Optimal Thresholding for the Automatic Recognition of Apple Fruits

Abstract

The aging and the decreasing number of farm workers in Japan have been apotential problem. That is why research on the automation of agricultural operations isconducted in recent years. One of these operations is the harvesting of fruit trees such as theapples. A robotic hand that could harvest the apple fruit similar to the human picker has beendeveloped, however the visual guidance of the developed hand has not yet been made. In thispaper, a machine vision system that would guide the robotic harvesting hand was studied. Themachine vision system consisted of a digital video camera and a personal computer.Images of a Fuji apple tree were analyzed and histograms of its luminance and color differenceof red were developed. The threshold for segmentation of the images to recognize the fruitportion was estimated from the histograms using the optimal thresholding method. Theestimated threshold effectively recognized the fruit portion. The threshold calculated from theluminance histogram using the optimal thresholding method was not effective in recognizing theFuji apple while the threshold selected from the color difference of red histogram was effectivein recognizing the Fuji apple.