边缘修正的多尺度卷积神经网络重建算法

程德强; 蔡迎春; 陈亮亮; 宋玉龙

doi:doi:10.3788/lop55.091003

激光与光电子学进展, 2018, 55 (9): 091003, 网络出版: 2018-09-08

边缘修正的多尺度卷积神经网络重建算法下载： 672次

Multi-Scale Convolutional Neural Network Reconstruction Algorithm Based on Edge Correction

论文大纲

程德强蔡迎春陈亮亮宋玉龙

作者单位

中国矿业大学信息与控制工程学院, 江苏徐州 221008

图像处理超分辨率重建边缘修正多尺度梯度信息 image processing super-resolution reconstruction edge correction multi-scale gradient information

AI 词云图 AI语音精读 AI语音超短摘要

注：本部分内容由 AI 自动生成，请您知悉。

摘要

目前, 基于卷积神经网络的超分辨率重建方法具有参数数量大, 时效性偏低, 边缘细节信息丢失的缺陷。针对该问题, 提出基于边缘修正的多尺度卷积神经网络超分辨率重建算法。首先在训练阶段, 利用低频信息的冗余性设置参数共享层, 将同一组滤波器应用到不同放大倍数的训练网络中, 构建多任务学习框架; 然后在重建阶段, 从样本训练库中学习可以高分辨率图像边缘修正系数, 采用邻域像素差值线性运算将边缘系数与重建的高分辨率图像进行融合, 矫正边缘信息的偏差, 弥补丢失细节; 最后根据随机梯度下降法和反向传播法, 利用梯度不断更新权重参数使网络达到最优化。实验结果表明, 该算法的重建效果较为显著, 边缘锐度较高, 消除了模糊和锯齿现象, 并且通过参数共享大幅减少参数量, 满足实时性的要求。

Abstract

At present, the super-resolution reconstruction methods based on convolutional neural network have the defects of large amount of parameters, low timeliness and loss of edge detail information. In order to solve these problems, we propose a super-resolution reconstruction algorithm of multiscale convolution neural network based on edge correction. Firstly, in the training phase, we set the parameter sharing layer by using the redundancy of low frequency information, In other words, the same set of filters applied to different magnification training networks to build the multi-task learning framework. In the reconstruction phase, the edge correction coefficient of high-resolution image is learned from the sample training library. The neighborhood pixel difference is used to fuse the edge coefficient and the reconstructed high resolution image, and to correct the deviation of the edge information and make up for the missing details. Finally, according to the stochastic gradient descent and back-propagation, we use the gradient to continuously update the weight parameters to make the network reach the maximum optimization. Experimental results show that the proposed algorithm has the significant reconstruction effect, high edge sharpness, elimination of blurring and aliasing, and greatly reduces the amount of parameters through parameter sharing to meet real-time requirements.

参考文献

[1] 肖进胜, 刘恩雨, 朱力, 等. 改进的基于卷积神经网络的图像超分辨率算法［J］. 光学学报, 2017, 37(3): 0318011.

Xiao J S, Liu E Y, Zhu L, et al. Improved image super-resolution algorithm based on convolutional neural network［J］. Acta Optica Sinica, 2017, 37(3): 0318011.

[2] Btz M, Eichenseer A, Seiler J, et al. Hybrid super-resolution combining example-based single-image and interpolation-based multi-image reconstruction approaches［C］∥Proceedings of IEEE International Conference on Image Processing, 2015: 58-62.

[3] 周靖鸿, 周璀, 朱建军, 等. 基于非下采样轮廓波变换遥感影像超分辨重建方法［J］. 光学学报, 2015, 35(1): 0110001.

Zhou J H, Zhou C, Zhu J J, et al. A method of super-resolution reconstruction for remote sensing image based on non-subsampled contourlet transform［J］. Acta Optica Sinica, 2015, 35(1): 0110001.

[4] 练秋生, 张伟. 基于图像块分类稀疏表示的超分辨率重构算法［J］. 电子学报, 2012, 40(5): 920-925．

Lian Q S, Zhang W. Image super-resolution algorithms based on sparse representation of classified image patches［J］. Acta Electronica Sinica, 2012, 40(5): 920-925．

[5] Yang J C, Wright J, Huang T, et al. Image super-resolution as sparse representation of raw image patches［C］∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2008: 1-8.

[6] Yang J C, Wright J, Huang T S, et al. Image super-resolution via sparse representation［J］. IEEE Transactions on Image Processing, 2010, 19(11): 2861-2873.

[7] Timofte R, De V, van Gool L. Anchored neighborhood regression for fast example-based super-resolution［C］∥Proceedings of IEEE International Conference on Computer Vision, 2013: 1920-1927.

[8] Timofte R, de Smet V, van Gool L. A+: Adjusted anchored neighborhood regression for fast super-resolution［C］∥Proceedings of Asian Conference on Computer Vision, 2014: 111-126.

[9] Dong C, Loy C C, He K M, et al. Image super-resolution using deep convolutional networks［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(2): 295-307.

[10] Dong C, Chen C L, Tang X. Accelerating the super-resolution convolutional neural network［C］∥Proceedings of European Conference on Computer Vision, 2016: 391-407.

[11] Kim J, Lee J K, Lee K M. Deeply-recursive convolutional network for image super-resolution［C］∥Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2016: 1637-1645.

[12] 孙超, 吕俊伟, 李健伟, 等. 基于去卷积的快速图像超分辨率方法［J］. 光学学报, 2017, 37(12): 1210004.

Sun C, Lü J W, Li J W, et al. Method of rapid image super-resolution based on deconvolution［J］. Acta Optica Sinica, 2017, 37(12): 1210004.

[13] Chang H, Yeung D Y, Xiong Y M. Super-resolution through neighbor embedding［C］∥Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004, 1: I.

[14] Aharon M, Elad M, Bruckstein A. rmK-SVD: An algorithm for designing overcomplete dictionaries for sparse representation［J］. IEEE Transactions on Signal Processing, 2006, 54(11): 4311-4322.

[15] Jia Y Q, Shelhamer E, Donahue J, et al. Caffe: Convolutional architecture for fast feature embedding［C］∥Proceedings of the 22nd ACM International Conference on Multimedia, 2014: 675-678.

[16] Bevilacqua M, Roumy A, Guillemot C, et al. Low-complexity single image super-resolution based on nonnegative neighbor embedding［EB/OL］. ［2018-02-05］http:∥eprints.imtlucca.it/2412/1/Bevilacqua_2012.pdf.

[17] Zeyde R, Elad M, Protter M. On single image scale-up using sparse-representations［C］∥Proceedings of International Conference on Curves and Surfaces, 2010: 711-730.

程德强, 蔡迎春, 陈亮亮, 宋玉龙. 边缘修正的多尺度卷积神经网络重建算法[J]. 激光与光电子学进展, 2018, 55(9): 091003. Cheng Deqiang, Cai Yingchun, Chen Liangliang, Song Yulong. Multi-Scale Convolutional Neural Network Reconstruction Algorithm Based on Edge Correction[J]. Laser & Optoelectronics Progress, 2018, 55(9): 091003.

边缘修正的多尺度卷积神经网络重建算法下载： 672次

关于本站 Cookie 的使用提示

全站搜索

边缘修正的多尺度卷积神经网络重建算法 下载： 672次

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索

边缘修正的多尺度卷积神经网络重建算法下载： 672次