Search

In order to improve the visualization effect of low resolution digital images, this study proposes a low resolution digital image reconstruction method based on Rational Function Model. Firstly, bilateral filtering is employed to preprocess low resolution digital images for denoising with preservation of considerable edge details of the images. Secondly, to make the digital images unlimited to the coordinate system in the reconstruction process, a Rational Function Model with general attributes is constructed. Data obtained from the model is taken as the input information, and the generative adversarial network is used to extract image features, which lays the data foundation for subsequent image reconstruction. Thirdly, SIFT algorithm and Difference of Gaussians function are used for accurate feature extraction to compensate for the extraction deviation caused by the defects of the training set itself in the generative adversarial network. Finally, the processes of feature point direction matching, wavelet transform, bilateral regularization processing, pixel correction, edge adaptive processing, etc. are carried out for ortho correction of the image function model. On this basis, image reconstruction is eventually established. The experiment shows the edges of the reconstructed digital image are non-aliased and relatively smooth, and the texture direction and shape in the original image are well maintained, so that the details of the target parts are greatly preserved, with high accuracy of feature point matching. Furthermore, the peak signal-to-noise ratio of the reconstructed digital image ranges from 90.3dB to 92.7dB, and the structural similarity index varies from 0.93 to 0.96, further demonstrating the effectiveness of this method. [ABSTRACT FROM AUTHOR]

The traditional image super-resolution (SR) reconstruction methods do not carry out sparse coding when performing SR reconstruction, resulting in poor structural similarity of image SR reconstruction. Therefore, this paper proposes a a super-resolution reconstruction method for polymorphic sparsity regularized images based on machine learning algorithms and builds a sparse representation model of polymorphic regularized images. For sparse representation models with different norm constraints, the orthogonal matching tracking in the greedy algorithm is used perform sparse coding to solve the sparse representation coefficient. The least square method in machine learning algorithm is used to solve the inverse operation of dictionary matrix update, construct the image super-resolution reconstruction minimization objective function, and solve the loss function between the super-resolution reconstruction image and the actual image, so as to complete the super-resolution reconstruction of polymorphic sparsity regularized image. The experimental results show that the edge recovery of the image under the proposed method is clearer, and the edge with more distinct edges and corners can be reconstructed. The Peak-Signal-to-Noise Ratio (PSNR) of the super-resolution reconstruction image is 60.5dB, and the Structural Similarity (SSIM) of the super-resolution reconstruction image is as high as 0.99, which effectively improves the image reconstruction effect. [ABSTRACT FROM AUTHOR]