Rice illness acknowledgment is vital in robotized rice illness conclusion frameworks. At display, profound convolutional neural arrange (CNN) is for most often regarded as the cutting-edge setup for image recognition. In this research, we suggest a brand-new CNN-based rice impact acknowledgment technique. For the reason of creating and checking the CNN show, the database consists of 2902 negative tests and 2906 positive tests. In addition, as part of our assessment We conduct testing for both qualitative and quantitative comparisons of the suggested plan's appropriateness. The results of the investigation suggest that high-level highlights mined from CNN are more accurate and beneficial than previously hand-crafted highlights such closest double designs histograms (LBPH) and Haar-WT (Wavelet Change). Additionally, the numerical evaluation's findings show that, compared to both LBPH and Softmax using CNN, Support Vector Machine (SVM) with CNN performs similarly, with higher precision, a larger area beneath curves (AUC), and better recipient working properties (ROC) curves. and Haar-WT, which both also use SVMs as the classifier. This might perhaps make our CNN programs the greatest method for detecting rice impact illness and might be applied in practical situations.