基于权重量化与信息压缩的车载图像超分辨率重建

doi:10.11772/j.issn.1001-9081.2019050804

计算机应用 ›› 2019, Vol. 39 ›› Issue (12): 3644-3649.DOI: 10.11772/j.issn.1001-9081.2019050804

• 虚拟现实与多媒体计算 • 上一篇下一篇

基于权重量化与信息压缩的车载图像超分辨率重建

许德智, 孙季丰, 罗莎莎

华南理工大学电子与信息学院, 广州 510641

收稿日期:2019-05-13 修回日期:2019-07-30 发布日期:2019-10-23 出版日期:2019-12-10
作者简介:许德智(1995-),男,湖北荆州人,硕士研究生,主要研究方向:机器学习、计算机视觉;孙季丰(1962-),男,广东广州人,教授,博士生导师,博士,主要研究方向:机器学习、模式识别、计算机视觉;罗莎莎(1994-),女,江西吉安人,硕士研究生,主要研究方向:计算机视觉、数据挖掘。
基金资助:
广东省科技计划项目（x2dxB216005）。

Vehicle-based image super-resolution reconstruction based on weight quantification and information compression

XU Dezhi, SUN Jifeng, LUO Shasha

School of Electronic and Information Engineering, South China University of Technology, Guangzhou Guangdong 510641, China

Received:2019-05-13 Revised:2019-07-30 Online:2019-10-23 Published:2019-12-10
Contact: 许德智
Supported by:
This work is partially supported by the Science and Technology Research Project of Guangdong (x2dxB216005).

摘要/Abstract

摘要： 针对智能驾驶领域中需要在内存受限的情况下得到高质量的超分辨率图像的问题，提出一种基于权重八位二进制量化的车载图像超分辨率重建算法。首先，基于八位二进制量化卷积设计信息压缩模块，减少内部冗余，增强网络内信息流动，提高重建速率；然后，整个网络由一个特征提取模块、多个堆叠的信息压缩模块和一个图像重建模块构成，并利用插值后超分辨率空间的信息与低分辨率空间重建后的图像融合，在不增加模型复杂度的基础上，提高网络表达能力；最后，算法中整个网络结构基于对抗生成网络（GAN）框架进行训练，使得到的图片有更好主观视觉效果。实验结果表明，所提算法的车载图像重建结果的峰值信噪比（PSNR）比基于GAN的超分辨率重建（SRGAN）算法提高了0.22 dB，同时其生成模型大小缩小为LapSRN的39%，重建速度提高为LapSRN的7.57倍。

关键词: 超分辨率重建, 车载图像, 八位二进制权重量化, 对抗生成网络, 信息压缩模块

Abstract: For the intelligent driving field, it is necessary to obtain high-quality super-resolution images under the condition of limited memory. Therefore, a vehicle-based image super-resolution reconstruction algorithm based on weighted eight-bit binary quantization was proposed. Firstly, the information compression module was designed based on the eight-bit binary quantization convolution, reducing the internal redundancy, enhancing the information flow in the network, and improving the reconstruction rate. Then, the whole network was composed of a feature extraction module, a plurality of stacked information compression modules and an image reconstruction module, and the information of the interpolated super-resolution space was fused with the image reconstructed by the low-resolution space, improving the network expression ability without increasing the complexity of the model. Finally, the entire network structure in the algorithm was trained based on the Generative Adversarial Network (GAN) framework, making the image have better subjective visual effect. The experimental results show that, the Peak Signal-to-Noise Ratio (PSNR) of the proposed algorithm for the reconstructed vehicle-based image is 0.22 dB higher than that of Super-Resolution using GAN (SRGAN), its generated model size is reduced to 39% of that of the Laplacian pyramid Networks for fast and accurate Super-Resolution (LapSRN), and the reconstruction speed is improved to 7.57 times of that of LapSRN.

Key words: super-resolution reconstruction, vehicle-based image, eight-bit binary weight quantification, Generative Adversarial Networks (GAN), information compression module

中图分类号:

TP391.4

许德智, 孙季丰, 罗莎莎. 基于权重量化与信息压缩的车载图像超分辨率重建[J]. 计算机应用, 2019, 39(12): 3644-3649.

