American Society of Agricultural and Biological Engineers



Click on “Download PDF” for the PDF version or on the title for the HTML version.


If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Intra-row weed recognition using plant spacing information in stereo images

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org

Citation:  Paper number  131592292,  2013 Kansas City, Missouri, July 21 - July 24, 2013. (doi: http://dx.doi.org/10.13031/aim.20131592292) @2013
Authors:   Yong Chen, Xiaojun Jin, Lie Tang, Jun Che, Yanxia Sun, Jun Chen
Keywords:   Stereo vision intra-row weed recognition height feature plant spacing information.

Abstract. Crop/weed recognition is a crucial step for selective herbicide application. A machine vision based sensing system was developed to detect intra-row weeds when crops were at their early growth stages. The proposed methods used color feature to extract vegetation from the background, whilst height and plant spacing information analysis techniques were applied to discriminate between crops and weeds. Firstly the identification of the weeds that were lower than crops was done by a height-based segmentation method using a stereo vision system. During the stereo matching process, correspondence search was performed on edged stereo images and disparity calculation was applied only to the edge pixels. This strategy could largely reduce the correspondence search range, thereby enhanced the weed recognition speed and accuracy. Afterwards, the higher weeds were distinguished from the crops by utilizing plant spacing characters. The histogram of plant pixels and their peak position were calculated from each pixel row of the segmented disparity image. Then plant centers were located and each weed region was further extracted based on the interplant distance in a row.

(Download PDF)    (Export to EndNotes)