This work governs that to detect diseased wheat leaf infected images. Agriculture's automatic leaf infection detection system contains the image gathering, image processing, image feature extraction, selection, and learning. This system suggestions the farmer with a fast and precise diagnosis of the plant infections. Automation of plant leaf disease identification system is an important for rushing crop diagnosis. This is research work finds that the Simple Color Histogram with Bayes Net model has highest performance. Actually, the highest accuracy is given by SCHFBN model, SCHFNB model and SCHFNBU model are producing 96.67% of accuracy. The highest positive predictive rate value is shown by by SCHFBN model, SCHFNB model and SCHFNBU model are producing 0.97 of positive predictive value. The maximum hit rate value is owned by SCHFBN model, SCHFNB model and SCHFNBU model are having 0.97 of hit rate value. The highest ROC value is given by SCHFBN model is having 1 of ROC value. The highest PRC value is given by SCHFBN model which is having 1 of PRC value. The SCHFNB model and SCHFNBU model are producing same PRC value which is 0.97 of PRC value. The highest time consumption is taking to build the GFBN model which is 0.09 seconds. The least time consumption is zero seconds to build SCHFNBU model and GFNB Updateable models. The highest F1-Score value is given by SCHFBN model, SCHFNB model and SCHFNBU model are securing is 0.97. The maximum Matthews Correlation Coefficient value is given by SCHFBN model, SCHFNB model and SCHFNBU model are holding 0.95 of MCC value. The SCHF produces lowest deviation compare with GF technique. This research work recommends that the simple color histogram filter using Bayes Net gives better result compare with other models.