1990 Volume 52 Issue 1 Pages 61-67
Fruits detecting methods based on digital color image processing algorithms using a personal computer (PC) were considered in order to develop a visual sensor of the agricultural robots for harvesting mandarin oranges.
Two different algorithms, one was the thresholding of RGB signals using the Discriminant Threshold Selection Method (DTSM) and the other was the color detection using chrominance signals, were selected to make good use of color TV signals. Both algorithms implemented in PC were applied to the real images that have been acquired in an outdoor orchard under sunlight. The experimental results showed that both algorithms performed well in the presence of interference caused by bright reflections and by shadows cast on the oranges.