On-Tree Detection and Counting of Apple Using Color Thresholding and CHT

Abstract

One of the challenges in continual fruit cultivation is to approximately calculate the number of fruits on a tree. This helps to get yield estimation for different farm operation. On-tree detection and counting of apple fruits in a correct manner is a challenging task in computer vision. This work implements circular Hough transform (CHT) and colour thresholding in an iterative manner in image processing method to detect and count each apple fruit. The fruit identified by CHT which were clearly visible are preferred for early harvesting than the partially visible fruits. The methodology proposed in this paper has the capability to detect and count fully visible as well as partially visible apple fruit. The precedence of Circular Hough Transform is to detect and count overlapping fruit. The count evaluated by the proposed algorithm results in 97.70% of accuracy.