该文探讨如何根据植物的叶片特征, 利用图像处理和机器学习的方法对植物进行分类。鉴于现有的叶片分类系统多采用单一的特征, 如几何和纹理等, 仅能在小规模数据库上得到较好的结果。然而, 随着样本种类的增多, 单一特征在不同种类叶片之间的相似性非常明显, 致使分类正确率降低。该研究使用多种复合特征, 并提出了原创的预处理方法以及宽度、叶缘频率特征, 较传统的几何特征更为详尽。研究结果显示, 复合特征可以有效避免算法过拟合问题, 使之适用于更大的数据库。通过提取21类植物的叶片宽度、颜色、叶缘和纹理共292维特征, 对1 915张数字图像进行了分类, 正确率达到93%, 并分析了各类特征对分类结果的影响。研究结果表明, 在不影响分类正确率前提下, 可将特征减少到约100维。
This study investigated how to classify plant species with compound leaf features using machine-learning approaches. Many traditional classification systems used a single feature, such as geometry or texture. Although such systems can achieve good results in small databases, with increasing records, the similarity in single features between different species will be remarkable, thus reducing the accuracy in large databases. This study examined how to extract compound features and proposes a novel preprocessing method and new ways to extract width and edge information, which are more detailed than with most of the state-of-the-art approaches. The compound features can reduce the influence of the over-fitting problem, so the algorithm can be used for larger databases. We examined up to 21 kinds of plants (extracting width, color, edge and texture data) and 1 915 digital images and achieved an accuracy of 93%. Finally, we analyzed the effect of each feature on classification results. We could reduce the feature’s dimension to about 100 without losing much classification accuracy.
[1]黄志开.彩色图像特征提取与植物分类研究[D].博士论文, 2006, 博士论文, 中国科学技术大学.
[2]魏蕾.基于图像处理和SVM的植物叶片分类研究[D].博士论文, 2012, 西北农林科技大学.
[3]顾振华.园林植物识别的策略与方法[J][J].现代农业科学, 2008, 15
(08):28-29
[4]邵新庆, 冯全, 邵世禄, 王敬轩, 王宇通.基于叶片图像的植物鉴别技术研究进展(综述)[J][J].甘肃农业大学学报, 2010, 45
(2):156-160
[5]Du, J.X. Wang and GZhang.Leaf shape based plant species recognition[J].Applied mathematics and computation, 2007, 185
(2):883-893
[6]Chia-Ling, L.and CShu-Yuan.Classification of leaf images[J].International Journal of Imaging Systems and Technology, 2006, 16
(1):15-23
[7]Wu, S.G., et al.A leaf recognition algorithm for plant classification using probabilistic neural network[M].Signal Processing and Information Technology, 2007 IEEE International Symposium on, 2007, 11-16
[8]Chaki, J.and RParekh.Plant leaf recognition using shape based features and neural network classifiers[J].International Journal of Advanced Computer Science and Applications, 2011, 2
(10):41-47
[9]朱静, 田兴军, 陈彬, 吕劲紫.植物叶形的计算机识别系统[J][J].植物学通报, 2005, 22
(5):599-604
[10]郑小东, 张晓煜, 薄树奎.植物叶裂特征自动提取研究[J][J].中国农学通报, 2012, 28
(27):152-156
[11]Mokhtarian, F.and SAbbasi.Matching shapes with self-intersections: Application to leaf classification[J].IEEE Transactions on Image Processing, 2004, 13
(5):653-661
[12]Fu, H.and Z. Chi.A two-stage approach for leaf vein extraction[M].Neural Networks and Signal Processing, 2003. Proceedings of the 2003 International Conference on, 2003, 1, 208-211
[13]Li, Y., et al.A leaf vein extraction method based on snakes technique[M], Neural Networks and Brain, 2005. ICNN\&B'05. International Conference on, 2005, 2, 885-888
[14]Backes, A.R.,DCasanova and O.M. Bruno,PLANT LEAF IDENTIFICATION BASED ON VOLUMETRIC FRACTAL DIMENSION[J].INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE, 2009, 23
(6):1145-1160
[15]杨辉军, 陈立伟.基于分形特征的植物识别[J][J].计算机工程与设计, 2010, 31
(24):5321-5323+5327
[16]Otsu, Nobuyuki.A threshold selection method from gray-level histograms[J].Automatica, 1975, 11
(285-296):23-27
首页