Research on Lightweight Road-Target-Recognition Algorithm in Complex Environment

doi:10.19678/j.issn.1000-3428.0067576

Abstract

Abstract:

Road-target recognition is a core technology used in intelligent transportation systems to solve urban congestion problems. However, existing algorithms exhibit unsatisfactory recognition performance in complex traffic environments, with numerous missed and false detections. Moreover, the model parameters are large, thus rendering them unsuitable for deployment on resource-limited mobile devices in practical scenarios. Hence, a lightweight road-target-recognition algorithm for complex environments is proposed in this study. A reconfigurable feature-extraction framework is designed based on the structure of the Single Shot Multi-Box Detector (SSD) algorithm. Three lightweight modules are used to construct shallow feature-extraction networks, and a custom Additional Block is used to construct deep-feature-extraction networks. The channel attention mechanism and a Lightweight Receptive Field Expansion (RFB-L) module are used to improve the detection performance of the model on targets of various sizes. Utilizing a custom pixel and a channel-information-fusion module to combine shallow and deep features enriches the information in the detection feature map. Meanwhile, a multi-feature, fusion learning-rate-adjustment algorithm is proposed to ensure the stable convergence of the model during training. A custom-developed dataset reflecting the complex and congested road of Hohhot_city is used to train and test the proposed algorithm. Comparative experimental results yielded by mainstream algorithms show that the proposed algorithm performs significantly better than YOLOv4-tiny and YOLOv5s algorithms under the same number of parameters. Its detection accuracy is similar to that of the YOLOv5m algorithm when the parameters are less than 40%. Additionally, its inference time and mean Average Precision (mAP) are 12.8 ms and 99.1%, respectively.

Key words: road-target recognition, feature extraction, feature fusion, channel attention, receptive field

摘要：

道路目标识别是智能交通系统解决城市拥堵问题的核心技术之一, 然而现有算法在复杂交通环境下识别效果较差, 存在大量漏检和误检情况, 且模型参数量大, 不适合在实际场景下部署于资源有限的移动端设备。针对以上问题, 提出一种复杂环境下的轻量化道路目标识别算法。基于SSD算法结构设计一种可重构的特征提取网络框架, 利用3种轻量化模块分别构建浅层特征提取网络, 以自定义的Additional Block构建深层特征提取网络, 并分别采用通道注意力机制和轻量化感受野扩大(RFB-L)模块提升模型对各尺寸目标的检测效果。利用自定义的像素与通道信息融合模块实现浅层与深层特征的融合, 丰富待检测特征图包含的信息。同时, 提出一种多特征融合的学习率调节算法, 使得训练过程中模型性能稳定地达到收敛。自制复杂拥堵道路数据集Hohhot_city用于算法训练和测试, 与主流算法的对比实验结果表明, 该算法性能明显优于参数量同级别的YOLOv4-tiny和YOLOv5s算法, 在参数量不到YOLOv5m算法40%的情况下与其检测精度接近, 并取得了12.8 ms的推理时间和99.1%的均值平均精度。

关键词: 道路目标识别, 特征提取, 特征融合, 通道注意力, 感受野

Zhenlu LI, Wei HUANG, Kai SUN. Research on Lightweight Road-Target-Recognition Algorithm in Complex Environment[J]. Computer Engineering, 2024, 50(4): 219-227.

李振鲁, 黄威, 孙锴. 复杂环境下的轻量化道路目标识别算法研究[J]. 计算机工程, 2024, 50(4): 219-227.

