计算机科学 ›› 2022, Vol. 49 ›› Issue (3): 185-191.doi: 10.11896/jsjkx.210100234
周颖1,2, 常明新1, 叶红1, 张燕1
ZHOU Ying1,2, CHANG Ming-xin1, YE Hong1, ZHANG Yan1
摘要: 针对太阳能电池板隐裂缺陷在进行光学检测时存在的特征不明显问题,以及小样本导致的训练不充分问题,提出了基于元迁移的太阳能电池板缺陷图像超分辨率重建方法,采用联合训练方法,利用内部图像和外部大规模图像信息分别作为不同阶段的训练数据。首先将引入的大量数据用于模型的初步训练,学习外部大规模数据的公共特征,然后通过元学习模型MAML进行多任务训练,为快速适应小样本无监督任务寻找一个适合图像内部学习的初始参数,提高模型的泛化能力,最后将预训练参数迁移至改进的ZSSR中进行自监督学习。在DIV2K、Set5、BSD100和太阳能电池板电致发光成像数据集上进行训练,实验结果表明,与传统的CARN,RCAN,IKC,ZSSR方法相比,该方法具有更高的峰值信噪比,最高达到36.66,参数量更小,相比ZSSR降低了70 000,图像重建时间更短,相比CARN降低了0.51 s,具有更好的重建效果,更高的重建效率。
中图分类号:
[1]JIANG J,LI M,ZHU M Q,et al.Crack detection of outdoorsolar panel based on convolutional neural network[J].Journal of Yangzhou University (Natural Science Edition),2020,23(1):49-53. [2]CHEN F M,CHENG X Y,YAO Z F.Design of Solar PanelDefect Detection Model Based on Deep Learning[J].Wireless Interconnection Technology,2019,16(23):56-61. [3]GUO B S,ZHUANG J C,ZHANG Q,et al.Color difference detection of polysilicon wafers based on multi-component convolutional neural network[J].China Mechanical Engineering,2021,23(18):1-10. [4]SHOCHER A,COHEN N,IRANI M.Zero-Shot Super-Resolu-tion Using Deep Internal Learning[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Salt Lake City,UT,2018:3118-3126. [5]MASTAN I D,RAMAN S.DCIL:Deep Contextual InternalLearning for Image Restoration and Image Retargeting[C]//2020 IEEE Winter Conference on Applications of Computer Vision (WACV).Snowmass Village,CO,USA,2020:2355-2364. [6]JUNZHI Y.Zero-Shot Super Resolution for Satellite RemoteSensing Images[C]//2019 IEEE International Conference on Signal,Information and Data Processing (ICSIDP).IEEE,2019:245-255. [7] FELUI-FABÁ J,FAN Y,YING L.Meta-learning pseudo-diffe-rential operators with deep neural networks[J].Journal of Computational Physics,2020,5(23):408-415. [8]PURI R.Few Shot Learning For Point Cloud Data Using Model Agnostic Meta Learning[J].2020 IEEE International Confe-rence on Image Processing (ICIP),2020,13(5):1906-1910. [9]DEMERTZIS K,ILIADIS L.GeoAI:A Model-Agnostic Meta-Ensemble Zero-Shot Learning Method for Hyperspectral Image Analysis and Classification[J].Algorithms,2020,13(3):61-68. [10]XU Z,CHEN X,TANG W,et al.Meta Weight Learning viaModel-Agnostic Meta-Learning[J].Neurocomputing,2020,432(7587):342-356. [11]EIRIKUR A,RADU T.Ntire 2017 challenge on single image super-resolution:Dataset and study[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops.2017:126-135. [12]LI Y M.Study on Bicubic Interpolation Algorithm for Images[D].Lanzhou:Lanzhou University,2020. [13]WANG J L,LIU L Q,ZHANG C M.Application of image processing in photovoltaic local shadow[J].Journal of Solar Energy,2020,41(2):284-289. [14]WU G,ZHAO L,WANG W,et al.PRED:A Parallel Network for Handling Multiple Degradations via Single Model in Single Image Super-Resolution[C]//2019 IEEE International Confe-rence on Image Processing (ICIP).Taipei,Taiwan,2019:5-10. [15]CHOLLET F.Xception:Deep Learning with Depthwise Separable Convolutions[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE,2017. [16]TAHIR S,JALAL A,KIM K.Wearable Inertial Sensors forDaily Activity Analysis Based on Adam Optimization and the Maximum Entropy Markov Model[J].Entropy (Basel,Swit-zerland),2020,22(5):1250-1258. [17]LAN R,SUN L,LIU Z,et al.Cascading and Enhanced Residual Networks for Accurate Single-Image Super-Resolution[J].IEEE Transactions on Cybernetics,2020,43(99):1-11. [18]MARCO B,ALINE R,CHRISTINE G,et al.Low-complexity single-image super-resolution based on nonnegative neighbor embedding[J].BMVC,2012,54(9):104-110. [19]DAVID M,CHARLESS F,DORON T,et al.A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics[J].Computer Vision,2001,19(6):416-423. [20]MEHTA J H.Relation between Entropy and Peak Signal toNoise Ratio in Prediction Error Expansion Considering Region of Interest[J].International Conference for Convergence in Technology (I2CT),2019,69(9):1-4. [21]ELACHKAR I,OUZIF H,LABRIJI H.Structural SimilarityMeasure of Users Profiles Based on A Weighted Bipartite