Vehicle and pedestrian detection method based on improved YOLOv4-tiny

Abstract

Aiming at the problem of low detection accuracy of vehicle and pedestrian detection models, this paper proposes an improved you only look once v4 (YOLOv4)-tiny vehicle and pedestrian target detection algorithm. Convolutional block attention module (CBAM) is introduced into cross stage partial Darknet-53 (CSPDarknet53)-tiny module to enhance feature extraction capabilities. In addition, the cross stage partial dense block layer (CSP-DBL) module is used to replace the original simple convolutional module superposition, which compensates for the high-resolution characteristic information and further improves the detection accuracy of the network. Finally, the test results on the BDD100K traffic dataset show that the mean average precision (mAP) value of the final network of the proposed method is 88.74%, and the detection speed reaches 63 frames per second (FPS), which improves the detection accuracy of the network and meets the real-time detection speed.

References

[1] GirshickR. Fast R-CNN[C], 2015, New York, IEEE: 1440-1448

[2] RenS Q, HeK M, GirshickR, et al.. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 39(6):1137-1149

[3] DaiJ F, LiY, HeK M, et al.. R-FCN object detection via region-based fully convolution networks[C], 2016, 379-387

[4] HeK M, GkioxariG, DollárP, et al.. Mask R-CNN[C], 2017, New York, IEEE: 2980-2988

[5] FangL P, HeH J, ZhouG M. Research overview of object detection methods[J]. Computer engineering and applications, 2018, 54(13):11-18(in Chinese)

[6] ShaoY H, ZhangD, ChuH Y, et al.. A review of YOLO object detection based on deep learning[J]. Journal of electronics & information technology, 2022, 44(10):3697-3708(in Chinese)

[7] LiJ Y, YangJ, KongB, et al.. Multi-scale vehicle-pedestrian detection algorithm based on attention mechanism[J]. Optics and precision engineering, 2021, 29(06):1448-1458 in Chinese)

[8] MengL X. Research on vehicle-pedestrian detection method based on deep learning[D], 2021, Taiyuan, North University of China(in Chinese)

[9] GuoZ J, LiJ Y, QiH J, et al.. Detection algorithm for infrared pedestrian and vehicle based on the improved YOLOv4[J]. Laser & infrared, 2023, 53(4):607-614(in Chinese)

[10] ZhouH P, WangJ, SunK L. Pedestrian detection algorithm based on improved YOLOv4-tiny[J]. Radio communications technology, 2021, 47(4):474-480(in Chinese)

[11] YiX, SongY H, ZhangY L. Enhanced darknet53 combine MLFPN based real-time defect detection in steel surface[J], 2020, Berlin, Heidelberg, Springer Science and Business Media Deutschland GmbH: 303-314

[12] LinT Y, DollarP, GirshickR, et al.. Feature pyramid networks for object detection[C], 2017, New York, IEEE: 936-944

[13] WooS, ParkJ, LeeJ Y, et al.. CBAM: convolutional block attention module[C], 2018, Berlin, Heidelberg, Springer Verlag: 3-19

[14] TanM M, PangRM, LeqV. EfficientDet: scalable and efficient object detection[C], 2020, New York, IEEE: 10778-10787Virtual

[15] ChienY W, HongY M L, YehI H, et al.. CSPNet: a new backbone that can enhance learning capability of CNN[C], 2020, New York, IEEE: 1571-1580Virtual