基于稀疏编码的图像超分辨率复原

图像超分辨率技术一直是计算机视觉领域研究的热点,为提高图像重建速度与精度,提出了一种稀疏编码与神经网络相结合的图像超分辨率算法。首先利用前馈神经网络严格对应稀疏编码过程中的每个步骤,然后通过反向传播算法对稀疏编码的所有组成部分进行联合训练,得到最为精确的高分辨率图像。级联多个稀疏编码网络增加了算法的灵活性,并减少了伪影。

Image Super-resolution Restoration Based on Sparse Coding

Image super-resolution technology has always been a hot spot in the field of computer vision. To improve the speed and accuracy of image reconstruction,an image super-resolution algorithm combining sparse coding and neural network is proposed. Firstly,the feedforward neural network is used to strictly correspond to each step in the sparse coding process,and then all components of the sparse coding are jointly trained by the back propagation algorithm to obtain the most accurate high-resolution image. Concatenating multiple Sparse coding networks increase the flexibility of the algorithm and reduce artifacts.