Title: Light leaf spots segmentation algorithm based on colour difference vectors

Authors: Sheng Miao; Weili Kou; Lihui Zhong

Addresses: College of Big Data and Intelligence Engineering, Southwest Forestry University, Kunming, Yunnan, 650224, China ' College of Big Data and Intelligence Engineering, Southwest Forestry University, Kunming, Yunnan, 650224, China ' College of Big Data and Intelligence Engineering, Southwest Forestry University, Kunming, Yunnan, 650224, China

Abstract: For plant diseases monitoring and diagnosing, leaf spots image segmentation and quantification is very crucial. Compared with dark spots, light leaf spots segmentation is hindered by the unobvious colour and similar leaf veins. This study specially focuses on the light leaf spots segmentation and vein removals. Firstly, vein and spot masks were respectively extracted from healthy and spots leaves by using a mask extraction model presented in this study. Secondly, vein ratio of vein masks was calculated to evaluate whether vein removal is needed or not. Thirdly, vein area and roundness were calculated for both vein and spots masks, and areas with high similarity to veins were removed from spots mask. The proposed algorithm was tested using leaves with light spots of 12 plants types. The results show that the presented algorithm has higher average accuracy (86%) than both traditional watershed (77%), Barbedo (2017) (82%) and CNN (81%).

Keywords: light leaf spots; image segmentation; vein removal; colour difference vector.

DOI: 10.1504/IJICA.2023.129354

International Journal of Innovative Computing and Applications, 2023 Vol.14 No.1/2, pp.12 - 21

Received: 15 Sep 2020
Accepted: 16 Mar 2021
Published online: 07 Mar 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article