/ / Recommend / Download Citations

URL: http://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0067576

http://www.ecice06.com/EN/Y2024/V50/I4/219

Figures/Tables 17

Fig.1 PAN structure

Fig.2 The structure of the algorithm in this paper

Fig.3 Reconfigurable feature extraction network

Fig.4 Base module structure

Fig.5 Additional module structure

Fig.6 ECA and improved In-residual block

Fig.7 Improved Additional block and RFB-L

Fig.8 Upsample and Concat

Fig.9 Improved PAN block and Fusion

Fig.10 Dataset target labeling

Fig.11 Dataset composition

Fig.12 Training curve after using the algorithm in this paper

Fig.13 The actual detection effect of the algorithm in this paper

References 28

1	AIT OUALLANE A, BAKALI A, BAHNASSE A, et al. Fusion of engineering insights and emerging trends: intelligent urban traffic management system. Information Fusion, 2022, 88, 218- 248. doi: 10.1016/j.inffus.2022.07.020
2	张小瑞, 陈旋, 孙伟, 等. 基于深度学习的车辆再识别研究进展. 计算机工程, 2020, 46(11): 1- 11. URL
	ZHANG X R, CHEN X, SUN W, et al. Progress of vehicle re-identification research based on deep learning. Computer Engineering, 2020, 46(11): 1- 11. URL
3	GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C. , USA: IEEE Press, 2014: 580-587.
4	REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137- 1149. doi: 10.1109/TPAMI.2016.2577031
5	HE K M, GKIOXARI G, DOLLAR P, et al. Mask R-CNN[C]//Proceedings of IEEE International Conference on Computer Vision. Washington D. C. , USA: IEEE Press, 2017: 2980-2988.
6	GHOSH R. On-road vehicle detection in varying weather conditions using Faster R-CNN with several region proposal networks. Multimedia Tools and Applications, 2021, 80(17): 25985- 25999. doi: 10.1007/s11042-021-10954-5
7	DAI X B, HU J P, ZHANG H M, et al. Multi-task Faster R-CNN for nighttime pedestrian detection and distance estimation. Infrared Physics & Technology, 2021, 115, 103694.
8	LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[C]//Proceedings of European Conference on Computer Vision. Berlin, Germany: Springer, 2016: 21-37.
9	REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C. , USA: IEEE Press, 2016: 779-788.
10	REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Washington D. C. , USA: IEEE Press, 2017: 6517-6525.
11	REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. [2023-04-05]. https:arxiv.org/abs/1701.06659.
12	LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C]//Proceedings of IEEE International Conference on Computer Vision. Washington D. C. , USA: IEEE Press, 2017: 2980-2988.
13	CHEN S L, HONG J, ZHANG T, et al. Object detection using deep learning: single shot detector with a refined feature-fusion structure[C]//Proceedings of IEEE International Conference on Real-time Computing and Robotics. Washington D. C. , USA: IEEE Press, 2019: 219-224.
14	李国进, 胡洁, 艾矫燕. 基于改进SSD算法的车辆检测. 计算机工程, 2022, 48(1): 266- 274. URL
	LI G J, HU J, AI J Y. Vehicle detection based on improved SSD algorithm. Computer Engineering, 2022, 48(1): 266- 274. URL
15	GUO X Y, LIU Q L, QIN Z K, et al. Target detection of forward vehicle based on improved SSD[C]//Proceedings of the 6th IEEE International Conference on Cloud Computing and Big Data Analytics. Washington D. C. , USA: IEEE Press, 2021: 466-468.
16	JAIN N, YERRAGOLLA S, GUHA T, et al. Performance analysis of object detection and tracking algorithms for traffic surveillance applications using neural networks[C]//Proceedings of the 3rd International Conference on I-SMAC. Washington D. C. , USA: IEEE Press, 2019: 690-696.
17	MAURI A, KHEMMAR R, DECOUX B, et al. Lightweight convolutional neural network for real-time 3D object detection in road and railway environments. Journal of Real-Time Image Processing, 2022, 19(3): 499- 516. doi: 10.1007/s11554-022-01202-6
18	JACOB B, KLIGYS S, CHEN B, et al. Quantization and training of neural networks for efficient integer-arithmetic-only inference[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C. , USA: IEEE Press, 2018: 2704-2713.
19	YOUNG S I, ZHE W, TAUBMAN D, et al. Transform quantization for CNN compression. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(9): 5700- 5714.
20	HAN S, POOL J, TRAN J, et al. Learning both weights and connections for efficient neural network[C]//Proceedings of the 28th International Conference on Neural Information Processing Systems. Cambridge, USA: MIT Press, 2015: 1135-1143.
21	RIERA M, ARNAU J M, GONZÁLEZ A. DNN pruning with principal component analysis and connection importance estimation. Journal of Systems Architecture, 2022, 122, 102336. doi: 10.1016/j.sysarc.2021.102336
22	KUSHAWAHA R K, KUMAR S, BANERJEE B, et al. Distilling spikes: knowledge distillation in spiking neural networks[C]//Proceedings of the 25th International Conference on Pattern Recognition. Washington D. C. , USA: IEEE Press, 2021: 4536-4543.
23	LIN Y, WANG C F, CHANG C Y, et al. An efficient framework for counting pedestrians crossing a line using low-cost devices: the benefits of distilling the knowledge in a neural network. Multimedia Tools and Applications, 2021, 80(3): 4037- 4051. doi: 10.1007/s11042-020-09276-9
24	HOWARD A G, ZHU M L, CHEN B, et al. MobileNets: efficient convolutional neural networks for mobile vision applications[EB/OL]. [2023-04-05]. https://doi.org/10.48550/arXiv.1704.04861.
25	GAO M, JIN L S, JIANG Y Y, et al. Multiple object tracking using a dual-attention network for autonomous driving. IET Intelligent Transport Systems, 2020, 14(8): 842- 848. doi: 10.1049/iet-its.2019.0536
26	HU J, SHEN L, ALBANIE S, et al. Squeeze-and-excitation networks[EB/OL]. [2023-04-05]. https://doi.org/10.48550/arXiv.1709.01507.
27	LIU S T, HUANG D, WANG Y H. Receptive field block net for accurate and fast object detection[EB/OL]. [2023-04-05]. https://doi.org/10.48550/arXiv.1711.07767.
28	姜竣, 翟东海. 基于空洞卷积与特征增强的单阶段目标检测算法. 计算机工程, 2021, 47(7): 232-238, 248. URL
	JIANG J, ZHAI D H. Single-stage object detection algorithm based on dilated convolution and feature enhancement. Computer Engineering, 2021, 47(7): 232-238, 248. URL

