Analysis of CNN Models using Augmented Data in Tomato Diseases

A vital source of nourishment for millions of people, tomatoes are one of the most significant vegetables in the world. However, many diseases may substantially decrease the yield and quality of tomato plants. Early blight, Late blight and Powdery mildew are the three most destructive diseases in tomatoes that have the capacity to drastically lower tomato crop yields. A trusted and efficient automated approach for identifying diseases needs to be created in order to address this problem. To recognize the three primary disorders that impact tomato plants based on leaf images, we created a model using a convolution neural network (CNN) and data augmentation. We also compared the performance of different CNN models, and assessed its efficacy. With an average accuracy rate of −61.75% and average loss of 6.75%, theResNet50 model is a valuable tool that tomato growers can use to keep the diseases mentioned above from affecting their tomatoes during the harvesting process.