XU Dezhi, SUN Jifeng, LUO Shasha. Vehicle-based image super-resolution reconstruction based on weight quantification and information compression[J]. Journal of Computer Applications, 2019, 39(12): 3644-3649.

参考文献

[1] EFRAT N, GLASNER D, APARTSIN A, et al. Accurate blur models vs. image priors in single image super resolution[C]//Proceedings of the 2013 IEEE International Conference on Computer Vision. Piscataway:IEEE, 2013:2832-2839.
[2] 张宁,王永成,张欣,等.基于深度学习的单张图片超分辨率重构研究进展[J/OL].自动化学报[2019-07-07].http://kns.cnki.net/kcms/detail/detail.aspx?doi=10.16383/j.aas.c190031. (ZHANG N, WANG Y C, ZHANG X, et al. A review of single image super-resolution based on deep learning[J/OL]. Acta Automatica Sinica[2019-07-07]. http://kns.cnki.net/kcms/detail/detail.aspx?doi=10.16383/j.aas.c190031.)
[3] DONG C, LOY C C, HE K, et al. Learning a deep convolutional network for image super-resolution[C]//Proceedings of the 2014 European Conference on Computer Vision, LNCS 8692. Cham:Springer, 2014:184-199.
[4] KIM J, LEE J K, LEE K M. Accurate image super-resolution using very deep convolutional networks[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2016:1646-1654.
[5] MAO X, SHEN C, YANG Y. Image restoration using very deep convolutional encoder-decoder networks with symmetric skip connections[C]//Proceedings of the 2016 Annual Conference on Neural Information Processing Systems. La Jolla:Neural Information Processing Systems Foundation, 2016:432-449.
[6] LAI W, HUANG J, AHUJA A, et al. Deep Laplacian pyramid networks for fast and accurate super-resolution[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2017:5835-5843.
[7] TAI Y, YANG J, LIU Y. Image super-resolution via deep recursive residual network[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2017:2790-2798.
[8] TONG T, LI G, LIU X, et al. Image super-resolution using dense skip connections[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2017:4809-4817.
[9] COURBARIAUX M, BENGIO Y, DAVID J P. BinaryConnect:training deep neural networks with binary weights during propagations[C]//Proceedings of the 2015 Annual Conference on Neural Information Processing Systems. La Jolla:Neural Information Processing Systems Foundation, 2015:3123-3131.
[10] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2016:770-778.
[11] 袁昆鹏,席志红.基于深度跳跃级联的图像超分辨率重建[J].光学学报,2019,39(7):0715003-1-0715003-10.(YUAN K P, XI Z H. Image super resolution based on depth jumping cascade[J]. Acta Optica Sinica, 2019, 39(7):0715003-1-0715003-10.)
[12] OODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets[C]//Proceedings of the 2014 Annual Conference on Neural Information Processing Systems. La Jolla:Neural Information Processing Systems Foundation, 2014:2672-2680.
[13] OHNSON J, ALAHI A, LI F. Perceptual losses for real-time style transfer and super-resolution[C]//Proceedings of the 2016 European Conference on Computer Vision, LNCS 9906. Cham:Springer, 2016:694-711.
[14] RASTEGARI M, ORDONEZ V, REDMON J, et al. XNOR-Net:ImageNet classification using binary convolutional neural networks[C]//Proceedings of the 2016 European Conference on Computer Vision, LNCS 9908. Cham:Springer, 2016:525-542.
[15] LIM B, SON S, KIM H, et al. Enhanced deep residual networks for single image super-resolution[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2017:1132-1140.
[16] IOFFE S, SZEGEDY S. Batch normalization:accelerating deep network training by reducing internal covariate shift[C]//Proceedings of the 201532nd International Conference on Machine Learning. New York:JMLR, 2015:448-456.
[17] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL].[2019-03-20]. https://arxiv.org/pdf/1409.1556.pdf.
[18] LEDIG C, THEIS L, HUSZÁR F, et al. Photo-realistic single image super-resolution using a generative adversarial network[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2017:105-114.
[19] WANG Z, BOVIK A C, SHEIKH H R, et al. Image quality assessment:from error visibility to structural similarity[J]. IEEE Transactions on Image Processing, 2004, 13(4):600-612.
[20] 刘杰平,杨业长,韦岗.结合暗通道先验的单幅图像快速去雾算法[J].华南理工大学学报(自然科学版),2018,46(3):86-91.(LIU J P, YANG Y Z, WEI G. A fast single image dehazing algorithm based on dark channel prior[J]. Journal of South China University of Technology (Natural Science Edition), 2018, 46(3):86-91.)
[21] LIN T Y, GOYAL P, GIRSHICK R. Focal loss for dense object detection[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway:IEEE, 2017:2999-3007

