FS-YOLOv5：轻量化红外目标检测方法

doi:10.3778/j.issn.1002-8331.2210-0487

摘要/Abstract

摘要： 针对传统目标识别算法复杂场景下的道路目标识别精度低、实时性差、小目标检测难度大等问题，提出了基于红外场景下FS-YOLOv5轻量化模型。采用单阶段目标检测网络YOLOv5s作为基础网络，提出了一种新的FS-MobileNetV3网络代替原网络中的CSPDarknet主干网络来提取特征图像；在原网络CIoU损失函数的基础上引入Power变换，替换为α-CIoU，提高网络对小目标的检测能力；将K-means++聚类算法应用在FLIR红外数据集上重新生成Anchor，最后利用DIoU-NMS替换原网络的NMS后处理方法，改善对遮挡物体的检测能力，降低了模型的漏检率。通过在FLIR红外数据集上的消融实验验证了FS-YOLOv5轻量化算法满足红外场景下的道路目标检测任务，与原网络相比，在平均精度仅降低0.37个百分点的前提下，FS-YOLOv5模型的大小减少了26%，参数量减少了29%，检测速度提升了11?FPS，满足了在不同场景下移动端部署的需求。

关键词: 轻量化, 红外目标检测, 损失函数, NMS算法, YOLOv5

Abstract: In order to solve the problems of traditional target recognition algorithm in complex scene, including low precision, poor real-time performance and difficulty in small target detection, an FS-YOLOv5s lightweight model based on infrared scene is proposed. A new FS-MobileNetV3 network is proposed to extract feature images instead of CSPDarknet backbone network, which is based on YOLOv5s, a one-stage target detection network. Based on the CIOU loss function of the original network, a Power transform is introduced, which is replaced by α-CIoU to improve the detection ability of the network to small targets. Then K-means++ clustering algorithm is applied to the FLIR infrared data set to regenerate the Anchor. DIoU-NMS is used to replace the NMS post-processing method of the original network to improve the detection ability of occluded objects and reduce the missed detection rate of the model. The ablation experiments on the FLIR infrared dataset have verified that the FS-YOLOv5s lightweight algorithm can meet the task of road target detection in infrared scenes. Compared with the original network, the average accuracy of the FS-YOLOv5s model is only reduced by 0.37?percentage points. The size is reduced by 26%, the number of parameters is reduced by 29%, and the detection speed is increased by 11?FPS, which meets the needs of mobile deployment in different scenarios.

Key words: lightweight, infrared target detection, loss function, non-maximum suppression（NMS） algorithm, YOLOv5

黄磊, 杨媛, 杨成煜, 杨威, 李耀华. FS-YOLOv5：轻量化红外目标检测方法[J]. 计算机工程与应用, 2023, 59(9): 215-224.

HUANG Lei, YANG Yuan, YANG Chengyu, YANG Wei, LI Yaohua. FS-YOLOv5：Lightweight Infrared Rode Target Detection Method[J]. Computer Engineering and Applications, 2023, 59(9): 215-224.

