基于改进YOLO v5算法的道路小目标检测

doi:10.12305/j.issn.1001-506X.2024.10.04

Abstract

Abstract:

In order to solve the problems that small targets on traffic roads faces including detection difficulty, low precision, detection failures, a multi-scale feature fusion target detection improvement algorithm based on the YOLO v5 (you only look once v5) algorithm is proposed. Firstly, the small target detection head is added for adapting to the small target size and alleviating the missed detection. Then, deformable convolutional networks V2 (DCN V2) is introduced to improve the model's learning ability for small targets in motion. The context augmentation module (CAM) is introduced to improve the recognition ability of small targets at a long distance. The replacement loss function is used to improve the bounding box's localization accuracy, and the spatial pyramid pooling and context spatial pyramid convolution_group (SPPCSPC_group) module is also used to improve the sensory field and feature expression ability of the network. The experiment results show that the proposed algorithm achieves an average accuracy of 95.2% in the category of small targets in the KITTI dataset, compared with the original YOLO v5 algorithm, the overall average accuracy is improved by 2.7%. For the detection of small targets, the average accuracy is improved by 3.1% with a better detection effect, which proves the effectiveness of the proposed algorithm for the detection of small targets on roads.

Key words: you only look once v5 (YOLO v5), small target detection, context augmentation module (CAM), deformable convolutional networks (DCN)

CLC Number:

TP311.5

Cunli SONG, Weiqin CHAI, Xuesong ZHANG. Road small target detection based on improved YOLO v5 algorithm[J]. Systems Engineering and Electronics, 2024, 46(10): 3271-3278.

Figures/Tables 10

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Table 1

Table 2

Fig.6

Fig.7

Table 3

References 31

1	GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2014: 580-587.
2	GIRSHICK R. Fast R-CNN[C]//Proc. of the IEEE International Conference on Computer Vision, 2015: 1440-1448.
3	REN S Q , HE K M , GIRSHICK R , et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Trans. on Pattern Analysis and Machine Intelligence, 2016, 39 (6): 1137- 1149.
4	HE K M, GKIOXARI G, DOLLAR P, et al. Mask R-CNN[C]//Proc. of the IEEE International Conference on Computer Vision, 2017: 2961-2969.
5	LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]//Proc. of the 14th European Conference of Computer Vision-ECCV, 2016: 21-37.
6	REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 779-788.
7	REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 7263-7271.
8	GE Z, LIU S T, WANG F, et al. YOLOX: exceeding YOLO series in 2021[EB/OL]. [2023-06-03]. https://arxiv.org/abs/2107.08430.html.
9	LI C Y, LI L L, JIANG H L, et al. YOLOv6: a single-stage object detection framework for industrial applications[EB/OL]. [2023-06-03]. https://arxiv.org/abs/2209.02976.html.
10	WANG C Y, BOCHKOVSKIY A, LIAOH Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023: 7464-7475.
11	皮骏, 刘宇恒, 李久昊. 基于YOLO v5s的轻量化森林火灾检测算法研究[J]. 图学学报, 2023, 44 (1): 26- 32.
	PI J , LIU Y H , LI J H . Research on lightweight forest fire detection algorithm based on YOLO v5s[J]. Journal of Graphics, 2023, 44 (1): 26- 32.
12	YAN F X, XU Y X. Improved target detection algorithm based on YOLO[C]//Proc. of the 4th International Conference on Robotics, Control and Automation Engineering, 2021: 21-25.
13	牛为华, 殷苗苗. 基于改进YOLO v5的道路小目标检测算法[J]. 传感技术学报, 2023, 36 (1): 36- 44.
	NIU W H , YIN M M . Road small target detection algorithm based on improved YOLO v5[J]. Chinese Journal of Sensors and Actuators, 2023, 36 (1): 36- 44.
14	XU X K , FENG Z J , CAO C Q , et al. An improved swin transformer-based model for remote sensing object detection and instance segmentation[J]. Remote Sensing, 2021, 13 (23): 4779. doi: 10.3390/rs13234779
15	LIU Z, LIN Y T, CAO Y, et al. Swin transformer: hierarchical vision transformer using shifted windows[C]//Proc. of the IEEE/CVF International Conference on Computer Vision, 2021: 9992-10002.
16	LI A , SUN S J , ZHANG Z Y , et al. A multi scale traffic object detection algorithm for road scenes based on improved YOLOv5[J]. Electronics, 2023, 12 (4): 878. doi: 10.3390/electronics12040878
17	WANG J Q, CHEN K, XU R, et al. CARAFE: content-aware reassembly of features[C]//Proc. of the IEEE/CVF International Conference on Computer Vision, 2019: 3007-3016.
18	SHEN L Z , TAO H F , NI Y Z , et al. Improved YOLOv3 model with feature map cropping for multi-scale road object detection[J]. Measurement Science and Technology, 2023, 34 (4): 045406. doi: 10.1088/1361-6501/acb075
19	LIU J , CAI Q Q , ZOU F M , et al. BiGA-YOLO: a lightweight object detection network based on YOLOv5 for autonomous driving[J]. Electronics, 2023, 12 (12): 2745. doi: 10.3390/electronics12122745
20	TAN M X, PANG R M, LE Q V. EfficientDet: scalable and efficient object detection[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 10781-10790.
21	ZHANG Y , GUO Z Y , WU J Q , et al. Real-time vehicle detection based on improved YOLO v5[J]. Sustainability, 2022, 14 (19): 12274. doi: 10.3390/su141912274
22	赵睿, 刘辉, 刘沛霖, 等. 基于改进YOLOv5s的安全帽检测算法[J]. 北京航空航天大学学报, 2023, 49 (8): 2050- 2061.
	ZHAO R , LIU H , LIU P L , et al. Safety hat detection algorithm based on improved YOLOv5s[J]. Journal of Beijing University of Aeronautics and Astronautics, 2023, 49 (8): 2050- 2061.
23	LIANG Q K , XIANG S , HU Y C , et al. PD2SE-Net: computer-assisted plant disease diagnosis and severity estimation net-work[J]. Computers and Electronics in Agriculture, 2019, 157, 518- 529. doi: 10.1016/j.compag.2019.01.034
24	ZHU X Z, HU H, LIN S, et al. Deformable convnets v2: more deformable, better results[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 9308-9316.
25	LUO Z X, SHEN T W, ZHOU L, et al. ContextDesc: local descriptor augmentation with cross-modality context[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 2527-2536.
26	GEIGER A , LENZ P , STILLER C , et al. Vision meets robotics: the KITTI dataset[J]. The International Journal of Robotics Research, 2013, 32 (11): 1231- 1237. doi: 10.1177/0278364913491297
27	ZHANG S F, CHI C, YAO Y Q, et al. Bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 9759-9768.
28	LIN T Y, DOLLAR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 2117-2125.
29	LIU S, QI L, QIN H F, et al. Path aggregation network for instance segmentation[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 8759-8768.
30	RAO Y M , ZHAO W L , TANG Y S , et al. HorNet: efficient high-order spatial interactions with recursive gated convolutions[J]. Advances in Neural Information Processing Systems, 2022, 35, 10353- 10366.
31	LIU Z, MAO H Z, WU C Y, et al. A ConvNet for the 2020s[C]//Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022: 11976-11986.

