Improved accuracy of pest detection using augmentation approach with Faster R-CNN

Abstract In agriculture, Pests are decreasing agricultural productivity. Identifying a pest is challenging process and subject to expert opinion. Nowadays, lots of work carried out for automatic detection. It becomes possible because emerging Deep Learning’s object detection architectures. This paper shows the multi-class using Faster R-CNN architecture compared performance results image augmentation with focused on accuracy along small dataset. We have used Horizontal Flip 90 Degree Rotation parameters solving class imbalance problem. found that trained model options can perform better an 91.02% architecture.