Plant diseases are a major global danger to agriculture, affecting food security as well as production. Accurate diagnosis and early detection are essential for effective illness management. Conventional techniques are frequently laborious and subjective and rely on visual assessment. Current developments in machine learning and computer vision provide promising substitutes. This research proposes an improved system that blends segmentation techniques with preprocessing for plant disease segmentation. While the segmentation stage uses the Adaptive Pixel Integration in Joint Segmentation (APIJS) method, the initial preprocessing stage refines the data using median filtering. This method, a variation of DJS, is intended to precisely identify disease-affected areas in plant photos. By improving the precision and effectiveness of plant disease segmentation, this framework contributes to advancing sustainable agriculture and strengthening global food security.