End-to-End AIoT Based Solution to Improve Crop Yield

The fast growth of the agricultural sector requires innovative keys to address problems such as low productivity, resource inefficiency, and the need for timely decisionmaking support for farmers. Motivated by these challenges, this research introducestwo novel systems-AIoTST-CR and VotTomNet-to enhance crop recommendationand disease diagnosis, thereby contributing to sustainable and precision agriculture.The proposed AIoTST-CR system integrates soil testing hardware with AI models torecommend the most appropriate crops based on soil parameters. This approach significantly reduces input costs and fertilizer usage while improving crop yield. Performanceevaluation revealed that Bayes method and Random Forest attained the maximum levelof accuracy, making sure that crop suggestions are trustworthy. A dedicated mobile application, Soil2Crop, was also developed to provide farmers with an intuitive platformfor accessing actionable insights.In parallel, the VotTomNet framework addresses tomato leaf disease diagnosis byensembling six up-to-date pre-trained deep learning models ( InceptionNet, VGG16, DenseNet, MobileNet, ResNet, & EfficientNet).