High-power laser facilities necessitate predicting incremental damage to final optics to identify evolving damage trends. In this study, we propose a surface damage detection method utilizing image segmentation employing ResNet-18 and a damage area estimation network employing U-Net++. Paired sets of online and offline images of optics obtained from a large laser facility are used to train the network. The trends of varying damage could be identified by incorporating additional experimental parameters. A key advantage of the proposed method is that the network can be trained end to end on small samples, eliminating the need for manual labeling or feature extraction. The software developed based on these models can facilitate the daily inspection and maintenance of optics in large laser facilities. By effectively applying deep learning techniques, we successfully addressed the challenges faced by traditional methods in handling complex environments, achieving the accurate identification and prediction of damages on optics.
High power laser facility requires prediction for incremental damage on final optics in order to find varying damage developing trends, then in terms of the trends we can appropriately formulate maintain plans for long-term physical experiment operation on the facility. This essay indicates an image process technology via deep learning. By comparing fit offline optics image with online optics image, the authenticity and the area of the damage can be correctly defined. Besides, due to the comparison we can establish a damage detection model. Through appropriate algorithm and image matching technology, we could manage element damages on the timeline. And the damage varying trends could be found when additionally combined experimental parameter with analysis on specific damage area. Thus the surplus value of assembly can be optimized. By programming a software for elements management this technology could be used on high power laser facility.
High-power laser facilities necessitate predicting incremental damage to final optics to identify evolving damage trends. In this study, we propose a surface damage detection method utilizing image segmentation employing ResNet-18, and a damage area estimation network employing U-Net++. Paired sets of online and offline images of optics obtained from a large laser facility is used to train the network. The trends of varying damage could be identified by incorporating additional experimental parameters. A key advantage of the proposed method is that the network can be trained end-to-end on small samples, eliminating the need for manual labeling or feature extraction. The software developed based on these models can facilitate the daily inspection and maintenance of optics in large laser facilities.