Title: Image-based automatic segmentation of leaf using clustering algorithm

Authors: Shivalika Sharma; Chinmay Chakraborty; Davinder Paul Singh; Shubham Mahajan; Amit Kant Pandit

Addresses: Department Computer Science Engineering, Yogananda College of Engineering and Technology, Jammu, J&K, 181205, India ' Electronics and Communication Engineering, Birla Institute of Technology, Mesra, Jharkhand, 835215, India ' Department of Computer Science and IT, Jammu University, Jammu, J&K, 180006, India ' School of Electronics & Communication, Shri Mata Vaishno Devi University, Katra, J&K, 182320, India; School of Engineering, Ajeenkya DY Patil University, DY Patil Knowledge City Rd., Charholi Budruk, Lohegaon, Pune, Maharashtra – 412105, India ' School of Electronics & Communication, Shri Mata Vaishno Devi University, Katra, J&K, 182320, India

Abstract: This paper proposes a productive strategy to separate the leaf area from different plant images. The image analysis in plants plays a critical job in reasonable and gainful farming. Plant image analysis is mainly used to record the plant development plant area and leaf zone etc. Plant leaf segmentation has become a significant task to be examined among these plant characters. For this reason, a strategy for leaf area segmentation from different plant images is made. In this paper we have proposed an algorithm which will segment plant on the basis of L*a*b colour spaces and cluster formation. When our proposed technique is compared to rosette tracker it was found that our proposed algorithm was more efficient in plant leaf segmentation. The results of our proposed algorithm are, accuracy is 79.23%, sensitivity is 85.84%, and specificity is 97.40%.

Keywords: colour space; leaf segmentation; plant development; plant image analysis; classification feature extraction; phenotypes traits.

DOI: 10.1504/IJNT.2022.128939

International Journal of Nanotechnology, 2022 Vol.19 No.6/7/8/9/10/11, pp.539 - 553

Received: 08 Jan 2021
Accepted: 03 Mar 2021
Published online: 13 Feb 2023 *

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