Image Segmentation in Liquid Argon Time Projection Chamber Detector
This addresses the need for efficient automatic reconstruction in neutrino studies, but it is incremental as it builds on existing segmentation techniques for a specific domain.
The paper tackles the problem of segmenting particle tracks from noise in Liquid Argon Time Projection Chamber detector images by proposing a method that uses pixel feature descriptors and a supervised classifier, achieving results on a hand-labeled dataset.
The Liquid Argon Time Projection Chamber (LAr-TPC) detectors provide excellent imaging and particle identification ability for studying neutrinos. An efficient and automatic reconstruction procedures are required to exploit potential of this imaging technology. Herein, a novel method for segmentation of images from LAr-TPC detectors is presented. The proposed approach computes a feature descriptor for each pixel in the image, which characterizes amplitude distribution in pixel and its neighbourhood. The supervised classifier is employed to distinguish between pixels representing particle's track and noise. The classifier is trained and evaluated on the hand-labeled dataset. The proposed approach can be a preprocessing step for reconstructing algorithms working directly on detector images.