基于改进YOLOv8n的手机屏幕瑕疵检测算法:PGS-YOLO

doi:10.19678/j.issn.1000-3428.0069259

摘要/Abstract

摘要： 手机屏幕作为人机交互的主窗口,已成为影响用户体验和终端整体性能的重要因素。因此,市场对解决手机屏幕瑕疵的需求日益增长。为满足这一需求,针对在手机屏幕瑕疵检测过程中存在检测精度低、小目标瑕疵漏检率高与检测速度慢的情况,提出一种以YOLOv8n作为基准模型的PGS-YOLO算法。PGS-YOLO通过增加一个专门的微小目标检测头,并结合SeaAttention注意力模块,有效提升对小目标的探测能力;将骨干网络和特征融合网络分别融入PConv与GhostNetV2轻量化模块,在保证精度的同时降低模型的参数量,提高瑕疵检测的速度与效率。实验结果表明,在北京大学手机屏幕表面瑕疵数据集中,相较于YOLOv8n,PGS-YOLO算法的mAP@0.5提升了2.5百分点,mAP@0.5∶0.95提升了2.2百分点,在手机屏幕瑕疵检测过程中不仅能对大片的瑕疵做到精准检测,还能对小瑕疵保持一定的准确度。此外,检测性能优于YOLOv5n、YOLOv8s等大部分YOLO系列算法。同时,参数量仅为2.0×10⁶,小于YOLOv8n,满足工业场景对手机屏幕瑕疵检测的需求。

关键词: YOLOv8n模型, 手机屏幕瑕疵检测, 小目标检测, 部分卷积, GhostNetV2轻量化模块, 挤压增强轴向注意力

Abstract: As the main window of human-computer interaction, the mobile phone screen has become an important factor affecting the user experience and the overall performance of the terminal. As a result, there is a growing demand to address defects in mobile phone screens. To meet this demand, in view of the low detection accuracy, high missed detection rate of small target defects, and slow detection speed in the process of defect detection on mobile phone screens, a PGS-YOLO algorithm is proposed, with YOLOv8n as the benchmark model. PGS-YOLO effectively improves the detection ability of small targets by adding a special small target detection head and combining it with the SeaAttention attention module. The backbone and feature fusion networks are integrated into PConv and GhostNetV2 lightweight modules, respectively, to ensure accuracy, reduce the number of model parameters, and improve the speed and efficiency of defect detection. The experimental results show that, in the dataset of mobile phone screen surface defects from Peking University, compared with the results of YOLOv8n, the mAP@0.5 and mAP@0.5∶0.95 of the PGS-YOLO algorithm are increased by 2.5 and 2.2 percentage points, respectively. The algorithm can accurately detect large defects in the process of mobile phone screen defect detection as well as maintain a certain degree of accuracy for small defects. In addition, the detection performance is better than that of most YOLO series algorithms, such as YOLOv5n and YOLOv8s. Simultaneously, the number of parameters is only 2.0×10⁶, which is smaller than that of YOLOv8n, meeting the needs of industrial scenarios for mobile phone screen defect detection.

Key words: YOLOv8n model, mobile phone screen defect detection, small target detection, partial convolution, GhostNetV2 lightweight module, squeeze enhances axial attention

中图分类号:

TP181

周思瑜, 徐慧英, 朱信忠, 黄晓, 盛轲, 曹雨淇, 陈晨. 基于改进YOLOv8n的手机屏幕瑕疵检测算法:PGS-YOLO[J]. 计算机工程, 2025, 51(5): 326-339.

ZHOU Siyu, XU Huiying, ZHU Xinzhong, HUANG Xiao, SHENG Ke, CAO Yuqi, CHEN Chen. Mobile Phone Screen Defect Detection Algorithm Based on Improved YOLOv8n: PGS-YOLO[J]. Computer Engineering, 2025, 51(5): 326-339.

