Machine Learning in Agriculture : A Neural Network Approach

Intelligent farming manages, focuses, and provides the agricultural industry with the infrastructure to leverage advanced technology. Agriculture provides a backbone to the economy by giving employment opportunities to people in developing countries like India. The traditional methods of agriculture are being modernized through advancements in farming. Intelligent farming increases productivity by predicting climatic conditions, soil testing, and cultivating crop recommendations are some elements of smart agriculture. Using InceptionV3 and XGBoost algorithms for image recognition and agricultural yield analysis, it enables individuals to nurture crops effectively and achieve great production at very little cost. The study used the Inception v3 deep learning model to classify plant images as healthy or diseased, achieving high accuracy rates of 87.80% on the train dataset and 85.61% on validation dataset. The model was also effective in accurately classifying images of pests, with even higher accuracy rates of 97.86% on the training set and 93.30% on the validation set. Additionally, the model performed well in detecting plant diseases, achieving an accuracy rate of 94.11% and 95.28% on the validation set. XGBoost was utilized to predict crop yield, with a 100% accuracy rate on the training set and a high accuracy rate of 0.99 on the testing set. These findings indicate that deep learning models can be effective tools for accurately classifying plant images and predicting crop yield.