Improving Durian Leaf Disease Detection Using LAB Color Space and CLAHE Technique with YOLOv8 Integration

Durian is considered a significant economic crop in Thailand, with an economic value of 1 1 0 billion baht. Leaf diseases in durian trees greatly impact the health of the trees and the quality of the durian fruit, especially its aroma and taste. Effective management of leaf diseases ensures that durian trees grow healthily and produce high-quality fruits. One way to diagnose and detect illnesses of durian leaves, such as algal leaf spot, leaf spot, and leaf blight, is through laboratory testing, can be done through laboratory testing, which is quite costly. Visual inspection requires regular monitoring, which is time-consuming and labor-intensive for industrial-scale cultivation. Therefore, this issue has gained attention by using image processing methods to identify diseases in durian leaves that presents an enhanced approach by improving the preprocessing steps using image sharpening, contrast enhancement, and conversion to LAB color space, followed by applying Contrast Limited Adaptive Histogram Equalization (CLAHE) to improve feature visibility and utilizing the YOLOv8 model for disease detection and classification. The experimental results found that the YOLOv8 model achieved an accuracy of 99.74% in detecting durian leaf diseases.