Host: The Japan Society of Mechanical Engineers
Name : [in Japanese]
Date : June 08, 2016 - June 11, 2016
This paper describes an image processing method to detect the center position of a cabbage head by using gray scale images. The method first selects candidate areas of the center positions using image edges, which are robust to illumination changes. Next, the candidate areas are validated by a binary discriminative function. To configure the function, the HOG (histograms of oriented gradients) feature that is robust to illumination changes and local shape differences is applied. Experimental results using a car-mounted camera show the effectiveness of the proposed method; we obtained 0.864 in F-measure and 92.2% success rate on position estimation of cabbage head center.
