E, Jelinek HF. Use of fractal theory in neuroscience: methods, advantages, and potential problems[J]. METHODS, 2001, 24: 309-321.
[24] Philippe PO, Ricarard CD. Performance evaluation for four classes of texture features[J]. Pattern Recognition, 1992, 25(8): 819-833.
[25] Dubuisson MP. Efficacy of fractal features in segmenting images of natural textures[J]. Pattern Recognition Letters, 1994, 15(4): 419-430.
[26] Kasparis T. Texture description using fractal and energy features[J]. Computers & Electrical Engineering, 1995, 21(1): 21-31.
[27] Chaudhuri BB, Sarkar N. Texture segmentation using fractal dimension [J]. IEEE Trans. on Pattern Analysis and Machine Intelligence, 1995, 17(1): 72-77.
[28] Gomm JB, Yu DL. Selecting radial basis function network centers with recursive orthogonal least squares training[J]. IEEE Transactions on Networks, 2000, 11(2): 306-314.
[29] Specht DF. Probabilistic neural networks for classification, mapping, orassociative memory[A]. IEEE International Conference on Neural Networks[C]. San Diego, CA, USA: Lockheed Palo Alto Res. Lab., CA, 1988: 525-532.
[30] Labonte G. On the efficiency of OLS reduced probabilistic neural networks for aircraft-flare discrimination[C]. Proceedings of the International Joint Conference on Neural Networks. New York. IEEE Press, 2003: 2306-2311.
[31] Hui C, Mats N, Sirkka L, et al. Evaluation of PCA methods with improved fault isolation capabilities on a paper machine simulator[J]. Chemometrics and Intelligent Laboratory Systems, 2008.
[32] Fatma L, Kemal P, Sadlk K, et al. Medical diagnosis of atherosclerosis from Carotid Artery Doppler Signals using principal component analysis (PCA), k-NN based weighting pre-processing and Artificial Immune Recognition System (AIRS)[J]. Journal of Biomedical Informatics, 2008, 1(41): 15-23.
[33] Zacharia P, Menti A, Zacharias T. Genetic algorithm-based optimal design of shunt compensators in the presence of harmonics[J]. Electric Power Systems Research, 2008, 4(78): 728-735.
[34] Buyck B. Taxonomists are an endangered species in Europe[J]. Nature, 1999, 401: 321.
[35] Soh LK, Tsatsoulis C. Texture analysis of SAR sea ice imagery using gray level co-occurrence matrices[J]. IEEE Trans. on Geoscience and Remote Sensing, 1999, 2(37): 780-795.
[36] Gibson D, Gaydecki RA. Definition and application of a Fourier domain texture measure: Applications to histological image segmentation[J]. Computers in Biology and Medicine, 1995, 6(25): 551-557.
[37] Diaz G, Cappai C, Setzu MD, et al. Nuclear pattern recognition by two-parameter texture analysis[J]. Computer Methods and Programs in Biomedicine, 1996, 1(49): 1-9.
[38] Tao B, Dickinson BW. Texture Recognition and Image Retrieval Using Gradient Indexing[J]. Journal of Visual Communication and Image Representation, 2000, 3(11): 327-342.
[39] McNitt-Gray MF, Wyckoff N, Sayre JW, et al. The effects of co-occurrence matrix based texture parameters on the classification of solitary pulmonary nodules imaged on computed tomography[J]. Computerized Medical Imaging and Graphics, 1999, 6(23): 339-348.
[40] Latif-Amet A, Ertuzun A, Ergil A. An efficient method for texture defect detection: sub-band domain co-occurrence matrices[J]. Image and Vision Computing, 2000, 6-7(18): 543-553.
[41] Gadelmawla ES. A vision system for surface roughness characterization using the gray level co-occurrence matrix[J]. NDT and E International, 2004, 7(37): 577-588.
[42] Mandal MK, Aboulnasr T, Panchanathan S. Fast Wavelet Histogram Techniques for Image Indexing[J]. Computer Vision and Image Understanding, 1999, 1-2(75): 99-110.
[43] Mausumi A, Malay KK. An adaptive approach to unsupervised texture segmentation using M-Band wavelet transform[J]. Signal Processing, 2001, 7(81): 1337-1356.
[44] Tang X, Stewart WK. Optical and Sonar Image Classification:Wavelet Packet Transform vs Fourier Transform[J]. Computer Vision and Image Understanding, 2000, 1(79): 25-46.