EIoU	CAM	SPPCSPC_group	4个检测头	C3-DCN	AP/%			mAP/%
EIoU	CAM	SPPCSPC_group	4个检测头	C3-DCN	车辆⁽¹⁾	骑行者⁽²⁾	行人⁽³⁾	mAP/%
×	×	×	×	×	97.5	93.8	87.2	92.5
√	×	×	×	×	97.8	94.4	88.3	93.5
√	√	×	×	×	97.9	94.8	88.4	93.7
√	√	√	×	×	97.9	95.2	88.6	93.9
√	√	√	√	×	98.1	96.5	90.0	94.8
√	√	√	√	√	98.1	96.9	90.3	95.2

算法模型	AP			mAP
算法模型	车辆	骑行者	行人	mAP
YOLO v5	97.5	93.8	87.2	92.5
本文算法	98.1	96.9	90.3	95.2

算法	主干网络	mAP/%
SSD	VGG16	77.4
YOLO v3	DarkNet53	85.1
YOLO v4	CSPDarknet53	88.4
YOLO v5	CSPDarknet53	92.5
Faster R-CNN	ResNet-50	72.9
YOLO v5-C2f	CSPDarknet53	93.0
YOLO v5-ECA	CSPDarknet53	92.9
HorNet	HorNet	94.1
ConvNeXt	ConvNeXt	93.0
本文算法	CSPDarknet53	95.2

[1]	Wenxiao WEI, Jieyu LIU, Qiang SHEN, Cheng LI. Feature fusion small target detection algorithm based on human eye view-point map [J]. Systems Engineering and Electronics, 2022, 44(4): 1120-1127.
[2]	Yanling SHI, Zipeng LIU, Xueliang ZHANG, Weiliang GU. Feature detection of floating small target on the sea surface based on EMD energy proportion [J]. Systems Engineering and Electronics, 2021, 43(2): 300-310.
[3]	Ruochen ZHAO, Jingdong WANG, Siyu LIN, Dongze GU. Small building detection algorithm based on convolutional neural network [J]. Systems Engineering and Electronics, 2021, 43(11): 3098-3106.
[4]	MENG Lingbo, GENG Xiurui. Improved principal kurtosis analysis algorithm and its application in hyperspectral imagery small target detection#br# [J]. Systems Engineering and Electronics, 2018, 40(12): 2855-2861.
[5]	IN Jian-lin， PING Xi-jian， MA De-bao. Suppression of driftscanning star based on standardized linear filter [J]. Journal of Systems Engineering and Electronics, 2012, 34(5): 864-870.
[6]	WU Yi-quan, WU Wen-yi, LUO Zi-juan. Infrared small target detection based on least absolute deviation and genetic algorithm [J]. Journal of Systems Engineering and Electronics, 2009, 31(3): 575-578.

Road small target detection based on improved YOLO v5 algorithm

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 31

Related Articles 6

Recommended Articles

Metrics

Comments