收稿日期:
2023-02-27
修回日期:
2023-03-29
接受日期:
2023-09-22
出版日期:
2024-01-15
发布日期:
2023-11-07
通讯作者:
王小平
E-mail:levine@nuaa.edu.cn
基金资助:
Received:
2023-02-27
Revised:
2023-03-29
Accepted:
2023-09-22
Online:
2024-01-15
Published:
2023-11-07
Contact:
Xiaoping WANG
E-mail:levine@nuaa.edu.cn
Supported by:
摘要:
自动铺丝构件在实际生产过程中会受到环境和工艺等因素的影响,从而导致缺陷的产生并对铺丝构件以及装配整体带来严重危害。由于自动铺丝成型技术的保密性以及制造工艺昂贵和耗时等因素,有关缺陷图像的样本数据集难以获取,造成了小样本学习的问题。提出通过虚拟样本生成的方法来扩容数据集并将其作为目标检测网络的输入训练样本。针对铺丝缺陷特点,在YOLOv5s中添加坐标注意力机制来优化设计,得到最终的目标检测网络模型并命名为Place-YOLO,其精度值为0.941,平均精度均值mAP为0.833,与原始YOLOv5s相比提升了1.8%。
中图分类号:
甘纪强, 王小平. 基于虚拟样本生成的铺丝表面缺陷检测[J]. 航空学报, 2024, 45(1): 428624-428624.
Jiqiang GAN, Xiaoping WANG. Surface defect detection of fiber placement based on virtual sample generation[J]. Acta Aeronautica et Astronautica Sinica, 2024, 45(1): 428624-428624.
表 3
ConSinGAN其他重要参数设置
参数 | 含义 | 设置值 |
---|---|---|
nfc | Number of filters per conv layer | 64 |
ker_size | Kernel size | 3 |
num_layer | Number of layers per stage | 3 |
padd_size | Net pad size | 0 |
nc_im | Image channels | 3 |
noise_amp | Additive noise cont weight | 0.1 |
min_size | Image minimal size at the coarser scale | 25 |
max_size | Image maximal size at the coarser scale | 250 |
train_depth | How many layers are trained if growing | 3 |
lr_g | Learning rate of generator | 0.000 5 |
lr_d | Learning rate of discriminator | 0.000 5 |
beta1 | Beta1 for adam | 0.5 |
lambda_grad | Gradient penalty weight | 0.1 |
alpha | Reconstruction loss weight | 10 |
1 | 原崇新, 李妍, 潘杰, 等. 自动铺丝过程中的典型缺陷及原因分析[J]. 航空制造技术, 2019, 62(4): 66-74. |
YUAN C X, LI Y, PAN J, et al. Typical defects and causes analysis of automated fiber placement[J]. Aeronautical Manufacturing Technology, 2019, 62(4): 66-74 (in Chinese). | |
2 | 张小辉, 朱玉祥, 张少秋, 等. 先进复合材料自动铺丝技术研究进展[J]. 航空制造技术, 2018, 61(7): 54-61. |
ZHANG X H, ZHU Y X, ZHANG S Q, et al. Research progress on automated fiber placement technology[J]. Aeronautical Manufacturing Technology, 2018, 61(7): 54-61 (in Chinese). | |
3 | WANIGASEKARA C, OROMIEHIE E, SWAIN A, et al. Machine learning based predictive model for AFP-based unidirectional composite laminates[J]. IEEE Transactions on Industrial Informatics, 2020,16(4): 2315-2324. |
4 | SMITH R, QURESHI Z, SCAIFE R, et al. Limitations of processing carbon fibre reinforced plastic/polymer material using automated fibre placement technology[J]. Reinforced Plastics & Composites, 2016, 35(21): 1527-1542. |
5 | HEINECKE F, WILLBERG C. Manufacturing-induced imperfections in composite parts manufactured via automated fiber placement[J]. Composites Science, 2019, 3(2): 56-56. |
