Generative Model For Material Irradiation Experiments Based On Prior Knowledge And Attention Mechanism

MinCong Luo, Li Liu

Abstract: Material irradiation experiment is dangerous and complex, which requires large number of high-level expertise in the manual processing of experimental images and data. In this paper, we propose a generative adversarial model based on prior knowledge and attention mechanism to achieve the generation of irradiated material images (data-to-image model), and a prediction model for corresponding industrial performance (image-to-data model). With the proposed models, researchers can skip the dangerous and complex irradiation experiments and obtain the irradiation images and industrial performance parameters directly by inputing some experimental parameters only. We also introduce a new dataset ISMD which contains 22000 irradiated images with 22,143 sets of corresponding parameters. Our model achieved high quality results by compared with several baseline models. The evaluation and detailed analysis are also performed.

Keywords: Generative Model, Images of Irradiation Experiments, Prior Knowledge, Attention Mechanism