DOI: 10.3724/SP.J.1249.2018.04405

Journal of Shenzhen University Science and Engineering (深圳大学学报理工版) 2018/35:4 PP.405-412

An BP neural network image restoration method based on differential evolution optimization

Aiming at the problems of slow convergence and local minimum of the back-propagating (BP) neural network, an image restoration method based on random scaling-differential evolution (RSDE) is proposed. In the proposed method, the noise-free images are blurred by the Gaussian noise. The blurred images and noise-free images are set to the training pairs, which are used to train and optimize the proposed RSDE-BP method. The optimized BP neural network is utilized to restore the testing images and remove the noises. The experimental results show that the convergence rate of the RSDE-BP algorithm is faster and the number of iterations is less than the traditional BP method, the PSO-BP method and the DE-BP method. In addition, the peak signal to noise ratio (PSNR) and the structural similarity (SSIM) are improved. Compared with the adaptive total variation (ATV) image deblurring method and the blurred image restoration method based on the second-order total generalized variation (TGV) regularization, the RSDE-BP method can effectively restore the images polluted by the noise and blur, while preserving the image texture and details more effectively.

Key words:image processing,blurred image,image restoration,BP neural network,differential evolution,peak signal to noise ratio,structural similarity

ReleaseDate:2018-07-26 10:51:03

[1] GONZALEZ R C, WOODS R E. 数字图像处理[M]. 阮秋琦,阮宇智,译. 3版.北京:电子工业出版社, 2017:197-201. GONZALEZ R C, WOODS R E. Digital image processing[M]. RUAN Qiuqi, RUAN Yuzhi, trans.. 3rd ed.. Beijing:Publishing House of Electronics Industry, 2017:197-201.(in Chinese)

[2] GUO Zhan,NILSSON P. Algorithm and implementation of the K-best sphere decoding for MIMO detection[J].IEEE Journal on Selected Areas in Communications,2006, 24(3):491-503.

[3] REN Wenqi, CAO Xiaochun, PAN Jinshan,et al.Image deblurring via enhanced low rank prior[J].IEEE Transactions on Image Processing,2016,25(7):3426-3437.

[4] PAN Jinshan, LIU Risheng, SU Zhixun, et al. Kernel estimation from salient structure for robust motion deblurring[J]. Signal Processing:Image Communication, 2013,28(9):1156-1170.

[5] YAN Ruomei, SHAO Ling. Blind image blur estimation via deep learning[J].IEEE Transactions on Image Processing, 2016, 25(4):1910-1921.

[6] PAN Jinshan, HU Zhe, SU Zhixun,et al. L0-regularized intensity and gradient prior for deblurring text images and beyond[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(2):342-355.

[7] 宋发兴,高留洋,刘东升,等.基于粒子群优化的BP神经网络图像复原方法[J].计算机技术与发展,2014, 24(6):149-152. SONG Faxing, GAO Liuyang, LIU Dongsheng,et al. A method of image restoration based on particle swarm optimization for BP neural network[J].Computer Technology and Development,2014,24(6):149-152.(in Chinese)

[8] RUMELLHART D E, HINTON G E, WILLIAMS R J. Learning representation by back-propagating errors[J]. Nature, 1986, 323:149-154.

[9] MALLIPEDDI R, SUGANTHAN P N, PAN Q K, et al. Differential evolution algorithm with ensemble of parameters and mutation strategies[J].Applied Soft Computing, 2011, 11(2):1679-1696.

[10] 刘琛,林盈,胡晓敏.差分演化算法各种更新策略的对比分析[J].计算机科学与探索,2013, 7(11):983-993. LIU Chen, LIN Ying, HU Xiaomin. Analyses and comparisons of different update strategies for differential evolution[J].Journal of Frontiers of Computer Science & Technology,2013,7(11):983-993.(in Chinese)

[11] SANCHEZ F C, AYALA H V H, COELHO L D S, et al. Multi-objective differential evolution algorithm for underwater image restoration[C]//IEEE Congress on Evolutionary Computation. Sendai, Japan:Institute of Electrical and Electronics Engineers Inc., 2015, 243-250.

[12] MAGOULAS G D, PLAGIANAKOS V P, VRAHATIS M N. Neural network-based colonoscopic diagnosis using on-line learning and differential evolution[J].Applied Soft Computing,2004,4(4):369-379.

[13] MA Lishan, YUAN Dekui, TAO Jianhua, et al. Prediction of ground water level based on DE-BP neutral network[C]//International Conference on Environmental Science and Information Application Technology. Wuhan, China:IEEE Computer Society, 2009, 258-261.

[14] DAI Qianwei, JIANG Feibo, DONG Li. Nonlinear inversion for electrical resistivity tomography based on chaotic DE-BP algorithm[J].Journal of Central South University,2014,21(5):2018-2025.

[15] GANG Shang, LUO Zhong, CHEN Liyao. Discussion about BP neural network structure and choice of samples training parameter[J].Journal of Wuhan University of Technology, 1997, 45(2):121-132.

[16] OLIVERIRA J P, BIOUCAS D J M, FIGUEIREDO M A T. Adaptive total variation image deblurring:a majorization-minimization approach[J].Signal Processing, 2009, 89(9):1683-1693.

[17] 任福全,邱天爽.基于二阶广义全变差正则项的模糊图像恢复算法[J].自动化学报,2015, 41(6):1166-1172. REN Fuquan, QIU Tianshuang.Blurred image restoration method based on second-order total generalized variation regularization[J].Acta Automatica Sinica,2015,41(6):1166-1172. (in Chinese)

[18] KNOLL F, BREDIES K, POCK T,et al.Second order total generalized variation (TGV) for MRI[J].Magnetic Resonance in Medicine,2011,65(2):480-491.