Potato disease detection and control measure recommendation using CNN

Potato cultivation will continue to be a crucial agricultural endeavor, significantly enrich to global food supply. and economic prosperity. However, the ongoing vulnerability of potatoes to diseases, particularly named early Blight and late Blight, will persist as a substantial threat to crop yield and quality. Traditional disease identification methods, reliant on expert visual inspection, are expected to remain time-consuming, error-prone, and inadequate for timely intervention. This paper will outline to find Potato diseases and recommend effective control measures using convolutional Neural-Networks (CNN). The project will adopt an object-oriented methodology to ensure modularity and maintainability, leveraging TensorFlow for comprehensive dataset collection, data cleaning, and model training. TensorFlow Lite will be applied to optimize and quantify the developed model, enhancing the efficiency of the deployment process. The frontend will be crafted with React Native to facilitate user accessibility and seamless interaction. The resulting deep learning technique is anticipated to exhibit high accuracy in identifying early-blight and late-blight in potato plants, enabling users to receive prompt and reliable disease predictions and actionable control suggestions. This innovative approach aims to downgrade crop losses, improve productivity, and foster sustainable agricultural practices, thereby bolstering global food security in the future.