State Detection Algorithm of Manipulator Based on Improved YOLOv4 Lightweight Network

doi:10.12068/j.issn.1005-3026.2024.06.002

Abstract

Abstract:

The YOLOv4 network is difficult to be widely used in industry due to its complex structure， many parameters， and large model size. In view of this problem， an improved lightweight network based on YOLOv4 is proposed.Firstly， GhostNet is used to replace the YOLOv4 backbone network， simplifying the network structure and reducing the number of model parameters； Secondly， in order to make up for the accuracy loss caused by network simplification， Spatial Pyramid Pooling structure is added after the other two output feature layers to enhance feature extraction. Thirdly， the attention mechanism of channel， which is Squeeze and Excitation Network， is added to improve the network’s ability to extract important information. Finally， the loss function CIOU is replaced by SIOU to accelerate the convergence of the model and thus produce a better model. Experimental results show that， on the premise of meeting industrial requirements， compared with YOLOv4 network， the improved lightweight network significantly reduces the number of model parameters and the amount of computation， while improving the detection speed， at the same time， at the expense of less detection accuracy， thus proving the effectiveness of the improved algorithm in the identification and detection of the clamping state of the manipulator in the optical fiber plugging task.

Key words: YOLOv4, GhostNet, deepthwise separable convolution, attention mechanism, loss function

CLC Number:

TP 391

Li-xin GUO, Su-tao BI, Ming-yang ZHAO. State Detection Algorithm of Manipulator Based on Improved YOLOv4 Lightweight Network[J]. Journal of Northeastern University(Natural Science), 2024, 45(6): 769-775.

Figures/Tables 13

Fig.1 Structure of YOLOv4 network

Fig.2 Structure of Ghost module

Fig.3 Structure of Ghost bottleneck

Fig.4 Backbone network of improved YOLOv4

Fig.5 Traditional convolution and depthwise

Fig.6 Structure of SENet

Fig.7 Structure of improved YOLOv4 lightweight network

Fig.8 Detection result of improved lightweight YOLOv4

Fig.9 Accuracy change curves of improved lightweight YOLOv4

Fig.10 Recall change curves of improved lightweight YOLOv4

Table 1 Comparison between improved algorithm

模型	参数量×10^-6	计算量×10^-9/FLOPs	FPS	mAP
YOLOv4	63.94	29.88	42	0.983 9
改进算法	11.71	3.97	54	0.980 3

Table 2 Result of ablation test

模块	SE	SPP	SIOU	参数量 ×10^-6	FPS	mAP
YOLOv4	—	—	—	63.94	42	0.983 9
YOLOv4_L	—	—	—	11.01	57	0.951 4
YOLOv4_L	√	—	—	11.05	56	0.960 7
YOLOv4_L	—	√	—	11.67	55	0.965 1
YOLOv4_L	—	—	√	11.01	57	0.959 3
YOLOv4_L	√	√	√	11.71	54	0.980 3

Fig.11 FPS and mAP of different models

References 13

1	Girshick R， Donahue J， Darrell T，et al.Rich feature hierarchies for accurate object detection and semantic segmentation［C］//IEEE Conference on Computer Vision and Pattern Recognition.Columbus：IEEE，2014：580-587.
2	Girshick R.Fast R-CNN［EB/OL］.（2015-04-30）［2022-12-25］.
3	Liu W， Anguelov D， Erhan D，et al.SSD：single shot MultiBox detector［C］//European Conference on Computer Vision.Cham：Springer，2016：21-37.
4	Redmon J， Farhadi A.Yolov3：an incremental improvement ［EB/OL］.（2018-04-08）［2022-12-25］..
5	Bochkovskiy A， Wang C Y， Liao H Y M.YOLOv4：Optimal speed and accuracy of object detection ［EB/OL］.（2020-04-23）［2022-12-25］..
6	He K M， Zhang X Y， Ren S Q，et al.Spatial pyramid pooling in deep convolutional networks for visual recognition ［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence，2015，37（9）：1904-1916.
7	Liu S， Qi L， Qin H F，et al.Path aggregation network for instance segmentation［C］//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City，2018：8759-8768.
8	Wang C Y， Liao H Y， Wu Y H，et al.CSPNet：a new backbone that can enhance learning capability of CNN［C］//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops （CVPRW）.Seattle，2020：1571-1580.
9	Han K， Wang Y H， Tian Q，et al.GhostNet：more features from cheap operations［C］//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）.Seattle，2020：1577-1586.
10	Chollet F.Xception：deep learning with depthwise separable convolutions［C］//2017 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）.Honolulu，2017：1800-1807.
11	Hu J， Shen L， Albanie S，et al.Squeeze‑and‑excitation networks［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence，2020，42（8）：2011-2023.
12	Zheng Z H， Wang P， Liu W，et al.Distance‑IOU loss：faster and better learning for bounding box regression［EB/OL］.（2019-11-19）［2022-12-25］..
	08287.pdf.
13	Gevorgyan Z，SIOU loss：more powerful learning for bounding box regression［EB/OL］.（2022-05-25）［2022-12-25］..

