以YOLOv8网络为基础提出一种多尺度渐近金字塔网络MSANet。首先使用带有分层特征融合结构的特征提取块M-Block替换常规卷积层,以增强网络对多尺度目标的特征提取能力;其次引入空间注意力机制（SRU）,抑制背景区域的特征冗余,使网络能更关注重点区域的同时减少参数量的引入;最后提出一种改进渐近金字塔网络AFPNa结构,缓解网络在特征融合过程中信息的丢失或退化问题,提升缺陷检测精度。实验结果表明,与YOLOv8原模型及RTMDET等7种先进检测网络相比,MSANet具有更高的检测精度,相较原模型均值平均精度提升5.7个百分点。
Abstract
A multi-scale asymptotic pyramid network called MSANet is proposed based on the YOLOv8 network. Initially, we replace conventional convolution layers with feature extraction blocks （M-Block） containing a hierarchical feature fusion structure to enhance the network's capability multi-scale feature extraction. Subsequently, we introduce the spatial attention mechanism SRU to suppress feature redundancy in background regions, allowing the network to focus more on crucial areas while reducing the introduction of parameters. Finally, we propose an improved asymptotic pyramid network structure, AFPNa, to mitigate information loss or degradation during the feature fusion process, thereby enhancing defect detection accuracy. Experimental results demonstrate that compared to the original YOLOv8 model and seven other advanced detection networks, including RTMDET, MSANet achieves higher detection accuracy, with a 5.7% improvement in mean average precision compared to the original model.
关键词
缺陷检测 /
深度学习 /
太阳电池 /
分层特征融合结构 /
多尺度渐近金字塔 /
空间注意力机制
Key words
defect detection /
deep learning /
solar cells /
hierarchical feature fusion structure /
multi-scale asymptotic pyramid /
spatial attention mechanism
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 周颖, 王如意, 袁梓桐, 等. 一种高效双路径注意力太阳电池缺陷检测网络[J]. 太阳能学报, 2023, 44(4): 407-413.
ZHOU Y, WANG R Y, YUAN Z T, et al.An efficient dual-path attention solar cell defect detection network[J]. Acta energiae solaris sinica, 2023, 44(4): 407-413.
[2] 张治, 邹鹏辉, 刘志锋, 等. 发射极掺杂工艺对产业化IBC太阳电池性能的影响[J]. 太阳能学报, 2022, 43(3): 158-162.
ZHANG Z, ZOU P H, LIU Z F, et al.Influence of emitter doping process on performance of industrial IBC solar cell[J]. Acta energiae solaris sinica, 2022, 43(3): 158-162.
[3] WANG Y L, HOU T, ZHANG X, et al.Surface defect detection of solar cell based on similarity non-maximum suppression mechanism[J]. Signal, image and video processing, 2023, 17(5): 2583-2593.
[4] 蒋尚俊, 易辉, 李红涛, 等. 基于迁移学习与ResNet的太阳电池缺陷检测方法[J]. 太阳能学报, 2023, 44(7): 116-121.
JIANG S J, YI H, LI H T, et al.Defect detection method of solar cells based on transfer learning and ResNet[J]. Acta energiae solaris sinica, 2023, 44(7): 116-121.
[5] PENG T, WEN F.Photovoltaic panel fault detection based on improved mask R-CNN[C]//2023 IEEE International Conference on Control, Electronics and Computer Technology (ICCECT). Jilin, China, 2023: 1187-1191.
[6] XU M Y.Solar cell defect detection based on improved G-SSD Network[J]. International journal of energy, 2023, 2(1): 68-71.
[7] 王道累, 李明山, 姚勇, 等. 改进SSD的光伏组件热斑缺陷检测方法[J]. 太阳能学报, 2023, 44(4): 420-425.
WANG D L, LI M S, YAO Y, et al.Method of hotspot detection of photovoltaic panels modules on improved SSD[J]. Acta energiae solaris sinica, 2023, 44(4): 420-425.
[8] CHEN Y M, YUAN X B, WU R Q, et al. YOLO-MS: rethinking multi-scale representation learning for real-time object detection[EB/OL].2023: 2308.05480.https://arxiv.org/abs/2308.05480v1.
[9] LI J F, WEN Y, HE L H.SCConv: spatial and channel reconstruction convolution for feature redundancy[C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Vancouver, BC, Canada, 2023: 6153-6162.
[10] YANG G Y, LEI J, ZHU Z K, et al.AFPN: asymptotic feature pyramid network for object detection[C]//2023 IEEE International Conference on Systems, Man, and Cybernetics (SMC). Honolulu, Oahu, HI, USA, 2023: 2184-2189.
[11] SU B Y, ZHOU Z, CHEN H Y.PVEL-AD: a large-scale open-world dataset for photovoltaic cell anomaly detection[J]. IEEE transactions on industrial informatics, 2023, 19(1): 404-413.
[12] REN S Q, HE K M, 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.
[13] TIAN Z, SHEN C H, CHEN H, et al.FCOS: fully convolutional one-stage object detection[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV). Seoul, Korea (South), 2019: 9626-9635.
[14] WANG C Y, BOCHKOVSKIY A, LIAO H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Vancouver, BC, Canada, 2023: 7464-7475.
[15] LYU C Q, ZHANG W W, HUANG H A, et al. RTMDet: an empirical study of designing real-time object detectors[EB/OL].2022: 2212.07784.https://arxiv.org/abs/2212.07784v2.
[16] ZHAO Y A, LYU W Y, XU S L, et al. DETRs beat YOLOs on real-time object detection[EB/OL].2023: 2304.08069.https://arxiv.org/abs/2304.08069v3.
[17] SELVARAJU R R, COGSWELL M, DAS A, et al.Grad-CAM: visual explanations from deep networks via gradient-based localization[J]. International journal of computer vision, 2020, 128(2): 336-359.
基金
国家自然科学基金（61971339）; 陕西省重点研发计划（2019GY-113）; 陕西省自然科学基础研究计划（2019JQ-361）
PDF(3292 KB)
