Small Target Detection Method Based on Dilated Convolution and Bidirectional Feature Fusion

In order to improve the effect of small target detection, this paper proposes a small target detection method based on dilated convolution and bidirectional feature fusion, on the basis of YOLOv7 algorithm, the sensory field of the feature map is enlarged by introducing dilated convolution, the bidirectional feature fusion mechanism is introduced for the feature fusion between the feature map hierarchy, and the network pays more attention to the small targets through the SE attention mechanism. Experiments are carried out on the VisDrone2019 datasets, and from the experimental results, it can be seen that the mAP value of the improved network model is improved by 3.3%, and the accuracy of some small target categories in the datasets is also significantly improved, which proves the effectiveness of the proposed method in the task of small target detection.