Improved YOLOv3 model with feature map cropping for multi-scale road object detection

• Published in: Measurement science & technology 2023-04, Vol.34 (4), p.45406
• Main Authors: Shen, Lingzhi, Tao, Hongfeng, Ni, Yuanzhi, Wang, Yue, Stojanovic, Vladimir
• Format: Article
• Language: English
• ISSN: 0957-0233; 1361-6501
• DOI: 10.1088/1361-6501/acb075

Road object detection is an essential and imperative step for driving intelligent vehicles. Generally, road objects, such as vehicles and pedestrians, present the characteristic of multi-scale and uncertain distribution which puts a high demand on the detection algorithm. Therefore, this paper proposes a YOLOv3 (You Only Look Once v3)-based method aimed at enhancing the capability of cross-scale detection and focusing on the valuable area. The proposed method fills an urgent need for multi-scale detection, and its individual components will be useful in road object detection. The K-means-GIoU algorithm is designed to generate a priori boxes whose shapes are close to real boxes. This greatly reduces the complexity of training, paving the way for fast convergence. Then, a detection branch is added to detect small targets, and a feature map cropping module is introduced into the newly added detection branch to remove the areas with high probability of background targets and easy-to-detect targets, and the cropped areas of the feature map are filled with a value of 0. Further, a channel attention module and spatial attention module are added to strengthen the network’s attention to major regions. The experiment results on the KITTI dataset show that the proposed method maintains a fast detection speed and increases the mAP (mean average precision) value by as much as 2.86 % compared with YOLOv3-ultralytics, and especially improves the detection performance for small-scale objects.