Various non-separable and difficult computing issues are resolved using pattern classifiers (PCs). Another of the main issues is accurately predicting a particular disease in a normal fruit tree. A grower could be able to take necessary preventative steps ahead of time with the aid of a prompt and properly apple tree illness that was anticipated. The green leaves plague and apples disease infections are predicted in this essay using a method for diagnosing apple illnesses. The system’s goal is to create a method for predicting apple disease. From different parts of Kashmir, several photographs of apple leaf disease have been gathered. In order to train the suggested human brain, these photos are used. An image can be used to extract a variety of features, including color, material, edge, and form characteristics. This study employs low grade and curve data to create an intelligent apple disease prediction system. The leaf ages are analyzed initially to obtain the key image information, including brightness, radiation, the IDM mean, standard deviation, perimeter, etc. Eleven apple leaf image characteristics and a multi-layer perceptron pattern classifier are used to train the suggested program’s architecture. The autonomous robot that performs pattern categorization is built using the gradual descending rear technique. The proposed approach has a 99.1% diagnostic average accuracy after being tested on several random samples. The proposed prediction model has a sensitivity of 98.1% and a specificity of 99.9%.