https://www.ecice06.com/CN/Y2025/V51/I5/326

参考文献

[1] 张伟,郑柯,唐娉,等.深度卷积神经网络特征提取用于地表覆盖分类初探[J].中国图象图形学报, 2017, 22(8):1144-1153. ZHANG W, ZHENG K, TANG P, et al. Deep convolutional neural network feature extraction is used for preliminary exploration of land cover classification[J]. China Journal of Image and Graphics, 2017, 22(8):1144-1153.(in Chinese)
[2] 徐天伦,李波,胡文杰,等.基于CA-EfficientDet的棉布瑕疵检测方法[J].中南民族大学学报(自然科学版), 2022, 41(2):243-250. XU T L, LI B, HU W J, et al. Cotton defect detection method based on CA-EfficientDet[J]. Journal of South-Central University for Nationalities (Natural Science Edition), 2022, 41(2):243-250.(in Chinese)
[3] 庄集超,郭保苏,吴凤和.基于可变形密集卷积神经网络的布匹瑕疵检测[J].计量学报, 2023, 44(2):178-185. ZHUANG J C, GUO B S, WU F H. Fabric defect detection based on deformable dense convolutional neural network[J]. Acta Metrologica Sinica, 2023, 44(2):178-185.(in Chinese)
[4] 赵楚,段先华,苏俊楷.改进Faster RCNN的瓷砖表面瑕疵检测研究[J].计算机工程与应用, 2023, 59(14):201-208. ZHAO C, DUAN X H, SU J K. Research on ceramic tile surface defect detection by improved faster RCNN[J]. Computer Engineering and Applications, 2023, 59(14):201-208.(in Chinese)
[5] LI F, LI F, XI Q G. DefectNet:toward fast and effective defect detection[J]. IEEE Transactions on Instrumentation Measurement, 2021, 70:3067221.
[6] 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.
[7] 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, 2017:7263-7271.
[8] REDMON J, FARHADI A. YOLOv3:an incremental improvement[EB/OL].[2023-09-10] . https://arxiv.org/abs/1804.02767v1.
[9] BOCHKOVSKIY A, WANG C Y, LIAO H M. YOLOv4:optimal speed and accuracy of object detection[EB/OL].[2023-09-10] . https://arxiv.org/abs/2004.10934v1.
[10] 陈益方,张上,冉秀康,等.基于改进YOLOv8的SAR图像飞机目标检测算法[J].电讯技术,2024,64(8):1206-1212. CHEN Y F, ZHANG S, RAN X K, et al. An aircraft target detection algorithm based on improved YOLOv8 in SAR Image[J]. Telecommunication Engineering,2024,64(8):1206-1212.(in Chinese)
[11] LI C Y, LI L, JIANG H L, et al. YOLOv6:a single-stage object detection framework for industrial applications[EB/OL].[2023-09-10] . https://arxiv.org/abs/2209.02976.
[12] WANG C Y, BOCHKOVSKIY A, LIAO H M. YOLOv7:trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2023:7464-7475.
[13] JOCHER G, CHAURASAI A, QIU J. Ultralytics YOLOv8(8.0.0)[EB/OL].[2023-09-10] . https://github.com/ultralytics/ultralytics.
[14] WANG C Y, YEH I H, LIAO H M. YOLOv9:learning what you want to learn using programmable gradient information[EB/OL].[2023-09-10] . https://arxiv.org/abs/2402.13616v2.
[15] 袁华清,刘桂华,王娅琼.基于改进YOLOv3-tiny的织物表面小目标瑕疵检测[J].制造业自动化, 2022, 44(12):172-176. YUAN H Q, LIU G H, WANG Y Q. Defect detection of small targets on fabric surface based on improved YOLOv3-tiny[J]. Manufacturing Automation, 2022, 44(12):172-176.(in Chinese)
[16] 郭峰,朱启兵,黄敏,等.基于改进YOLOV4的陶瓷基板瑕疵检测[J].光学精密工程, 2022, 30(13):1631. GUO F, ZHU Q B, HUANG M, et al. Defect detection in ceramic substrate based on improved YOLOV4[J]. Optics and Precision Engineering, 2022, 30(13):1631.(in Chinese)
[17] 刘俊豪,王美林,谢兴,等.基于改进YOLOv5的皮革瑕疵检测算法[J].计算机工程, 2023, 49(8):240-249. LIU J H, WANG M L, XIE X, et al. Leather defect detection algorithm based on improved YOLOv5[J]. Computer Engineering, 2023, 49(8):240-249.(in Chinese)
[18] LI B Y, LIU Y, WANG X G, et al. Gradient harmonized single-stage detector[C]//Proceedings of the 33rd AAAI Conference on Artificial Intelligence and 31st Innovative Applications of Artificial Intelligence Conference and Ninth AAAI Symposium on Educational Advances in Artificial Intelligence. New York, USA:ACM Press, 2019:8577-8584.
[19] WANG X, GAO H, JIA Z, et al. BL-YOLOv8:an improved road defect detection model based on YOLOv8[J]. Sensors, 2023, 23(20):8361.
[20] WAN Q, HUANG Z L, LU J C, et al. SeaFormer++:squeeze-enhanced axial transformer for mobile visual recognition[EB/OL].[2023-09-10] . https://arxiv.org/abs/2301.13156v5.
[21] CHEN J R, KAO S H, HE H, et al. Run, don't walk:chasing higher FLOPS for faster neural networks[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2023:12021-12031.
[22] TANG Y, HAN K, GUO J, et al. GhostNetv2:enhance cheap operation with long-range attention[C]//Proceedings of Advances in Neural Information Processing Systems. Cambridge, USA:MIT Press, 2022:9969-9982.
[23] ZHU C C, HE Y H, SAVVIDES M. Feature selective anchor-free module for single-shot object detection[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2019:840-849.