6 | 丁希仑, 罗伟恒, 刘斐, 等. 自动铺丝成型构件缺陷在线检测技术进展[J]. 北京航空航天大学学报, 2022, 48(9): 1721-1733. |
DING X L, LUO W H, LIU F, et al. Review on automated fiber placement induced defects and their online monitoring technology[J]. Journal of Beijing University of Aeronautics and Astronautics, 2022, 48(9): 1721-1733 (in Chinese). | |
7 | 倪金辉. 基于机器视觉的预浸纱缺陷检测系统的研究[D]. 南京: 南京航空航天大学, 2015. |
NI J H, Defect detection system for prepreg based on machine vision[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2015 (in Chinese). | |
8 | 康硕, 柯臻铮, 王璇, 等. 基于红外和可见光图像融合的铺丝缺陷检测方法[J]. 航空学报, 2022, 43(3): 425187. |
KANG S, KE Z Z, WANG X, et al. Detection method of defects in automatic fiber placement based on fusion of infrared and visible images[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(3): 425187 (in Chinese). | |
9 | 蔡志强, 肖军, 文立伟, 等. 基于预浸纱自动铺放缺陷的分割算法[J]. 航空材料学报, 2017, 37(2): 21-27. |
CAI Z Q, XIAO J, WEN L W, et al. Algorithm of defect segmentation for AFP based on prepregs[J]. Journal of Aeronautical Materials, 2017, 37(2): 21-27 (in Chinese). | |
10 | 吕昊远, 俞璐, 周星宇, 等. 半监督深度学习图像分类方法研究综述[J]. 计算机科学与探索, 2021, 15(6): 1038-1048. |
LV H Y, YU L, ZHOU X Y, et al. Review of semi-supervised deep learning image classification methods[J]. Journal of Frontiers of Computer Science & Technology, 2021, 15(6): 1038-1048 (in Chinese). | |
11 | 赵朗月, 吴一全. 基于机器视觉的表面缺陷检测方法研究进展[J]. 仪器仪表学报, 2022, 43(1): 198-219. |
ZHAO L Y, WU Y Q. Research progress of surface defect detection methods based on machine vision[J]. Chinese Journal of Scientific Instrument, 2022, 43(1): 198-219 (in Chinese). | |
12 | 汤勃, 孔建益, 伍世虔. 机器视觉表面缺陷检测综述[J]. 中国图象图形学报, 2017, 22(12): 1640-1663. |
TANG B, KONG J Y, WU S Q. Review of surface defect detection based on machine vision[J]. Journal of Image and Graphics, 2017, 22(12): 1640-1663 (in Chinese). | |
13 | SHAHAM R T, DEKEL T, MICHAELI T. SinGAN: learning a generative model from a single natural image [C]∥Proceedings of the IEEE International Conference on Computer Vision. 2019: 4570-4580. |
14 | 刘剑超, 相洁, 张玲, 等. 一种自约束的小样本缺损图像分割方法[J]. 小型微型计算机系统, 2022, 43(8): 1732-1738. |
LIU J C, XIANG J, ZHANG L, et al. Level-set rectified U-net for few-shot fouling image segmentation[J]. Journal of Chinese Mini-Micro Computer Systems, 2022, 43(8): 1732-1738 (in Chinese). | |
15 | 王坤峰, 苟超, 段艳杰, 等. 生成式对抗网络GAN的研究进展与展望[J]. 自动化学报, 2017, 43(3): 321-332. |
WANG K F, GOU C, DUAN Y J, et al. Generative adversarial networks: The state of the art and beyond[J]. Acta Automatica Sinica, 2017, 43(3): 321-332 (in Chinese). | |
16 | HINZ T, FISHER M, WANG O, et al. Improved techniques for training single-image GANs[C]∥2021 IEEE Winter Conference on Applications of Computer Vision. 2021: 1299-1308. |
