Organic acids, amino acids, proteins and carbohydrates are enclosed in coffee leaves. In various applications like coffee leaf tea, therapeutic agents, tobacco substitutes, packaging material, personal hygienic products, ethno medicine and animal feed, coffee leaves were wielded. For preventing disease in coffee plants, research has been done for high potential benefits. Generally, for coffee leaf disease classification, the utilized Deep Learning (DL) models were effectively applied. But, to predict a tiny portion of the disease in the overall leaf area, the prevailing techniques were not effective. Thus, this work proposes a novel technique of Weibull distributed-Golden Flower Pollination Optimization (W-GFPO) YoloV3 and Jaccard Index Fuzzy C-means Clustering (J-FCM)-centric Multi-disease Prediction in Coffee Leaf (WJMPCL). Primarily, to enhance the edge, the images are pre-processed. Then, the image is given to the color spacing model, which enhanced the image and is further given to feature extraction. For training purposes, the features are given to W-GFPO YoloV3 and segmentation is done with labeling. After that, for clustering the images into healthy leaf clusters, coffee leaf miner, phoma tarda, coffee leaf rust, and Iron spot, the labeled image is given to J-FCM. Utilizing coffee leaf images, the proposed mechanism was assessed, which attained superior performance in all experiments when analogized with the prevailing approaches.

coffee leaf miner and Iron spot, Jaccard Index Fuzzy C-means Clustering (J-FCM), phoma tarda, Weibull distributed- Golden Flower Pollination Optimization (W-GFPO),