Graphs[J].ISPRS International Archives of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2020,66(5):1-9. [22]LAN R,SUN L,LIU Z,et al.Cascading and Enhanced Residual Networks for Accurate Single-Image Super-Resolution[J].IEEE Transactions on Cybernetics,2020,66(99):146-155. [23]QI Y,GU J,LI W,et al.Pulmonary nodule image super-resolution using multi-scale deep residual channel attention network with joint optimization[J].The Journal of Supercomputing,2019,76(3):1508-1515. [24]DENIZ K,HAKAN A.A Novel Short-Term Photovoltaic Power Forecasting Approach based on Deep Convolutional Neural Network[J].International Journal of Green Energy,2021,18(5):34-39. [25]LEE K,LEE E,CHOI B,et al.Automatic Pharyngeal PhaseRecognition in Untrimmed Videofluoroscopic Swallowing Study Using Transfer Learning with Deep Convolutional Neural Networks[J].Diagnostics,2021,11(2):622-630. [26]GU J,LU H,ZUO W,et al.Blind Super-Resolution with Iterative Kernel Correction[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).2019:1245-1250. |
[1] | 周乐员, 张剑华, 袁甜甜, 陈胜勇. 多层注意力机制融合的序列到序列中国连续手语识别和翻译 Sequence-to-Sequence Chinese Continuous Sign Language Recognition and Translation with Multi- layer Attention Mechanism Fusion 计算机科学, 2022, 49(9): 155-161. https://doi.org/10.11896/jsjkx.210800026 |
[2] | 李宗民, 张玉鹏, 刘玉杰, 李华. 基于可变形图卷积的点云表征学习 Deformable Graph Convolutional Networks Based Point Cloud Representation Learning 计算机科学, 2022, 49(8): 273-278. https://doi.org/10.11896/jsjkx.210900023 |
[3] | 陈泳全, 姜瑛. 基于卷积神经网络的APP用户行为分析方法 Analysis Method of APP User Behavior Based on Convolutional Neural Network 计算机科学, 2022, 49(8): 78-85. https://doi.org/10.11896/jsjkx.210700121 |
[4] | 朱承璋, 黄嘉儿, 肖亚龙, 王晗, 邹北骥. 基于注意力机制的医学影像深度哈希检索算法 Deep Hash Retrieval Algorithm for Medical Images Based on Attention Mechanism 计算机科学, 2022, 49(8): 113-119. https://doi.org/10.11896/jsjkx.210700153 |
[5] | 檀莹莹, 王俊丽, 张超波. 基于图卷积神经网络的文本分类方法研究综述 Review of Text Classification Methods Based on Graph Convolutional Network 计算机科学, 2022, 49(8): 205-216. https://doi.org/10.11896/jsjkx.210800064 |
[6] | 齐秀秀, 王佳昊, 李文雄, 周帆. 基于概率元学习的矩阵补全预测融合算法 Fusion Algorithm for Matrix Completion Prediction Based on Probabilistic Meta-learning 计算机科学, 2022, 49(7): 18-24. https://doi.org/10.11896/jsjkx.210600126 |
[7] | 张颖涛, 张杰, 张睿, 张文强. 全局信息引导的真实图像风格迁移 Photorealistic Style Transfer Guided by Global Information 计算机科学, 2022, 49(7): 100-105. https://doi.org/10.11896/jsjkx.210600036 |
[8] | 戴朝霞, 李锦欣, 张向东, 徐旭, 梅林, 张亮. 基于DNGAN的磁共振图像超分辨率重建算法 Super-resolution Reconstruction of MRI Based on DNGAN 计算机科学, 2022, 49(7): 113-119. https://doi.org/10.11896/jsjkx.210600105 |
[9] | 刘月红, 牛少华, 神显豪. 基于卷积神经网络的虚拟现实视频帧内预测编码 Virtual Reality Video Intraframe Prediction Coding Based on Convolutional Neural Network 计算机科学, 2022, 49(7): 127-131. https://doi.org/10.11896/jsjkx.211100179 |
[10] | 徐鸣珂, 张帆. Head Fusion:一种提高语音情绪识别的准确性和鲁棒性的方法 Head Fusion:A Method to Improve Accuracy and Robustness of Speech Emotion Recognition 计算机科学, 2022, 49(7): 132-141. https://doi.org/10.11896/jsjkx.210100085 |
[11] | 金方焱, 王秀利. 融合RACNN和BiLSTM的金融领域事件隐式因果关系抽取 Implicit Causality Extraction of Financial Events Integrating RACNN and BiLSTM 计算机科学, 2022, 49(7): 179-186. https://doi.org/10.11896/jsjkx.210500190 |
[12] | 王杉, 徐楚怡, 师春香, 张瑛. 基于CNN-LSTM的卫星云图云分类方法研究 Study on Cloud Classification Method of Satellite Cloud Images Based on CNN-LSTM 计算机科学, 2022, 49(6A): 675-679. https://doi.org/10.11896/jsjkx.210300177 |
[13] | 杨玥, 冯涛, 梁虹, 杨扬. 融合交叉注意力机制的图像任意风格迁移 Image Arbitrary Style Transfer via Criss-cross Attention 计算机科学, 2022, 49(6A): 345-352. https://doi.org/10.11896/jsjkx.210700236 |
[14] | 杨健楠, 张帆. 一种结合双注意力机制和层次网络结构的细碎农作物分类方法 Classification Method for Small Crops Combining Dual Attention Mechanisms and Hierarchical Network Structure 计算机科学, 2022, 49(6A): 353-357. https://doi.org/10.11896/jsjkx.210200169 |
[15] | 张嘉淏, 刘峰, 齐佳音. 一种基于Bottleneck Transformer的轻量级微表情识别架构 Lightweight Micro-expression Recognition Architecture Based on Bottleneck Transformer 计算机科学, 2022, 49(6A): 370-377. https://doi.org/10.11896/jsjkx.210500023 |
|