Apple defect detection using statistical histogram based EM algorithm

Segmentation of an image into its components plays an important role in most of the image processing applications. In this article an important application of image processing in determination of apple quality is studied, and an automatic algorithm is proposed in order to determine apples skin color defects. First, this image is converted from RGB to color space L* a* b*. Then fruit shape is extracted by A CM algorithm. Finally, the image has segmented using SHEM a lgor ithm. Ex per im ental r esults on the a cquir ed im ages show that both EM and SHEM spend the same iterations to accomplish the segmentation process and get the same results. However, the proposed SHEM algorithm consumes less time than the standard EM algorithm. Accuracy of the proposed algorithm on the acquired images is 91.72% and 94.86% for healthy pixels and defected ones, respectively.