| 68 | 0 | 43 |
| 下载次数 | 被引频次 | 阅读次数 |
提出了一种基于退化模型的超分辨率重建方法，旨在解决航空精密零件图像采集过程中因失焦模糊导致的图像质量下降问题。首先，设计了一种真实的图像退化模型，模型包括失焦模糊、下采样和噪声等多种因素，通过扩展退化空间来模拟实际应用中的多样化退化情况。基于该模型生成合成数据，训练了超分辨率重建模型，以提高精密零件失焦图像的质量。为了提取图像特征，提出了一种新的基于Swin Transformer的图像超分辨率网络，堆叠了多个残差Swin Transformer通道注意力(residual swin transformer channel attention, RST-CA)模块，为每两个Swin Transformer模块引入通道注意力模块。实验结果表明，在恢复航空精密零件失焦图像细节方面优于传统的退化模型，特别是在零件纹理和微圆孔细节的重建上，显示出明显的清晰度提升。在客观评价指标上，相较于传统退化模型与经典超分方法(如EDSR、SRGAN、SwinIR等),该方法在PSNR和SSIM指标上平均提升约3.2 dB与0.05,尤其在微小孔结构的细节恢复方面表现更为优越。该研究为航空精密零件的失焦恢复和超分辨率重建提供了一种有效的技术路径，具有较大的应用前景和工业应用价值。
Abstract：A super-resolution reconstruction method based on a degradation model is proposed, aiming at solving the problem of image quality degradation due to out-of-focus blurring during image acquisition of aerospace precision parts.First, a realistic image degradation model is designed, which includes various factors such as out-of-focus blur, downsampling and noise, and simulates diverse degradation situations in real applications by extending the degradation space.Based on this model, synthetic data are generated and a super-resolution reconstruction model is trained to improve the quality of out-of-focus images of precision parts.In order to extract image features, a new swin transformer-based image super-resolution network is proposed, stacking multiple residual swin transformer channel attention(RST-CA) modules for every two swin transformer modules to introduce the channel attention module.The experimental results show that it outperforms the traditional degradation model in recovering the out-of-focus image details of aerospace precision parts, especially in the reconstruction of part texture and microcircular hole details, which show obvious clarity improvement.In terms of objective evaluation metrics, compared with the traditional degradation model and classical super-segmentation methods(e.g.,EDSR,SRGAN,SwinIR,etc.),the method shows an average enhancement of about 3.2 dB and 0.05 in the PSNR and SSIM metrics, and is especially superior in the detail recovery of micro-miniature hole structures.This study provides an effective technical path for out-of-focus recovery and super-resolution reconstruction of aerospace precision parts, which has a large application prospect and industrial application value.
[1] GISARIO A,KAZARIAN M,MARTINA F,et al.Metal additive manufacturing in the commercial aviation industry:a review[J].Journal of Manufacturing Systems,2019,53:124-149.
[2] KHAN K,SOHAIB M,RASHID A,et al.Recent trends and challenges in predictive maintenance of aircraft’s engine and hydraulic system[J].Journal of the Brazilian Society of Mechanical Sciences and Engineering,2021,43:1-17.
[3] 李文，周海蔚.基于GA-BP神经网络的工业CT缺陷检测[J].计量与测试技术，2024,51(3):96-99.
[4] DENG L,LIU G,ZHANG Y L.A review of machine vision applications in aerospace manufacturing quality inspection[C]//2024 4th International Conference on Computer,Control and Robotics (ICCCR).New York:IEEE,2024:31-39.
[5] 蒋舜耀，彭珍瑞，殷红.基于计算机视觉的桥梁拉索索力测量[J].兰州交通大学学报，2024,43(6):155-164.
[6] 李相格.基于改进LOG算子与Otsu算法结合的物体表面图像残损检测方法[J].兰州交通大学学报，2024,43(1):59-63.
[7] 王珮，朱宇，闫庆森，等.真实场景图像去模糊：挑战与展望[J].中国图象图形学报，2024,29(12):3501-3528.
[8] EL ABBADI N K,ABDULKHALEQ A H,AL HASSANI S A.A survey on blind de-blurring of digital image[J].Iraqi Journal of Science,2022,63(1):338-352.
[9] LIU Y Q,DU X,SHEN H L,et al.Estimating generalized gaussian blur kernels for out-of-focus image deblurring[J].IEEE Transactions on Circuits and Systems for Video Technology,2020,31(3):829-843.
[10] 侯礼杰，沈寅松，刘晓晨，等.三维高斯溅射超分辨视觉场景构建算法[J].中国测试，2024,50(9):13-20.
[11] ZHANG K,ZUO W M,GU S H,et al.Learning deep CNN denoiser prior for image restoration[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.New York:IEEE,2017:3929-3938.
[12] VASWANI A,SHAZEER N,PARMAR N,et al.Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems.New York:Curran Associates Inc.,2017:6000-6010.
[13] LIU A R,LIU Y H,GU J J,et al.Blind image super-resolution:a survey and beyond[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2022,45(5):5461-5480.
[14] CHEN H T,WANG Y H,GUO T Y,et al.Pre-trained image processing transformer[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.New York:IEEE,2021:12299-12310.
[15] SU H,LI Y,XU Y F,et al.A review of deep-learning-based super-resolution:from methods to applications[J].Pattern Recognition,2024:110935.
[16] ZHANG K,GOOL L V,TIMOFTE R.Deep unfolding network for image super-resolution[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.New York:IEEE,2020:3217-3226.
[17] KAUR R,KARMAKAR G,XIA F,et al.Deep learning:survey of environmental and camera impacts on internet of things images[J].Artificial Intelligence Review,2023,56(9):9605-9638.
[18] ZHANG K,LIANG J Y,VAN GOOL L V,et al.Designing a practical degradation model for deep blind image super-resolution[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision,New York:IEEE,2021:4791-4800.
[19] LIANG J Y,CAO J Z,SUN G L,et al.Swinir:image restoration using swin transformer[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision,New York:IEEE,2021:1833-1844.
[20] WANG Q L,WU B G,ZHU P F,et al.ECA-Net:efficient channel attention for deep convolutional neural networks[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR),New York:IEEE,2020:11531-11539.
[21] LIU Z,LIN Y T,CAO Y,et al.Swin transformer:hierarchical vision transformer using shifted windows[C]//2021 IEEE/CVF International Conference on Computer Vision,New York:IEEE,2021:9992-10002.
[22] HU B,LI L D,WU J J,et al.Subjective and objective quality assessment for image restoration:a critical survey[J].Signal Processing Image Communication,2020,85:115839.
[23] SARA U,AKTER M,UDDIN M S.Image quality assessment through FSIM,SSIM,MSE and PSNR——a comparative study[J].Journal of Computer and Communications,2019,7(3):8-18.
[24] SETIADI D R I M.PSNR vs SSIM:imperceptibility quality assessment for image steganography[J].Multimedia Tools and Applications,2021,80:8423-8444.
[25] REZNIK Y.Another look at SSIM image quality metric[J].Electronic Imaging,2023,35:1-7.
基本信息:
DOI：
中图分类号:TP391.41;V261
引用信息:
[1]杨海马,肖啸天,刘瑾等.基于改进退化网络的航空零件超分辨率重建方法[J].兰州交通大学学报,2025,44(03):1-8+28.
基金信息:
国家自然科学基金(U1831133); 中国科学院空间主动光电技术重点实验室开放基金(2021ZDKF4); 上海市科技创新行动计划(22dz1201300); 上海市浦江人才计划(23PJD067)