Research on Image Restoration of Ancient Buildings Based on Deep Convolution Network

Inhalt:
Handicraft repair has appeared hundreds of years ago. Through this technology, people can repair damaged art products and make the repaired items meet the visual requirements of human beings as much as possible. Image restoration technology, as a technology related to handicraft restoration, has been paid attention to by many people in recent years. Digital image restoration is an important research direction in the field of computer vision. In recent years, with the excellent performance of deep learning in image processing, more and more research teams begin to use the relevant methods of deep learning to deal with the problem of image restoration. Image restoration technology is an efficient image tampering technology. This tampering method will not leave obvious traces, which brings severe challenges to forensics. Image restoration passive forensics technology judges the authenticity and integrity of the image by detecting the extracted image features. In view of the remarkable achievements of deep neural network in automatically finding the features needed for classification. The vision system can quickly find the object of interest in the picture and send it back to the brain for further processing. Introducing the attention mechanism of the vision system into the computer can greatly improve the performance of the computer. Saliency calculation in visual mechanism can provide the fast computing ability to locate the prominent foreground, exclude the background, and accurately complete various image analysis tasks, which is widely used in various computer vision tasks.