Leaf margin is one of the main characteristics to identify plant species. Compared to leaf shape features, leaf margin features are much more subtle, so they are often indispensable in multiscale recognition of plant species as either dependent features or supplements for others. The progresses include designing 7 new margin feature descriptors, taking hierarchical classification organized by some semantic dictionaries to reach a better classification accuracy, and finally deciding plant species of a leaf node member by similarity evaluation and retrieval. Our experiments have revealed that the descriptors, named as the ratio of residual convex to leaf area and the ratio of right edge to left edge, are efficient to distinguish between different nonlobedleaf species and different nonintegrifoliousleaf species; the mean value of residual convex etc., is of other examples of useful descriptors to the identification between different nonintegrifoliousleaf species. By using the hierarchical classification in the feature space of multi leaf margin descriptors, 30 nonlobedleaf species have been divided into several leaf nodes, and the mean overall accuracy is better than 81.21%. The test of assessing the similarity between the new assigned leaf node member and the known samples has further demonstrated that the framework of jointly using the hierarchical classification and the image retrieval is effective for the identification of plant species.
YAN Yi-zhen,ZHOU Jian-hua
. Plant image classification and retrieval based on leaf margin features[J]. Journal of East China Normal University(Natural Science), 2015
, 2015(4)
: 154
-163
.
DOI: 10.3969/j.issn.1000-5641.2015.04.016
［1］NAM Y, HWANG E, KIM D. A similaritybased leaf image retrieval scheme: Joining shape and venation features［J］. Computer Vision And Image Understanding, 2008, 110: 245259.
［2］MACLEOD N, BENFILELD M, CULVERHOUSE P. Time to automate identification［J］. Nature, 2010, 467: 154155.
［3］JAMES S C, DAVID C, JONATHAN Y C, et al. Plant species identification using digital morphometrics: A review［J］. Expert Systems with Applications, 2012, 39: 75627573.
［4］祁亨年,寿韬,金水虎,等.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):281284.
［5］朱静,田兴军,陈彬,等.植物叶形的计算机识别系统［J］.植物学通报,2005,22(5):599604.
［6］赵国庆,刘循,王勇,等.导数在提取植物叶片锯齿特征上的应用［J］.四川大学学报:自然科学版,2009,46(4):941946.
［7］徐辉,王忠芝,黄心渊,等.基于角点检测的叶缘锯齿快速识别［J］.北京林业大学学报,2010,32(6):8589.
［8］郑小东,王晓洁,高洁,等.SUSAN算法在植物叶缘特征提取中的应用［J］.中国农学通报,2011,27(27):174178.
［9］王晓洁,于浩杰,郑小东,等.凸包在植物叶锯齿与叶裂位置识别中的应用［J］.农机化研究,2013(3):214217.
［10］CLARK J Y. Identification of botanical specimens using artificial neural networks［C］//Computational Intelligence in Bioinformatics and Computational Biology, 2004. Proceedings of the 2004 IEEE Symposium on, 2004: 8794.
［11］CLARK J Y. Plant identification from characters and measurements using artificial neural networks［M］//MACLEODN.Automated Taxon Identification in Systematics: Theory, Approaches and Applications. FL:CRC Press, 2007.
［12］CLARK J Y. Neural networks and cluster analysis for unsupervised classification of cultivated species of Tilia (Malvaceae)［J］. Botanical Journal of the Linnean Society, 2009, 159: 300314.
［13］RUMPUNEN K, BARTISH I V. Comparison of differentiation estimates based on morphometric and molecular data, exemplified by various leaf shape descriptors and RAPDs in the genus Chaenomeles［J］. Taxon, 2002, 51: 6982.
［14］OTSU N. A threshold selection method from graylevel histogram［J］. Automatica, 1975, 11(285296): 2327.
［15］贺鹏,黄林.植物叶片特征提取及识别［J］.农机化研究,2008(6):168170.
［16］周坚华.遥感图像分析与空间数据挖掘［M］.上海:上海科技教育出版社,2010:109.
［17］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):190193.
中国综合性科技类核心期刊(北大核心)