Neuro-explicit semantic segmentation of the diffusion cloud chamber

For decades, in diffusion cloud chambers, different types of subatomic particle tracks from radioactive sources or cosmic radiation had to be identified with the naked eye which limited amount data that could processed. In order allow these classical detectors enter digital era, we successfully developed a neuro-explicit artificial intelligence model that, given an image chamber, automatically annotates most visible according type process created it. To achieve this goal, combined attention U-Net neural network architecture methods shape detected tracks. Our experiments show effectively detects and approach decreases misclassification rate rare particles by 73% compared solely using U-Net.