High-energy flash radiography uses the penetrating power of pulsed X-rays and their interaction with matter to diagnose the geometric and physical properties of the target. Flash radiography is used to study the fluid dynamics of the high-speed motion process inside dense objects. The process of flash radiography imaging generates noise and blurring that significantly degrade image quality, lead to loss of image information, and interfere with accurate diagnosis of objects. Therefore, the use of image restoration techniques is essential to remove noise and recovering potentially clear images from the original radiographic images. Traditional radiation image restoration methods lack flexibility and generalization. Furthermore, neural network methods based on supervised learning require many actual image data sets to be obtained in advance, which limits network training. In this paper, we propose an unsupervised radiation image restoration method based on a deep image prior. The classical DeepRED framework is extended by adding weight-adaptive variational regularization. Automatic strategies are employed to estimate local regularization parameters, resulting in better noise removal and preservation of image edges and texture structures. The proposed method is evaluated through numerical simulations and experimental studies of flash radiographic images. The results indicate that the proposed method is effective in restoring flash radiographic images, removing noise, and preserving boundary region information.

Flash radiography,Image restoration,Unsupervised,Regularized