Plant leaf recognition by integrating shape and texture features

作者：

Highlights：

• We propose a novel shape description approach named multiscale triangle descriptor MTD for plant species recognition.

• The MTD feature is represented by two triangle features: signed area of a triangle and the center distance of a triangle.

• We present a plant leaf recognition scheme that effectively combines the MTD shape and LBP-HF texture features.

• The experimental results show that the proposed method outperforms the state-of-the-art plant species recognition approaches.

摘要

•We propose a novel shape description approach named multiscale triangle descriptor MTD for plant species recognition.•The MTD feature is represented by two triangle features: signed area of a triangle and the center distance of a triangle.•We present a plant leaf recognition scheme that effectively combines the MTD shape and LBP-HF texture features.•The experimental results show that the proposed method outperforms the state-of-the-art plant species recognition approaches.

论文关键词：Plant leaf recognition,Multiscale triangle descriptor,Local binary pattern,Shape descriptor,Texture feature

论文评审过程：Received 1 November 2019, Revised 12 December 2020, Accepted 18 December 2020, Available online 4 January 2021, Version of Record 17 January 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107809