Target Recognition Technology Based on Improved Faster RCNN
Hai-rong Hu, Jin-xin Duan, Jian Quan Bi, Lu-lu Fang · 2022 3rd International Conference on Electronic Communication and Artificial Intelligence (IWECAI) · 2022
Based on the Faster RCNN detector, this article first analyzes the advantages and disadvantages of using a variety of mainstream convolutional neural networks in target recognition. An improved model with “difficult samples” augmentation features and lighter network model is proposed. And through experiments, it is verified that this model has certain advantages in comprehensive evaluation indicators such as accuracy, recall, and detection speed compared with other models.