Plant leaf classification and retrieval using multi-scale shape descriptor

作     者：Xu, Guoqing Li, Chen 

作者机构：Nanyang Inst Technol Sch Informat Engn 80 ChangJiang Rd Nanyang Peoples R China North China Univ Technol Sch Comp Sci 5 Jinyuanzhuang Rd Beijing Peoples R China 

出 版 物：《JOURNAL OF ENGINEERING-JOE》 (J. Eng.)

年 卷 期：2021年第2021卷第8期

页      面：467-475页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 08[工学] 

基　　金：Universities Key Scientific Research Project of He'nan [19A520029] Interdisciplinary Sciences Project Nanyang Institute of Technology, research project of Beijing Municipal Education Commission [KM201810009005] Beijing Young Topnotch Talents Cultivation Program [CITTCD201904009] Beijing Talents Support Program (Backbone Talent Program) 

主　　题：Optical, image and video signal processing Image recognition Other topics in statistics Computer vision and image processing techniques Information retrieval techniques Biology and medical computing 

摘      要：Plant leaf classification is a significant and challenging research problem in computer vision area. In this study, an original multi-scale shape descriptor is presented to perform leaf classification and retrieval. Firstly, a novel iterative rule is proposed as scales generation method, which is parameter free. Secondly, leaf contour points are represented by angle information which is calculated using their neighbour points under each scale. The angle information representation is invariant to image rotation, translation and scaling. More importantly, it can describe leaf in a hierarchical way by capturing leaf features from global to local variations. Then Fast Fourier Transform operation is applied to make the representation more compact and independent from starting point. Subsequently, for leaf retrieval the dissimilarity of each pair of leaf under each scale is computed using city block metric. And support vector machine is used as classifier for leaf classification. Finally, experiments and comparisons with multiple state-of-the-art approaches are performed. The classification accuracy was 96.85% and 93.56% respectively on Swedish and Flavia leaf datasets. The mean average precision score was 66.42%, 76.69% and 44.14% respectively on Flavia, Swedish and MEW2012 leaf datasets. The results demonstrate that the proposed method has excellent performance.