Apple Fruit Disease Identification Using a Novel Gray-Scale Segmentation Algorithm and Sorting Using a Naive Bayesian Classifier

Objectives: Apple fruit classification is crucial in the post-harvest process to ensure quality control. This paper proposes a novel gray-scale segmentation algorithm for accurately identifying diseased regions in apple images. Methods: The algorithm converts color images to grayscale, normalizes the histogram, and applies a threshold based on inter-class frequency summation to segment defective areas. Extracted texture features from segmented images are fed into a Naive Bayesian classifier to distinguish between sound and diseased apples. Findings: Experiments show that the algorithm outperforms existing methods like Otsu’s and Kapur’s regarding visual accuracy and execution speed. The Naive Bayesian classifier achieves a remarkable accuracy of 95%, demonstrating the efficacy of the proposed approach. Novelty: This work presents a significant step towards enhancing the accuracy and speed of apple fruit sorting machines . Keywords: Thresholding Algorithm, Apple Fruit, Segmentation, Classification, Naive Bayesian classifier