Abstract
Considering the problems of low resolution and rough details in existing mural images, this paper proposes a superresolution reconstruction algorithm for enhancing artistic mural images, thereby optimizing mural images. The algorithm takes a generative adversarial network (GAN) as the framework. First, a convolutional neural network (CNN) is used to extract image feature information, and then, the features are mapped to the high-resolution image space of the same size as the original image. Finally, the reconstructed high-resolution image is output to complete the design of the generative network. Then, a CNN with deep and residual modules is used for image feature extraction to determine whether the output of the generative network is an authentic, high-resolution mural image. In detail, the depth of the network increases, the residual module is introduced, the batch standardization of the network convolution layer is deleted, and the subpixel convolution is used to realize upsampling. Additionally, a combination of multiple loss functions and staged construction of the network model is adopted to further optimize the mural image. A mural dataset is set up by the current team. Compared with several existing image superresolution algorithms, the peak signal-to-noise ratio (PSNR) of the proposed algorithm increases by an average of 1.2–3.3 dB and the structural similarity (SSIM) increases by 0.04 = 0.13; it is also superior to other algorithms in terms of subjective scoring. The proposed method in this study is effective in the superresolution reconstruction of mural images, which contributes to the further optimization of ancient mural images.
Highlights
Ancient murals are the bright pearl in the treasure house of cultural heritage
Based on the aforementioned information, this study proposes a new superresolution reconstruction algorithm, which is applied to the superresolution reconstruction of ancient mural images. e improvement of the proposed algorithm is mainly as follows: (1) e network design takes generative adversarial network (GAN) as the basic framework, including the generative network and the discriminate network; MSE loss, VGg loss, and adversarial loss functions are introduced to optimize the network in two stages
Aiming at mural image optimization, this study proposed a superresolution reconstruction algorithm to enhance the artistry of mural images
Summary
Ancient murals are the bright pearl in the treasure house of cultural heritage. At present, the protection of murals mostly focuses on the field research of ancient murals and the restoration of damaged areas of murals. Convolution layers capture the abstract content of the image and eliminate the damage; deconvolution layers upsample the features and restore image details; symmetric skips are introduced, which makes the training converge faster This algorithm is likely to produce overfitting in superresolution reconstruction of mural image datasets. Zhang and An [18] introduced a superresolution reconstruction method based on migration learning and deep learning, which can obtain high-quality, high-resolution images and reduce the time cost of model construction. Ledig et al [19] proposed the SRGAN algorithm and designed a loss function to enhance the reality of the restored image In their method, the adversarial loss function of the generative adversarial network (GAN) was combined, which enables the output superresolution image to be more authentic. (3) e discriminant network increases the number of network layers, and residual modules are integrated to enable the network to extract more image information, and the expression ability of the discriminant network is increased to further optimize the generative network model
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
