Identifying the Fertilization of Crops by Detecting Its Disease Using Image Processing

Agriculture is main reason for the development of our country. Technology has proven its fullest in this area too. But also manyfarmers are still unaware of what disease does the required crop is affected of. Identifying the crop disease requires large amount of time and needed much knowledge in it. The proposed method focuses on providing the name of the crop disease and what fertilizers can be used for it. At the initial phase the input image is pre-processed. Features like Grey level Co-occurrence matrix (GLCM), are withdrawn from the pre-processed image. Then the fertilizer decision making is done using Layered Recurrent Neural Network (LRNN). This whole process is simulated using MATLAB 2016b. The farmer can simply upload the picture of the infected crop and can see the information about that particular crop. The result displays what is the name of the disease that has been affected to the crop and what fertilizers can be used to avoid the disease. It also displays the required amount of fertilizers that must be given to the crop. The performance is noted that the detection rate and data protection is higher than existing SVM classifier.