An Effective Stage of Preprocessing Design for Identifying Mango Leaf Disease via Hybrid Filtration

Mango, a vital fruit crop cultivated across various regions globally, faces health and productivity challenges due to several diseases. Among the prevalent diseases are black sooty mold, bacterial canker, and anthracnose, which are distinguished by leaf lesions. Employing methods for processing images offers a means to identify and categorize anthracnose signs on mango leaves. This study introduces a procedure to enhance Mango leaf illustration quality for more effective processing. The approach encompasses preparing pictures for processing by reducing noise and sharpening, followed by segmentation using the thresholding of Otsu. The suggested approach underwent testing on a dataset of mango leaf images, exhibiting improved quality metrics such as PSNR and MSE. The findings underscore the possibility of employing these methods of processing for early disease identification and control of mango trees, ultimately enhancing the yield and quality of crops.