[45] Angel P, Morris C. Analyzing the Mallat Wavelet Transform to Delineate Contour and Textural Features[J]. Computer Vision and Image Understanding, 2000, 3(80): 267-288.
[46] Liapis S, Sifakis E, Tziritas G. Color and texture segmentation using wavelet frame analysis, deterministic relaxation, and fast marching algorithms[J]. Journal of Visual Communication and Image Representation, 2004, 1(15): 1-26.
[47] 耿 楠, 何东健. 小麦生长信息计算机视觉检测技术研究[J]. 农业工程学报, 2001, 17(1): 136-139.
[48] 中华木材商务网. 阔叶树. http://knology.chinaccm.com/phrase-2006011014310300322.html.
[49] 李 巍, 张春生. 浅谈增加阔叶树种比重在更新造林中的作用[J]. 林业勘查设计, 2002, 3: 47-48.
[50] 李正理, 殷宏章. 植物形态.植物生理[M]. 中国大百科全书出版社, 1988: 174.
[51] 张 彪, 淮虎银, 金银根. 植物分类学实验[M]. 东南大学出版社, 2007: 3-5.
[52] 周世权, 马恩伟. 植物分类学[M]. 北京: 林业出版社, 1995.
[53] (英)V·H·海吾德, 植物分类学(第二版)[M]. 北京: 科学出版社, 1979.
[54] 徐克学.生物数学[M]. 北京: 科学出版社, 1998.
[55] 唐守正, 李 勇. 生物数学模型的统计学基础[M]. 北京: 科学出版社, 2002.
[56] 洪德元, 葛 颂, 张大明等. 植物濒危机制研究的原理和方法[A]. 生物多样性研究进展: 首届全国生物多样性保护与持续利用研讨会论文集.
[57] 孙德敏, 秦琳琳, 王 永, 等. 基于纹理的植物叶片红外图像的模式识别[J]. 模式识别与人工智能, 2004, 17(1): 77-81.
[58] 秦琳琳. 植物叶温的模式识别与营养液中 pH 值的监控研究[D]. 中国科学技术大学硕士学位论文, 2003.
[59] 刘 敏. 基于分形的田间杂草图像分析与识别的研究[D]. 北京工业大学硕士学位论文, 2002.
[60] 张健钦. 计算机视觉技术在杂草识别中的应用研究[D]. 河北农业大学硕士学位论文, 2003.
[61] 于 峰. 计算机图像处理技术在植物氮营养诊断中的应用及其软件开发[D]. 中国农业大学硕士学位论文, 2003.
[62] 傅 星, 卢汉清. 应用计算机进行植物自动分类的初步研究[J], 生态学杂志, 1994, 13(2): 69-71.
[63] 傅 弘, 池哲儒, 常 杰, 等. 基于人工神经网络的叶脉信息提取[J]. 植物学通报, 2004, 21(4): 429-436.
[64] 祁亨年, 寿 韬, 金水虎. 基于叶片特征的计算机辅助植物识别模型[J]. 浙江林学院学报, 2003, 20(3): 281-284.
[65] 祁亨年. 植物外观特征自动获取及计算机辅助植物分类与识别[J]. 浙江林学院学报, 2004, 21(2): 222-227.
[66] 王晓峰, 黄德双, 杜吉祥, 等. 叶片图像特征提取与识别技术的研究[J]. 计算机工程与应用, 2006, 3: 191-192.
[67] 朱 静, 田兴军, 陈 彬, 等. 植物叶型的计算机识别系统[J]. 植物学通报, 2005, 22(5): 599-604.
[68] 张 蕾. 基于叶片特征的计算机自动植物种类识别研究[D]. 东北师范大学硕士学位论文, 2007.
[69] 杨顺辽. 弗里曼链码法在图像处理中的应用[J]. 中国水运(理论版). 2006, 1(4): 24
[70] 周培德. 计算几何——算法分析与设计[M]. 北京: 清华大学山版社, 2003.
[71] 何东健, 耿 楠, 张义宽. 数字图象处理[M]. 西安:西安电子科技大学出版社, 2003.
[72] 贾永红. 数字图像处理[M]. 武汉: 武汉大学出版社, 2003.
