Pest Detection and Eco-Friendly Pesticide Sorting Using Deep Learning

Effective pest management techniques are becoming important as global food demand rises and agriculture expands. This research work proposes a deep learning approach achieving a promising 94% accuracy in identifying and classifying nine distinct crop pest species. These include aphids, armyworms, beetles, bollworms, mites, mosquitoes, grasshoppers, sawflies, and stem borers. The system utilizes Convolutional Neural Networks (CNNs) to analyze crop images and detect potential infestation. An accuracy of 94% is achieved in identifying nine distinct pest species. This model utilizes the ReLU function to introduce non-linearity in the neural networks and the SoftMax activation function to generate a probability distribution over several eco-friendly pesticide alternatives. This pest identification capability, coupled with the model’s ability to recommend suitable eco-friendly pesticides, empowers farmers with a precise and sustainable solution. The recommendations consider both the identified pest and its infestation level, promoting targeted control measures that minimize environmental impact. This pest identification capability, coupled with the model’s ability to recommend suitable eco-friendly pesticides, empowers farmers with a precise and sustainable solution. The recommendations consider both the identified pest and its infestation level, promoting targeted control measures that minimize environmental impact.