An Algorithm to Extract the Defective Areas of Potato Tubers Infected with Black Scab Disease Using Fuzzy C Means Clustering for Automatic Grading

Abstract

Estimating the surface area of defects of diseased potatoes is a key factor in the automatic grading of this product. In this article, an algorithm has been developed using fuzzy clustering method and image processing functions to estimate the defective areas of potato tubers infected with black scab disease. Fuzzy clustering, which is an unsupervised method, was used to segment color images and extract defective areas of potatoes, and image processing functions have been used to extract the total area of potatoes. In the segmentation method based on fuzzy clustering, the data matrix related to potato images were divided into separate clusters in a fuzzy way, in which the boundaries of the clusters are defined in a fuzzy way instead of being definite and specific. The results showed that this algorithm is very efficient for extracting black scab disease and can be used to extract the amount of diseases that can be used for automatic grading of this product based on the American standards.

Keywords

• black scab disease
• fcm clustering algorithm
• potato disease
• image processing