Terahertz image detection combining asymmetric feature attention and feature fusion

Wenjian Zeng; Yan Zhu; Tao Shen; Kai Zeng; Yingli Liu

doi:10.11834/jig.210095

Image Understanding and Computer Vision | Views : 0 下载量: 212 CSCD: 2

PDF
Export
Share
Collection
Album

Terahertz image detection combining asymmetric feature attention and feature fusion
Vol. 27, Issue 8, Pages: 2496-2505(2022)
Received：01 March 2021，

Revised：2021-4-22，

Accepted：29 April 2021，

Published：16 August 2022
DOI： 10.11834/jig.210095
稿件说明：

移动端阅览

DOI：

Wenjian Zeng, Yan Zhu, Tao Shen, Kai Zeng, Yingli Liu. Terahertz image detection combining asymmetric feature attention and feature fusion[J]. Journal of Image and Graphics, 2022, 27(8): 2496-2505. DOI： 10.11834/jig.210095.

摘要

目的

太赫兹由于穿透性强、对人体无害等特性在安检领域中得到了广泛关注。太赫兹图像中目标尺寸较小、特征有限，且图像分辨率低，目标边缘信息模糊，目标信息容易和背景信息混淆，为太赫兹图像检测带来了一定困难。

方法

本文在YOLO(you only look once)算法的基础上提出了一种融合非对称特征注意力和特征融合的目标检测网络AFA-YOLO(asymmetric feature attention-YOLO)。在特征提取网络CSPDarkNet53(cross stage paritial DarkNet53)中设计了非对称特征注意力模块。该模块在浅层网络中采用非对称卷积强化了网络的特征提取能力，帮助网络模型在目标特征有限的太赫兹图像中提取到更有效的目标信息；使用通道注意力和空间注意力机制使网络更加关注图像中目标的重要信息，抑制与目标无关的背景信息；AFA-YOLO通过增加网络中低层到高层的信息传输路径对高层特征进行特征融合，充分利用到低层高分辨率特征进行小目标的检测。

结果

本文在太赫兹数据集上进行了相关实验，相比原YOLOv4算法，AFA-YOLO对phone的检测精度为81.15%，提升了4.12%，knife的检测精度为83.06%，提升了3.72%。模型平均精度均值(mean average precision

mAP)为82.36%，提升了3.92%，漏警率(missing alarm

MA)为12.78%，降低了2.65%，帧率为32.26帧/s，降低了4.06帧/s。同时，本文在太赫兹数据集上对比了不同的检测算法，综合检测速度、检测精度和漏警率，AFA-YOLO优于其他目标检测算法。

结论

本文提出的AFA-YOLO算法在保证实时性检测的同时有效提升了太赫兹图像中目标的检测精度并降低了漏警率。

Abstract

Objective

Terahertz technology has great application potentials in related to wireless communications

biomedicine

and non-destructive testing. Some terahertz imaging features are suitable for hidden objects detection for human security inspections because terahertz waves can penetrate the substance like ceramics

plastics and cloths and are largely absorbed or reflected by metals

liquids and other substances with no harmless on human body. With the development of terahertz imaging technology and the increasing flow of people in application scenarios

the use of artificial recognition of terahertz images is no longer applicable. In order to solve the problem of hidden objects and dangerous goods detection

current research have focused on using deep learning method to classify and analyze them. Due to the resolution and contrast of the terahertz image are low

the edge information of the target in the image is blurred

the target information is easily confused with the background information

and the target information is unclear in the terahertz image

and the feature information is limited. Therefore

the effectiveness issue of feature information to detect the target in terahertz image is challenged for terahertz image detection.

Method

We facilitates a target detection framework asymmetric feature attention-you only look once(AFA-YOLO) that combines asymmetric feature attention and feature fusion based on the you only look once v4(YOLOv4) algorithm to resolve the barriers of small-scale target detection in terahertz images.cross stage paritial DarkNet53(CSPDarkNet53) is as a feature extraction network for AFA-YOLO and an asymmetric feature attention module in CSPDarkNet53 is designed. First

this module uses asymmetric convolution in the shallow network to enhance the feature extraction capabilities of the network

helping the network model to extract more effective target information from the terahertz image with limited target features; Second

the module melts convolutional block attention module(CBAM) attention force mechanism via using the channel attention mechanism to make the model pay more attention to the important information of the target in the image

suppress unrelated background information to the target

and use the spatial attention mechanism to pay attention to the position information of the target in the terahertz image

allowing the model to optimize target contexts. AFA-YOLO has carried out feature fusion operations as well

the high-level features are enhanced through increasing the information transmission path from the low-level to the high-level in the network. The high-level feature map can obtain fine-grained target appearance information and the positioning and detection of small-scale targets in terahertz images can be optimized.

