Fast Generalized Chan-Vese Model for plant/soil segmentation to estimate percentage of ground cover in agricultural images

Y. Boutiche, A. Abdessalem, N. Ramou, N. Chetih

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

Computer vision systems have been widely used in agriculture to perform tasks requiring segmentation and classification. In this paper, we describe a fast multi-channel implicit active contour method for performing plant/soil segmentation in color agriculture images. First, the Chan-Vese model is generalized to segment images in L∗ a∗ b∗ color space, then a level set method, capable of dealing with topology changes, is used to segment the whole image. Finally, a functional is optimized via a 'sweeping' algorithm for fast convergence. Percentage of Ground Cover (PGC) is then easily estimated from the segmented image. Several experiments have been conducted to validate the proposed algorithm.

Original languageEnglish
Title of host publication2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728153414
DOIs
Publication statusPublished - Dec 2019
Externally publishedYes
Event19th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2019 - Ajman, United Arab Emirates
Duration: Dec 10 2019Dec 12 2019

Publication series

Name2019 IEEE 19th International Symposium on Signal Processing and Information Technology, ISSPIT 2019

Conference

Conference19th IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2019
Country/TerritoryUnited Arab Emirates
CityAjman
Period12/10/1912/12/19

Keywords

  • Agricultural images
  • color spaces
  • Percentage of Ground Cover
  • segmentation

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing
  • Information Systems and Management

Cite this