17 | 刘伟韬. 基于SinGAN的图像生成模型[J]. 曲阜师范大学学报(自然科学版), 2020, 46(2): 67-71. |
LIU W T. Image generative model based on single generative adversarial network[J]. Journal of Qufu Normal University (Natural Science Edition), 2020, 46(2): 67-71 (in Chinese). | |
18 | 程国建, 张福临. 基于SinGAN的岩石薄片图像超分辨率重建[J]. 西安石油大学学报(自然科学版), 2021, 36(2): 116-121. |
CHENG G J, ZHANG F L. Super-resolution reconstruction of rock slice image based on SinGAN[J]. Journal of Xi’an Shiyou University (Natural Science Edition), 2021, 36(2): 116-121 (in Chinese). | |
19 | SONGWEI G, RUI Z, HONGXIA L, et al. Improved SinGAN integrated with an attentional mechanism for remote sensing image classification[J]. Remote Sensing, 2021, 13(9): 1713. |
20 | 王星, 高峰, 陈吉, 等. 基于GAN网络的煤岩图像样本生成方法[J]. 煤炭学报, 2021, 46(9): 3066-3078. |
WANG X, GAO F, CHEN J, et al. Generative adversarial networks based sample generation of coal and rock images[J]. Journal of China Coal Society, 2021, 46(9): 3066-3078 (in Chinese). | |
21 | 尹宏鹏, 陈波, 柴毅, 等. 基于视觉的目标检测与跟踪综述[J]. 自动化学报, 2016, 42(10): 1466-1489. |
YIN H P, CHEN B, CHAI Y, et al. Vision-based object detection and tracking: A review[J]. Acta Automatica Sinica, 2016, 42(10): 1466-1489 (in Chinese). | |
22 | 王玲敏, 段军, 辛立伟. 引入注意力机制的YOLOv5安全帽佩戴检测方法[J]. 计算机工程与应用, 2022, 58(9): 303-312. |
WANG L M, DUAN J, XIN L W. YOLOv5 helmet wear detection method with introduction of attention mechanism[J]. Computer Engineering and Applications, 2022, 58(9): 303-312 (in Chinese). | |
23 | HOU Q, ZHOU D, FENG J. Coordinate attention for efficient mobile network design[C]∥2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021: 13708-13717. |
24 | 任欢, 王旭光. 注意力机制综述[J]. 计算机应用, 2021, 41(S1): 1-6. |
REN H, WANG X G. Review of attention mechanism[J]. Journal of Computer Applications, 2021, 41(S1): 1-6 (in Chinese). | |
25 | 赵璐璐, 王学营, 张翼, 等. 基于YOLOv5s融合SENet的车辆目标检测技术研究[J]. 图学学报, 2022, 43(5): 776-782. |
ZHAO L L, WANG X Y, ZHANG Y, et al. Vehicle target detection based on YOLOv5s fusion SENet[J]. Journal of Graphics, 2022, 43(5): 776-782 (in Chinese). | |
26 | 陈志军, 胡军楠, 冷姚, 等. 基于轻量化网络和注意力机制的智能车快速目标识别方法[J]. 交通运输系统工程与信息, 2022, 22(6): 105-113. |
CHEN Z J, HU J N, LENG Y, et al. Intelligent vehicle target fast recognition based on lightweight network and attention mechanism[J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22(6): 105-113 (in Chinese). | |
27 | 于海洋, 景鹏, 张文涛, 等. 基于残差与注意力机制的道路裂缝检测U-Net改进模型[J]. 计算机工程, 2023, 49(6): 265-273. |
