Plant identification using leaf shapes—A pattern counting approach
作者：
Highlights：
• Doing statistics over learned shape patterns, rather than matching contour points.
• The proposed method can classify both simple leaves and compound leaves.
• The learned shape patterns measure leaf shapes from different aspects.
• We do NOT require any pre-processing of the leaf contours, i.e. petioles removal.
• Experiments on public datasets show the effectiveness of the proposed method.
摘要
•Doing statistics over learned shape patterns, rather than matching contour points.•The proposed method can classify both simple leaves and compound leaves.•The learned shape patterns measure leaf shapes from different aspects.•We do NOT require any pre-processing of the leaf contours, i.e. petioles removal.•Experiments on public datasets show the effectiveness of the proposed method.
论文关键词：Plant identification,Foliage image recognition,Shape matching,Sparse representation,Dictionary learning
论文评审过程：Received 4 October 2014, Revised 21 March 2015, Accepted 2 April 2015, Available online 16 April 2015, Version of Record 17 June 2015.
论文官网地址：https://doi.org/10.1016/j.patcog.2015.04.004
