Weed detection in sugar beet fields using machine vision. | [International Journal of Agriculture and Biology • 2006]

Author(s):

1. Abdolabbas Jafari: Department of Bio-System Engineering, Tehran University, Karaj, Iran

2. Seyed Saeid Mohtasebi: Department of Bio-System Engineering, Tehran University, Karaj, Iran

3. Mahmoud Omid: Department of Bio-System Engineering, Tehran University, Karaj, Iran

4. Hasan Eghbali Jahromi: Department of Computer Engineering, Shiraz University, Shiraz, Iran

Abstract:

Machine vision is a useful method for segmentation of different objects in agricultural applications, especially shape recognition methods, but it has more difficulty for weed detection due to leaves occlusion and their overlaps. Many indices have been investigated by researchers to perform weed segmentation based on color information of the images. In this study the relation between three main components (red, green & blue) of the images, which constitute the true color of different plants have been extracted from image data using discriminant analysis. 300 digital images of sugar beet plants and seven types of common sugar beet weeds at different normal lighting conditions were used to provide enough information to feed the discriminant analysis procedure. Discriminant functions and their success rate in weed detection and segmentation of different plant species have been evaluated.

Page(s): 602-605

DOI: DOI not available

Published: Journal: International Journal of Agriculture and Biology, Volume: 8, Issue: 5, Year: 2006

Keywords:

Keywords are not available for this article.

References:

References are not available for this document.

Citations

Citations are not available for this document.

Citations

Downloads

Views