The Mogao Grottoes, located at the western end of the Hexi Corridor in Dunhuang, constitute a splendid artistic treasure trove of ancient Chinese civilization. However, due to various external factors, such as the environment and human activities, existing murals commonly suffer from fading and discoloration problems. The utilization of color restoration techniques grounded in deep learning promise perpetual preservation of mural images. Nevertheless, it is challenging to acquire a substantial number of authentic fading images and corresponding reference images as paired data for training, which constrains the scope of research and development in the domain of mural color restoration. The images generated by the current color restoration methods based on cycle generative adversarial networks suffer from poor semantic consistency, blurred edges, false colors and artifacts, and other mismatches with human visual perception. This paper proposes a novel approach of color restoration for unpaired mural images that uses cycle-consistent generative adversarial networks with an attention mechanism and a spectral normalization discriminator to address key challenges. First, to reduce the false colors and artifacts in the restored image caused by the insufficient extraction of mural detail features, a global attention mechanism based on a combination of average pooling and maximum pooling is constructed in the generator. This mechanism is designed to learn the effective information of the feature maps adaptively. In addition, we employ an SN-Patch discriminator to enhance the training stability and convergence speed of the model and improve the clarity of the color restored image. Finally, to further optimize the generated images, the network applies a composite loss function, a linear combination of adversarial loss, cycle consistency loss, identity mapping loss, and cyclic perceptual consistency loss, which contribute to improving texture quality and generating visually more natural color-restored images. The findings from the experiments conducted on both simulated and genuine faded mural images illustrate that the introduced approach outperforms others in terms of both objective and subjective evaluation standards, achieving a more precise recreation of the colors and intricate details present in the faded mural imagery.

Cultural heritage digitization is of great significance for the protection, restoration, and rejuvenation of cultural relics. In particular, the investigation of fading mechanisms is essential for virtual restoration to accurately recreate the original appearance of artifacts and facilitate humanistic and historical research. For the purpose of investigating the color fading mechanism of pigments, we propose a color fading time series model using a combined long short-term memory recurrent neural network modified by the gray wolf optimization algorithm (GWOAD-LSTM). First, the general gray wolf algorithm was scaled up to two dimensions and combined with an LSTM model for optimal parameter search. Second, six pigments commonly used in painted artifacts were subjected to simulated aging experiments. Third, by applying the experimental data to different predictors, the results of the Back Propagation Neural Network (BPNN), Long Short-Term Memory (LSTM), Long Short-Term Memory on Gray Wolf Optimizer (GWO-LSTM), and GWOAD-LSTM models were compared. The results showed that our proposed GWOAD-LSTM model outperformed other models in terms of accuracy and generalization ability, especially in predicting hLC color attributes. Our study aimed to provide a new application tool for the restoration and rejuvenation of painted artifacts.

The pigments of cultural color paintings have faded with the passage of time. Color segmentations are essential for digital color reconstruction, but the complexity of color paintings makes it challenging to achieve high-precision segmentation using previous methods. To address the challenges of color painting segmentation, an optimized strategy based on U-Net is proposed in this paper. The residual blocks of a residual network (ResNet) are added to the original U-Net architecture, and a UR-Net is constructed for the semantic segmentation of color paintings. The following steps are taken. First, datasets of color paintings are obtained as training and test samples and are labeled with the two following pixel colors: earth red and ultramarine blue. Second, residual blocks are improved and added to fit the U-Net architecture. Then, a UR-Net is constructed and trained using the samples to obtain the semantic segmentation model. Finally, the effectiveness of the trained UR-Net model for segmenting the test samples is evaluated, and it is compared with the K-means clustering algorithm, ResNet, and U-Net. Data from several studies suggest that the segmentation accuracy of the UR-Net model is higher than that of other methods for the color segmentation of painted images, and the IoUs of the segmented earth red and ultramarine blue pixels are 0.9346 and 0.9259, respectively, achieving the desired results. The proposed UR-Net model provides theoretical and methodological support for further in-depth research on color recognition and segmentation of cultural color paintings.