HNNL: Experimental Analysis of Wheat Crop Disease Detection Using Hyper Neural Network Based Learning Methodology

Many new ways of solving certain serious problems have emerged in the agricultural sector as a result of the fast growth of technology. Our method automates the identification of several wheat leaf diseases with amazing efficiency using the Python programming language and architecture based on neural networks. Disease identification in wheat harvests is an old subject that has been studied using many approaches. This study demonstrates that a new model called Hyper Neural Network Learning (HNNL) can identify and categorize problems with wheat harvests. This is when the suggested method is tested against an existing learning model known as a Convolutional Neural Network (CNN) to see how well it performs. The suggested model is utilized by the programme to recognize important features of the photos; it makes use of models that has been trained before. The proposed approach may then use the main parameters mentioned earlier to distinguish between healthy wheat crops and wheat crops damaged by disease. Results showed a dependability of 96.46 percent using a total of 1500 images in this investigation. There was one picture class showing healthy crops and eleven classes showing sick crops in these photographs. We used varied camera angles to flip the photographs in the collection, which helped the suggested model identify and classify illnesses more effectively. Our preliminary research suggests that a combination of HNNL and CNN may substantially enhance the precision of wheat crop diagnosis of diseases.