Improved Lightweight Target Detection Algorithm for Complex Roads with YOLOv5

Aiming at the problems of false detection and missed detection caused by dense occluded targets and small targets in complex road background, and the existing target detection algorithms are not strong enough, a complex road lightweight target detection algorithm based on improved YOLOv5 is proposed. Firstly, Quality Focal Loss is introduced, and the classification score is combined with the quality prediction of location to improve the positioning accuracy of dense occluded targets; Secondly, a shallow detection layer is added as the detection layer of smaller targets, the three-scale detection of the original algorithm is changed to four scales, and the feature fusion part is also changed accordingly, which improves the learning ability of the algorithm to the features of small targets; Then, referring to the lightweight idea of GhostNet, the Bottleneck of the Neck part of the original algorithm is replaced by GhostBottleneck, which reduces the volume of the algorithm model and the amount of parameters and calculations on the premise of maintaining the detection accuracy; Finally, the Backbone part of the original algorithm is replaced by Mobilenetv3, which further lightens the algorithm and improves the detection speed of the algorithm in the CPU environment. The experimental results show that on the public autopilot data set KITTI and the self-made rider helmet data set Helmet, compared with the original YOLOv5s algorithm, the volume of the improved algorithm in this paper is reduced by 60%, the amount of parameters is reduced by 62%, and the amount of calculation is reduced by 63%. The detection speed in the CPU environment is increased by 6FPS. Compared with some mainstream target detection algorithms, the improved algorithm in this paper has certain advantages.