[24] TIAN Z, SHEN C H, CHEN H, et al. FCOS:fully convolutional one-stage object detection[C]//Proceedings of IEEE/CVF International Conference on Computer Vision. Washington D. C., USA:IEEE Press, 2019:9627-9636.
[25] KONG T, SUN F C, LIU H P, et al. FoveaBox:beyound anchor-based object detection[J]. IEEE Transactions on Image Processing, 2020, 29:7389-7398.
[26] LIU W, LIAO S C, REN W Q, et al. High-level semantic feature detection:a new perspective for pedestrian detection[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2019:5187-5196.
[27] WANG J Q, CHEN K, YANG S, et al. Region proposal by guided anchoring[C]// Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2019:2965-2974.
[28] HAN K, WANG Y H, TIAN Q, et al. GhostNet:more features from cheap operations[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2020:1580-1589.
[29] ZHANG J, BAN M. PKU-market-phone[EB/OL].[2023-09-10] . https://github.com/jianzhang96/MSD.
[30] GHIASI G, CUI Y, SRINIVAS A, et al. Simple copy-paste is a strong data augmentation method for instance segmentation[C]// Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2021:2918-2928.
[31] JIANG Y Q, TAN Z Y, WANG J Y, et al. GiraffeDet:a heavy-neck paradigm for object detection[EB/OL].[2023-09-10] . https://arxiv.org/abs/2202.04256v2.
[32] HAN K, WANG Y H, XU C, et al. GhostNets on heterogeneous devices via cheap operations[J]. International Journal of Computer Vision, 2022, 130(4):1050-1069.
[33] CHEN C P, GUO Z C, ZENG H E, et al. RepGhost:a hardware-efficient ghost module via re-parameterization[EB/OL].[2023-09-10] . https://arxiv.org/abs/2211.06088v2.
[34] LIU Y C, SHAO Z R, HOFFMANN N. Global attention mechanism:retain information to enhance channel-spatial interactions[EB/OL].[2023-09-10] . https://arxiv.org/abs/2112.05561v1.
[35] YANG L, ZHANG R Y, LI L, et al. Simam:a simple, parameter-free attention module for convolutional neural networks[C]//Proceedings of International Conference on Machine Learning. Washington D. C., USA:IEEE Press, 2021:11863-11874.
[36] WOO S, PARK J, LEE J Y, et al. CBAM:convolutional block attention module[C]//Proceedings of International Conference on Computer Vision. Berlin, Germany:Springer, 2018:3-19.
[37] LI Y X, HOU Q B, ZHENG Z H, et al. Large selective kernel network for remote sensing object detection[EB/OL].[2023-09-10] . https://arxiv.org/abs/2303.09030v2.
[38] LIU Y C, SHAO Z R, TENG Y Y, et al. NAM:normalization-based attention module[EB/OL].[2023-09-10] . https://arxiv.org/abs/2111.12419v1.
[39] YU Y, ZHANG Y, CHENG Z, et al. MCA:multidimensional collaborative attention in deep convolutional neural networks for image recognition[EB/OL].[2023-09-10] . https://github.com/ndsclark/MCANet.
[40] CAI Z W, VASCONCELOS N. Cascade R-CNN:delving into high quality object detection[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA:IEEE Press, 2018:6154-6162.
[41] 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 and Machine Intelligence, 2017, 39(6):1137-1149.
[42] LIU W, ANGUELOV D, ERHAN D, et al. SSD:single shot MultiBox detector[C]//Proceedings of the 14th European Conference on Computer Vision. Berlin, Germany:Springer, 2016:21-37.
[43] ZHAO Y A, LV W Y, XU S L, et al. DETRs beat YOLOs on real-time object detection[EB/OL].[2023-09-10] .https://arxiv.org/abs/2304.08069v3.
[44] 倪富陶,李倩,聂云靖,等.基于改进YOLOv8的轻量化钢筋端面检测算法研究[J].太原理工大学学报, 2024, 55(4):696-704. NI F T, LI Q, NIE Y J, et al. Lightweight rebar end detection algorithm based on improved YOLOv8[J]. Taiyuan University of Technology, 2024, 55(4):696-704.(in Chinese)
[45] 刘子洋,徐慧英,朱信忠,等. Bi-YOLO:一种基于YOLOv8改进的轻量化目标检测算法[J].计算机工程与科学, 2024, 46(8):1444-1454. LIU Z Y, XU H Y, ZHU Z X, et al. Bi-YOLO:an improved lightweight object detection algorithm based on YOLOv8n[J]. Computer Engineering and Science, 2024, 46(8):1444-1454.(in Chinese)
[46] WANG C C, HE W, NIE Y, et al. Gold-YOLO: efficient object detector via gather-and-distribute mechanism[EB/OL].[2023-09-10] . https://arxiv.org/abs/2309.11331.
[47] DING R W, DAI L H, LI G P, et al. TDD-Net:a tiny defect detection network for printed circuit boards[J]. CAAI Transactions on Intelligence Technology, 2019, 4(2):110-116.
[48] SELVARAJU R R, COGSWELL M, DAS A, et al. Grad-CAM:visual explanations from deep networks via gradient-based localization[C]//Proceedings of IEEE International Conference on Computer Vision. Washington D. C., USA:IEEE Press, 2017:618-626.

选择文件类型/文献管理软件名称

选择包含的内容