基于改进Paik型Boltzmann机的图像复原

图像复原可以看作是一个极小化问题, 利用Boltzmann机固有的随机神经网络性质可以保证搜索过程中网络不陷入局部极小点. 对传统Boltzmann机做如下改进：首先将Paik算法与Boltzmann机结合, 其次为了加快速度, 将串行模式推广到并行模式. 第三为了增加计算精度, 使用了亚单位步长增进技术. 最后为折中收敛速度与收敛精度这一对矛盾, 采用了自适应步长策略. 对于算法的每一处改进均有详细的理论验证、收敛性分析和残差变化讨论. 实验表明该方法能够无限逼近能量最小点, 复原结果优于改进Boltzmann机0.5-0.8dB, 且收敛速度仅有1/3.

Image Restoration Based on Modified Paik Boltzmann Machine

Image restoration can be regarded as a minimization problem, which can be solved by Boltzmann machine, since it is certainly fall into global minimum not local minimum by its inherent random neural network. Following improvements are done on traditional Boltzmann Machine: firstly integrate Paik into Boltzmann Machine; secondly extend serial model to parallel model; thirdly subunit step is adopted to increase precision; finally make the step adaptive to trade off the contradiction of convergence precision and convergence velocity. Each improvement was expatiated via theories, convergence analysis, and residual. Experiments demonstrate that this proposed method can approximate the overall minimum of the energy function infinitely, and the restored image by this method is superior to that of traditional Boltzmann.