[1]	Hu FENG, Ke-chen SONG, Wen-qi CUI, Yun-hui YAN. Few-Shot Semantic Segmentation of Strip Steel Surface Defects Based on Meta-Learning [J]. Journal of Northeastern University(Natural Science), 2024, 45(3): 354-360.
[2]	Ying SUN, Ya-ru ZHOU, Xue-ying ZHANG. Speech Emotion Recognition Fusing Functional Paralanguage Proportion Coefficient [J]. Journal of Northeastern University(Natural Science), 2024, 45(1): 40-48.
[3]	JIANG Yang， LIU Cheng， DING Qi-chuan， WANG Li. Segmentation of COVID-19 CT Images Based on Dual Attention Mechanism [J]. Journal of Northeastern University(Natural Science), 2023, 44(9): 1259-1268.
[4]	ZHOU Song， GAO Tian-han. EEG Recognition Method for Epileptic Patients Based on RNN Model with Attention Mechanism [J]. Journal of Northeastern University(Natural Science), 2023, 44(8): 1098-1103.
[5]	DING Qi-chuan， WANG Li， LIU Cheng. Classification of Pulmonary Nodule by Combining Long-Distance Channel Attention and Pathological Feature [J]. Journal of Northeastern University(Natural Science), 2023, 44(4): 476-485.
[6]	CHEN Cheng， SHI Pei-xin， WANG Zhan-sheng， JIA Peng-jiao. Shield Load Prediction Method Based on Deep Learning with Multiattention Mechanism [J]. Journal of Northeastern University(Natural Science), 2023, 44(11): 1631-1638.
[7]	LIN Qing-yang， CHEN Xiao-fang， XIE Yong-fang. An Superheat Identification Method in Aluminium Electrolysis Based on Residual Convolutional Self-Attention Neural Network [J]. Journal of Northeastern University(Natural Science), 2023, 44(1): 8-17.
[8]	LIU Yang， YAN Dong-mei， MENG Fan-wei. Improved Two-Branch Person Re-identification Algorithm Based on Transformer [J]. Journal of Northeastern University(Natural Science), 2023, 44(1): 26-32.
[9]	DAI Yin， LIU Wei-bin， DONG Xin-yang， SONG Yu-meng. U-Net CSF Cells Segmentation Based on Attention Mechanism [J]. Journal of Northeastern University(Natural Science), 2022, 43(7): 944-950.
[10]	ZHANG Xue-feng， XU Hua-wen， YANG Mian-zimei. High Perceptual Image Compression Based on Conditional GAN [J]. Journal of Northeastern University(Natural Science), 2022, 43(6): 783-791.
[11]	YU Zhe-zhou， LIU Yan， LIU Yuan-ning. Improved Iris Locating Algorithm Based on YOLOV3 [J]. Journal of Northeastern University(Natural Science), 2022, 43(4): 496-501.
[12]	LI Da-peng， ZHAO Qi-hui， XING Tie-jun， ZHAO Da-zhe. Prison Term Prediction of Judicial Cases Based on Hierarchical Attentive Recurrent Neural Network [J]. Journal of Northeastern University(Natural Science), 2022, 43(3): 344-349.
[13]	GUO Yu， SUN Zhili， PAN Ershun. Online Control and Optimization Strategy of Technological Parameters of Reflow Soldering [J]. Journal of Northeastern University Natural Science, 2014, 35(4): 555-558.