[73] 彭瑞东, 谢和平, 鞠 杨. 二维数字图像分形维数的计算方法[J]. 中国矿业大学学报, 2004, 33(1): 19-25.
[74] 王印培, 陈 进, 孙晓明, 等. 珠光体球化的分形研究[J]. 理化检验—物理分册, 2003, 39(3): 129-133.
[75] 屈朝霞, 张汉谦. 材料科学中的分形理论应用进展[J]. 宇航材料工艺, 1999(5): 5-9.
[76] 李留仁, 赵艳艳, 李忠兴, 等. 多孔介质微观孔隙结构分形特征及分形维数的意义[J]. 石油大学学报(自然科学版), 2004, 28(3): 105-107.
[77] 何隆华, 赵 宏. 水系的分形维数及其含义[J]. 地理科学, 1996, 16(2): 124-128.
[78] 贺 伟, 钟孚勋. 储层岩石孔隙的分形结构研究和应用[J]. 天然气工业, 2000, 20(2): 67-70.
[79] 孙博文, 张本祥. 上证指数的分形结构及其盒维数测量的研究[J]. 哈尔滨理工大学学报, 2001, 33(6): 25-28.
[80] 曼德尔布洛特. 分形对象—形、机遇和维数[M]. 文志英, 苏虹, 译. 北京:世界图书出版公司, 1999: 34-38.
[81] 孙博玲. 分形维数及其测量方法[J]. 东北林业大学学报, 2004, 32(3): 116-119.
[82] 许 东, 吴 铮. 基于 MATLAB6.x 的系统分析与设计——神经网络[M]. 西安电子科技大学出版社, 2002: 24.
[83] 邢 杰, 萧德云. 基于 PCA 的概率神经网络结构优化[J]. 清华大学学报(自然科学版), 2008, 1(48): 141-144.
[84] 陈 波, 郭壮志. 基于优化平滑因子σ 的概率神经网络的变压器故障诊断方法研究[J]. 现代电力, 2007, 2(24): 46-46.
[85] 李双虎, 张艳秋. 基于 PCA 的指纹识别算法研究[J]. 河北省科学院学报, 2007, 2(24): 1-2.
[86] 杨 静, 毛宗源. 基于 PCA 和神经网络的识别方法研究[J]. 计算机工程与应用, 2007, 43(25): 246.
[87] 施光林, 史维祥. 遗传算法及其研究与应用新进展[J]. 科技导报, 1997, 4: 35-37.
[88] 杨晓鹏, 宗 明. Visual C++实用编程技术[M]. 北京: 中国水利水电出版社, 2002.
[89] 何 强, 何 英. MATLAB 扩展编程[M]. 北京: 清华大学出版社, 2002.
[90] 郭 虹, 薄云飞, 林 冬. VC++与 MATLAB 混合编程技术研究[J]. 计算机工程, 2002, 9(28): 269-271.
[91] 李志明, 曹 杰, 谢求成. VC++与 MATLAB 的混合编程[J]. 遥测遥控, 1999, 6(20): 59-61.
[92] 许雪开. VC++与 MATLAB 混合编程[J]. 机电工程, 2007, 2(24): 26-28.
[93] 谢 芳. CV 与 MATLAB 数学库和图形库的混合编程应用[J]. 2005, 1(24): 38-40.
[94] 新电脑课堂编委会. Visual C++编程篇[M]. 北京: 电子工业大学出版社, 2004: 133-271.
[95] 尹立民, 王兴东. VC++应用编程 150 例[M]. 北京: 电子工业出版社, 2004: 1-135.
[96] 伍 怡. 三维植物器官的几何建模[D]. 中国科学院自动化研究所硕士学位论文, 2001.
[97] 严红平. 基于子结构和水力模型的虚拟植物建模和可视化研究[D]. 中国科学院自动化研究所博士后论文, 2002.
[98] 王 矼. 虚拟植物器官库几何造型软件与相关问题研究[D]. 中国科学院自动化研究所硕士学位论文, 2003.
[99] 赵 星. 忠实于植物学的虚拟植物生长研究[D]. 中国科学技术大学博士学位论文, 2001.
[100] 赵 星. 虚拟植物生长中几个问题的研究[D]. 中国科学院自动化研究所博士后论文, 2003.
[101] 丁维龙. 虚拟植物生长模型及其与智能系统集成研究[D]. 中国科学技术大学博士学位论文, 2004.