基于Fourier神经网络的图像复原算法  

Image restoration algorithm based on Fourier neural network

作　　者：田启川[1] 田茂新[1] 

机构地区：[1]太原科技大学电子信息工程学院,太原030024

出　　处：《计算机应用研究》2012年第3期1143-1145,共3页Application Research of Computers

基　　金：山西省自然科学基金资助项目(2008011030);山西省回国留学人员科研资助项目(2011-075);太原市大学生创新创业专项项目(2010;2011);太原科技大学校UIT项目(XJ2010040)

摘　　要：由于退化图像的点扩散函数难以准确确定,提出一种基于Fourier正交基函数的前向神经网络图像复原模型,该模型以一组Fourier正交基为隐层神经元的激励函数,根据误差传递算法进行权值修正,达到收敛目标。给出Fourier神经网络及其相应的衍生算法的图像恢复实现步骤。实验表明,该方法能较好地实现图像的复原。According to the fact that PSF （ point spread function） of the degraded image can＇ t obtain accurately, this paper constructed a feed-forward neural network for image restoration based on the Fourier orthogonal function. The hidden-layer neurons were activated by a series of Fourier orthogonal functions, updated its weight by the error back-propagation training algorithm and finally reached convergence target. This paper applied the Fourier neural network and its hidden-neuron growing algorithm to recover the fuzzy image. Experiments show they have better performances on image restoration.

关 键 词：图像复原 傅里叶正交基函数 傅里叶神经网络 衍生算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]