参考文献

[1] REN S，HE K，GIRSHICK R，et al.Faster R-CNN：towards real-time object detection with region proposal networks[J].IEEE Transactions on Pattern Analysis & Machine Intelligence，2016，39（6）：1137-1149.
[2] LIU W，ANGUELOV D，ERHAN D，et al.SSD：single shot multibox detector[C]//European Conference on Computer Vision，Amsterdam，the Netherlands，October 11-14，2016.Heidelberg：Springer，2016：21-37.
[3] REDMON J，DIVVALA S，GIRSHICK R，et al.You only look once：unified，real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，Las Vegas，NV，USA，June 27-30，2016.Piscataway：IEEE，2016：779-788.
[4] CHEN Y Y，JHONG S Y，LI G Y，et al.Thermal-based pedestrian detection using faster R-CNN and region decomposition branch[C]//2019 International Symposium on Intelligent Signal Processing and Communication Systems（ISPACS），Taipei，Taiwan，China，December 3-6，2019.Piscataway：IEEE，2019：1-2.
[5] 张汝榛，张建林，祁小平，等.复杂场景下的红外目标检测[J].光电工程，2020，47（10）：1-10.
ZHANG R Z，ZHANG J L，QI X P，et al.Infrared target detection and recognition in complex scene[J].Opto-Electronic Engineering，2020，47（10）：1-10.
[6] NARAYANAN A，KUMAR R D，ROSELINKIRUBA R，et al.Study and analysis of pedestrian detection in thermal images using YOLO and SVM[C]//2021 Sixth International Conference on Wireless Communications，Signal Processing and Networking（WiSPNET），Chennai，India，25-27 March 2021.Piscataway：IEEE，2021：431-434.
[7] 赵明，张浩然.一种基于跨域融合网络的红外目标检测方法[J].光子学报，2021，50（11）：331-341.
ZHAO M，ZHANG H R.An infrared obiect detection method based on cross-domain fusion network[J].Acta Photonica Siniica，2021，50（11）：331-341.
[8] 何丽，张红艳，房婉琳.融合多尺度边界特征的显著实例分割[J].计算机科学与探索，2022，16（8）：1865-1876.
HE L，FANG W L，ZHANG H Y.Salient instance segmentation via multiscale boundary characteristic network[J].Journal of Frontiers of Computer Science and Technology，2022，16（8）：1865-1876.
[9] 朱炳宇，刘朕，张景祥.融合Grad-CAM和卷积神经网络的COVID-19检测算法[J].计算机科学与探索，2022，16（9）：2108-2120.
ZHU B Y，LIU Z，ZHANG J X.COVID-19 detection algorithm combining grad-CAM and convolutional neural network[J].Journal of Frontiers of Computer Science and Technology，2022，16（9）：2108-2120.
[10] BOCHKOVSKIY A，WANG C Y，LIAO H Y M.Yolov4：Optimal speed and accuracy of object detection[J].arXiv：2004.10934，2020.
[11] LIN T Y，DOLLáR P，GIRSHICK R，et al.Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，Honolulu，HI，USA，21-26 July 2017.Piscataway：IEEE 2017：2117-2125.
[12] LI H，XIONG P，AN J，et al.Pyramid attention network for semantic segmentation[J].arXiv：1805.10180，2018.
[13] HOWARD A，SANDLER M，CHU G，et al.Searching for mobilenetv3[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision，Seoul，Korea（South），Oct 27-Nov 3，2019.Piscataway：IEEE，2019：1314-1324.
[14] MA N，ZHANG X，ZHENG H T，et al.Shufflenet v2：practical guidelines for efficient CNN architecture design[C]//Proceedings of the European Conference on Computer Vision（ECCV），Munich，Germany，September 8-14，2018.Berlin：Springer，2018：116-131.
[15] LASKAR M T R，HUANG J X，SMETANA V，et al.Extending isolation forest for anomaly detection in big data via K-means[J].ACM Transactions on Cyber-Physical Systems（TCPS），2021，5（4）：1-26.
[16] REZATOFIGHI H，TSOI N，GWAK J Y，et al.Generalized intersection over union：a metric and a loss for bounding box regression[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition，Long Beach Convention & Entertainment Center，Los Angeles CA，United States，Jun 15-21，2019.Piscataway：IEEE，2019：658-666.
[17] ZHENG Z，WANG P，REN D，et al.Enhancing geometric factors in model learning and inference for object detection and instance segmentation[J].IEEE Transactions on Cybernetics，2021：1-13.
[18] HE J，ERFANI S，MA X，et al.Alpha IoU：a family of power intersection over union losses for bounding box regression[C]//Advances in Neural Information Processing Systems，2021：20230-20242.
[19] ZHENG Z，WANG P，LIU W，et al.Distance-IoU loss：faster and better learning for bounding box regression[C]//Proceedings of the AAAI Conference on Artificial Intelligence，2020：12993-13000.