基于权重量化与信息压缩的车载图像超分辨率重建

Vehicle-based image super-resolution reconstruction based on weight quantification and information compression

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	牛康力, 谌雨章, 沈君凤, 曾张帆, 潘永才, 王绎冲. 基于深度学习的双通道夜视图像复原方法[J]. 计算机应用, 2021, 41(6): 1775-1784.
[2]	梁敏, 王昊榕, 张瑶, 李杰. 基于加速残差网络的图像超分辨率重建方法[J]. 计算机应用, 2021, 41(5): 1438-1444.
[3]	段友祥, 张含笑, 孙歧峰, 孙友凯. 基于拉普拉斯金字塔生成对抗网络的图像超分辨率重建算法[J]. 计算机应用, 2021, 41(4): 1020-1026.
[4]	王永金, 左羽, 吴恋, 崔忠伟, 赵晨洁. 基于注意力机制的图像超分辨率重建[J]. 计算机应用, 2021, 41(3): 845-850.
[5]	林静, 黄玉清, 李磊民. 基于球形矩匹配与特征判别的图像超分辨率重建[J]. 计算机应用, 2020, 40(8): 2345-2350.
[6]	张豪, 张强, 邵思羽, 丁海斌. 深度学习在单图像三维模型重建的应用[J]. 计算机应用, 2020, 40(8): 2351-2357.
[7]	胡雪影, 郭海儒, 朱蓉. 基于混合深度卷积网络的图像超分辨率重建[J]. 计算机应用, 2020, 40(7): 2069-2076.
[8]	姚鲁, 宋慧慧, 张开华. 混合阶通道注意力网络的单图像超分辨率重建[J]. 计算机应用, 2020, 40(10): 3048-3053.
[9]	欧阳宁, 韦羽, 林乐平. 结合感知边缘约束与多尺度融合网络的图像超分辨率重建方法[J]. 计算机应用, 2020, 40(10): 3041-3047.
[10]	陈龙彪, 谌雨章, 王晓晨, 邹鹏, 胡学敏. 基于深度学习的水下图像超分辨率重建方法[J]. 计算机应用, 2019, 39(9): 2738-2743.
[11]	刘月峰, 杨涵晰, 蔡爽, 张晨荣. 基于改进卷积神经网络的单幅图像超分辨率重建方法[J]. 计算机应用, 2019, 39(5): 1440-1447.
[12]	莫赞, 盖彦蓉, 樊冠龙. 基于GAN-AdaBoost-DT不平衡分类算法的信用卡欺诈分类[J]. 计算机应用, 2019, 39(2): 618-622.
[13]	孙毅堂, 宋慧慧, 张开华, 严飞. 基于极深卷积神经网络的人脸超分辨率重建算法[J]. 计算机应用, 2018, 38(4): 1141-1145.
[14]	彭政, 陈东方, 王晓峰. 正则化超分辨率重建过程的自适应阈值去噪[J]. 计算机应用, 2017, 37(7): 2084-2088.
[15]	陈伟业, 孙权森. 结合压缩感知与非局部信息的图像超分辨率重建[J]. 计算机应用, 2016, 36(9): 2570-2575.