Result

Our research uses the detection accuracy map

missed alarm (MA) rate and detection speed frames per second(FPS) as indicators to carry out related experiments on the terahertz data set. The detection accuracy of AFA-YOLO for the phone is 81.15% compare to the original YOLOv4 algorithm

which is an increase of 4.12%. The detection accuracy of knife is 83.06%

an increased ratio of 3.72%. The model mean average precision(mAP) is 82.36%

which is an increased ratio of 3.92%. The MA is 12.78%

which is a decreased ratio of 2.65%

and the FPS is 32.26

which is lower to 4.06. Additionally

we conduct comparative analysis of different detection algorithms on the terahertz dataset. AFA-YOLO optimized target detection algorithms in terms of recognized detection speed

detection accuracy and missed alarm rate.

Conclusion

We facilitate an AFA-YOLO detection framework that combines asymmetric feature attention and feature fusion. Our YOLOv4-based framework melts asymmetric feature attention module into the shallow network and enhances the target information on the high-level feature map. The optimal information ensures real-time detection through improving the detection accuracy of the target in the terahertz image effectively and lowering the missed alarm rate.

关键词

Keywords

references

Bochkovskiy A, Wang C Y and Liao H Y M. 2020. YOLOv4: optimal speed and accuracy of object detection [EB/OL ] . [2021-03-01 ] . https://arxiv.org/pdf/2004.10934.pdf https://arxiv.org/pdf/2004.10934.pdf

Chen B Q and Deng T. 2019. The vehicle object detection research based on improved SSD algorithm. Journal of Chongqing University of Technology (Natural Science), 33(1): 58-63, 129

陈冰曲, 邓涛. 2019. 基于改进型SSD算法的目标车辆检测研究. 重庆理工大学学报(自然科学), 33(1): 58-63,129 [DOI: 10.3969/j.issn.1674-8425(z).2019.01.009]

Ding X H, Guo Y, Ding G G and Han J G. 2019. ACNet: strengthening the kernel skeletons for powerful CNN via asymmetric convolution blocks//Proceedings of 2019 IEEE/CVF International Conference on Computer Vision. Seoul, Korea (South): IEEE: 1911-1920 [ DOI: 10.1109/ICCV.2019.00200 http://dx.doi.org/10.1109/ICCV.2019.00200 ]

Feng H, An D Y, Tu H, Bu W H, Wang W J, Zhang Y H, Zhang H K, Meng X X, Wei W, Gao B X and Wu S. 2020. A passive video-rate terahertz human body imager with real-time calibration for security applications. Applied Physics B, 126(8): #143 [DOI: 10.1007/s00340-020-07496-3]

Gao M F, Yu R C, Li A, Morariu V I and Davis L S. 2018. Dynamic zoom-in network for fast object detection in large images//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA: IEEE: 6926-6935 [ DOI: 10.1109/CVPR.2018.00724 http://dx.doi.org/10.1109/CVPR.2018.00724 ]

Guo F, Zhang Y X, Tang J and Li W Q. 2021. YOLOv3-A: a traffic sign detection network based on attention mechanism. Journal on Communications, 42(1): 87-99

郭璠, 张泳祥, 唐琎, 李伟清. 2021. YOLOv3-A: 基于注意力机制的交通标志检测网络. 通信学报, 42(1): 87-99 [DOI: 10.11959/j.issn.1000-436x.2021031]

He K M, Gkioxari G, Dollár P and Girshick R. 2017. Mask R-CNN//Proceedings of 2017 IEEE International Conference on Computer Vision. Venice, Italy: IEEE: 2980-2988 [ DOI: 10.1109/ICCV.2017.322 http://dx.doi.org/10.1109/ICCV.2017.322 ]

Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu C Y and Berg A C. 2016. SSD: single shot MultiBox detector//Proceedings of the 14th European Conference on Computer Vision. Amsterdam, the Netherlands: Springer: 21-37 [ DOI: 10.1007/978-3-319-46448-0_2 http://dx.doi.org/10.1007/978-3-319-46448-0_2 ]

Liu Z G, Li D Y, Ge S S and Tian F. 2020. Small traffic sign detection from large image. Applied Intelligence, 50(1): 1-13 [DOI: 10.1007/s10489-019-01511-7]

Mittleman D M. 2017. Perspective: terahertz science and technology. Journal of Applied Physics, 122(23): #230901 [DOI: 10.1063/1.5007683]

Redmon J, Divvala S, Girshick R and Farhadi A. 2016. You only look once: unified, real-time object detection//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Vegas, USA: IEEE: 779-788 [ DOI: 10.1109/CVPR.2016.91 http://dx.doi.org/10.1109/CVPR.2016.91 ]

Redmon J and Farhadi A. 2017. YOLO9000: better, faster, stronger//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA: IEEE: 6517-6525 [ DOI: 10.1109/CVPR.2017.690 http://dx.doi.org/10.1109/CVPR.2017.690 ]

Redmon J and Farhadi A. 2018. YOLOv3: an incremental improvement[EB/OL ] . [2021-03-01 ] . https://arxiv.org/pdf/1804.02767.pdf http://dx.doi.org/https://arxiv.org/pdf/1804.02767.pdf

Ren S Q, He K M, Girshick R and Sun J. 2017. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(6): 1137-1149 [DOI: 10.1109/TPAMI.2016.2577031]

Sommer L W, Schuchert T and Beyerer J. 2017. Fast deep vehicle detection in aerial images//Proceedings of 2017 IEEE Winter Conference on Applications of Computer Vision (WACV). Santa Rosa, USA: IEEE: 311-319 [ DOI: 10.1109/WACV.2017.41 http://dx.doi.org/10.1109/WACV.2017.41 ]

Wang K X, Liew J H, Zou Y T, Zhou D Q and Feng J S. 2019. PANet: few-shot image semantic segmentation with prototype alignment//Proceedings of 2019 IEEE/CVF International Conference on Computer Vision. Seoul, Korea (South): IEEE: 9196-9205 [ DOI: 10.1109/ICCV.2019.00929 http://dx.doi.org/10.1109/ICCV.2019.00929 ]

Woo S, Park J, Lee J Y and Kweon I S. 2018. CBAM: convolutional block attention module//Proceedings of the 15th European Conference on Computer Vision. Munich, Germany: Springer: 3-19 [ DOI: 10.1007/978-3-030-01234-2_1 http://dx.doi.org/10.1007/978-3-030-01234-2_1 ]

Xiao H, Zhang R Y, Wang H, Zhu F, Zhang C, Dai H N and Zhou Y B. 2018. R-PCNN method to rapidly detect objects on THz images in human body security checks//2018 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Cloud and Big Data Computing, Internet of People and Smart City Innovation. Guangzhou, China: IEEE: 1777-1782 [ DOI: 10.1109/SmartWorld.2018.00300 http://dx.doi.org/10.1109/SmartWorld.2018.00300 ]

Yang X, Wu T, Zhang L, Yang D, Wang N N, Song B and Gao X B. 2019. CNN with spatio-temporal information for fast suspicious object detection and recognition in THz security images. Signal Processing, 160: 202-214 [DOI: 10.1016/j.sigpro.2019.02.029]

Zhang J S, Xing W J, Xing M D and Sun G C. 2018. Terahertz image detection with the improved faster region-based convolutional neural network. Sensors, 18(7): #2327 [DOI: 10.3390/s18072327]

Alert me when the article has been cited

提交

Lightweight spaceborne remote sensing object detection algorithm with multi-attention mechanism

Cross-modal feature fusion and detail-enhanced RGB-D salient object detection

Offline recognition of handwritten mathematical expressions： a survey

Two-stage vision transformer for fusing global and local features in distracted driving behavior recognition

Multi-scale attention encoder-decoder network for low-dose dental CBCT denoising