Paddy Crop Pest Identification and Classification Techniques Using Deep Learning

The presence of pests in crops has a significant negative influence on both the quality and the quantity. As a result, taking steps to control and prevent bug infestations is more crucial than ever. Early detection of pests can increase agricultural production, which is of utmost significance to farmers. Finding pests manually may take time and effort for farmers, particularly those who operate bigger farms. Therefore, it is crucial to develop pest identification system based on computer vision in order to reduce the damage due to pests through proper classification of the pests and targeted management strategies. This may be accomplished by targeting management techniques towards specific pests. Deep convolutional neural networks (CNNs) are capable of extracting features and learn complicated, high-level qualities for images categorization, they are being used to address this problem. This study aims to provide a comparative analysis of several deep-learning methodologies. The Inception V3 model has outstanding performance, with an accuracy of 93.64%. It is clear that resources for determining pests impacting the yield of paddy and assessing their effectiveness are accessible to everyone.