YU H Y, JING P, ZHANG W T, et al. Improved U-Net model for road crack detection based on residual and attention mechanism[J]. Computer Engineering, 2023, 49(6): 265-273 (in Chinese). | |
28 | 佟雨兵, 张其善, 祁云平. 基于PSNR与SSIM联合的图像质量评价模型[J]. 中国图象图形学报, 2006(12): 1758-1763. |
TONG Y B, ZHANG Q S, QI Y P. Image quality assessing by combining PSNR with SSIM[J]. Journal of Image and Graphics, 2006(12): 1758-1763 (in Chinese). | |
29 | 高瑶, 肖卫国, 王力, 等. 关于红外图像目标检测评价仿真研究[J]. 计算机仿真, 2016, 33(11): 33-36, 52. |
GAO Y, XIAO W G, WANG L, et al. Study on simulation and evaluation of infrared image target detection[J]. Computer Integrated Manufacturing Systems, 2016, 33(11): 33-36, 52 (in Chinese). | |
30 | 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. |
31 | BERG A C, FU C Y, SZEGEDY C, et al. SSD: Single Shot MultiBox Detector [DB/OL]. arXiv:. |
32 | REDMON J, FARHADI A. YOLOv3: An incremental improvement[C]∥Conference on Computer Vision and Pattern Recognition. 2018: 1-6 |
33 | BOCHKOVSKIY A, WANG C Y, LIAO H. YOLOv4: Optimal speed and accuracy of object detection [DB/OL]. arXiv:. |
[1] | 王传云, 苏阳, 王琳霖, 王田, 王静静, 高骞. 面向反制无人机集群的多目标连续鲁棒跟踪算法[J]. 航空学报, 2024, 45(7): 329017-329017. |
[2] | 罗旭东, 吴一全, 陈金林. 无人机航拍影像目标检测与语义分割的深度学习方法研究进展[J]. 航空学报, 2024, 45(6): 28822-028822. |
[3] | 杨倩, 王彦哲, 杨迪, 李泽众, 曲巍崴. 基于数据驱动的纤维增强复合材料自动铺放速度预测与规划[J]. 航空学报, 2024, 45(10): 429313-429313. |
[4] | 陈立群, 邹旭, 张磊, 朱颖盼, 王港, 陈金勇. 基于国产商用器件的星载智能目标检测技术[J]. 航空学报, 2023, 44(S2): 728860-728860. |
[5] | 奉志强, 谢志军, 包正伟, 陈科伟. 基于改进YOLOv5的无人机实时密集小目标检测算法[J]. 航空学报, 2023, 44(7): 327106-327106. |
[6] | 冒国韬, 邓天民, 于楠晶. 基于多尺度分割注意力的无人机航拍图像目标检测算法[J]. 航空学报, 2023, 44(5): 326738-326738. |
[7] | 贾宝惠, 姜番, 王玉鑫, 王杜. 基于民机维修文本数据的故障诊断方法[J]. 航空学报, 2023, 44(5): 326598-326598. |
[8] | 姚俞成, 李旭, 徐启敏, 孔栋. 基于特征增强和校准的航拍车辆实例分割方法[J]. 航空学报, 2023, 44(24): 328397-328397. |
[9] | 李晓航, 周建江. 基于自适应记忆长度的多尺度模态融合网络[J]. 航空学报, 2023, 44(22): 628977-628977. |
[10] | 张毅, 张焱, 张宇, 张勇, 刘荻. 基于多级特征增强融合的红外飞机目标检测方法[J]. 航空学报, 2023, 44(22): 629220-629220. |
[11] | 苑玉彬, 吴一全, 赵朗月, 陈金林, 赵其昌. 基于深度学习的无人机航拍视频多目标检测与跟踪研究进展[J]. 航空学报, 2023, 44(18): 28334-028334. |
[12] | 田洁华, 孙迪, 屈峰, 白俊强. 基于CST⁃GAN的翼型参数化方法[J]. 航空学报, 2023, 44(18): 128280-128280. |
[13] | 宋玉存, 葛泉波, 朱军龙, 陆振宇. 基于梯度差自适应学习率优化的改进YOLOX目标检测算法[J]. 航空学报, 2023, 44(14): 327951-327951. |
[14] | 李子豪, 王正平, 贺云涛. 基于自适应协同注意力机制的航拍密集小目标检测算法[J]. 航空学报, 2023, 44(13): 327944-327944. |
[15] | 王强, 吴乐天, 王勇, 王欢, 杨万扣. 基于关键点检测的红外弱小目标检测[J]. 航空学报, 2023, 44(10): 328173-328173. |
阅读次数 | ||||||||||||||||||||||||||||||||||||||||||||||||||
全文 152
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||
摘要 310
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||
版权所有 © 航空学报编辑部
版权所有 © 2011航空学报杂志社
主管单位:中国科学技术协会 主办单位:中国航空学会 北京航空航天大学