[1]	CUI Liqun, CAO Huawei. Target Detection of Remote-Sensing Images Based on Improved YOLOv5 [J]. Computer Engineering, 2024, 50(4): 228-236.
[2]	Yu AN, Haibo GE, Wenhao HE, Sai MA, Mengyang CHENG. Siamese Network Tracking Algorithm Based on Compensated Attention Mechanism [J]. Computer Engineering, 2024, 50(4): 187-196.
[3]	LIU Yanhong, YANG Qiuxiang, HU Shuai. Research on Multi-Scale Feature Fusion Dehazing Network Based on Feature Differences [J]. Computer Engineering, 2024, 50(4): 247-257.
[4]	Wentao YUAN, Wentao WEI, Demin GAO. Research on Multiview Convolutional Gesture Recognition with Fusion Attention Mechanism [J]. Computer Engineering, 2024, 50(3): 208-215.
[5]	Yanzhou FENG, Jianxia LIU, Haiyi WANG, Guohao FENG, Yu BAI. Real Image Denoising Method Based on Multi-Level Residual Information Distillation [J]. Computer Engineering, 2024, 50(3): 216-223.
[6]	Shuaikang XIE, Fengguang XIONG, Xinjie ZHU, Ningdong SONG, Wenqing LI, Tingfeng WANG. Three-Dimensional Point Cloud Registration Method Based on Spatial Deformable Transformer [J]. Computer Engineering, 2024, 50(3): 224-232.
[7]	Weijian LI, Huijun HU. Video Description Generation Method Based on Latent Feature Augmented Network [J]. Computer Engineering, 2024, 50(2): 266-272.
[8]	Haochen XU, Manhua LIU. Facial Landmark Detection Based on Hierarchical Self-Attention Network [J]. Computer Engineering, 2024, 50(2): 239-246.
[9]	Long SUN, Rongfen ZHANG, Yuhong LIU, Tingli RAO. Mask Wearing Detection Algorithm for Dense Crowds from a Monitoring Perspective [J]. Computer Engineering, 2023, 49(9): 313-320.
[10]	Wenzhuo FAN, Tao WU, Junping XU, Qingqing LI, Jianlin ZHANG, Meihui LI, Yuxing WEI. Super-Resolution Reconstruction of Arbitrary Scale Images Based on Multi-Resolution Feature Fusion [J]. Computer Engineering, 2023, 49(9): 217-225.
[11]	Lu HAN, Weigang HUO, Yonghui ZHANG, Tao LIU. Multivariate Time Series Forecasting Based on Multi-Scale Feature Fusion and Dual-Attention Mechanism [J]. Computer Engineering, 2023, 49(9): 99-108.
[12]	Jiaxin LI, Jin HOU, Boying SHENG, Yuhang ZHOU. Remote Sensing Small Object Detection Network Based on Improved YOLOv5 [J]. Computer Engineering, 2023, 49(9): 256-264.
[13]	Shanshu BAO, Bo CHE, Linhong DENG. Lung Sound Signal Recognition Based on Dual-Source Domain Transfer Learning [J]. Computer Engineering, 2023, 49(9): 295-302, 312.
[14]	Zhina SONG, Sha LI, Jianming YANG, Chuan XU. Remote Sensing Ship Target Detection Based on Feature and Region Localization Enhancement [J]. Computer Engineering, 2023, 49(8): 257-264.
[15]	Ang LU, Jun CHU, Lu LENG. Image Dehazing Based on High-Frequency and Low-Frequency Feature Enhancement [J]. Computer Engineering, 2023, 49(8): 174-181.

Please choose a citation manager

Content to export