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.

Rice plant detection in pre-heading term for autonomous robot navigation

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

Citation:  2010 Pittsburgh, Pennsylvania, June 20 - June 23, 2010  1009419.(doi:10.13031/2013.29903)
Authors:   Ryohei Masuda, Kazuo Nomura, Kenta Nakayama
Keywords:   Gabor filter, Autonomous robot navigation, Paddy field, Image processing

For autonomous robot navigation on levee in paddy field, it is necessary for the robot to distinguish between rice plant and weed. Our objective is to develop a method to detect rice plant in image of paddy field. We address paddy field in pre-heading term. Distinction between rice plant and weed is a challenging problem because colors of them are very similar. We focused on the fact that rice plant leaves have directional feature in the image and used Gabor filter to capture it. We used images of paddy fields taken by digital camera in this research. We found that Gabor function matched rice plant leaves in the image more strongly when it had nearly vertical direction and on the other hand, it matched weeds in rather many directions. This feature enabled us to develop a detection method based on Gabor filter. The results of experiments showed that our method could function to some extent, but there were still some misjudgment and it took long time to process. We discuss about further improvements.

(Download PDF)    (Export to EndNotes)