Categorization and Recognition of Pest in Crops and Remedial action for Smart Agriculture

Plant-related diseases and insects can seriously harm a plant’s quality and productivity. Plant diseases and pests can be identified using digital image processing. Recent advances in the research of image processing have demonstrated deep learning has significantly surpassed more traditional methods in this regard. Scientist are currently exploring that how deep learning techniques could enhance the identification of plant diseases and pests. This paper outlines the claims integrated with analysing pests and plant pathogens and evaluates and compares the various diagnostic techniques currently in use. This article summarizes three viewpoints from latest research on deep learning implemented in plant disease and pest detection, each based on a distinct network design. Recognising pest-borne diseases with a convolutional neural network (CNN) driven methodology using Dataset. In identifying pests by comparing its effectiveness with other existing methods. The CNN models, which were trained exclusively on the pest dataset, scored an impressive accuracy of $\mathbf{9 2. 6 6 \%}$.