Hamburg 2016 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

T: Fachverband Teilchenphysik

T 38: Gammaastronomie II

T 38.5: Talk

Monday, February 29, 2016, 17:45–18:00, VMP9 SR 27

FACT - New Image Parameters Based on the Watershed-Algorithm — •Lena Linhoff, Kai Arno Bruegge, and Jens Buss for the FACT collaboration — TU Dortmund, Dortmund, Deutschland, Experimentelle Physik 5b

FACT, the First G-APD Cherenkov Telescope, is the first imaging atmospheric Cherenkov telescope that is using Geiger-mode avalanche photodiodes (G-APDs) as photo sensors. The raw data produced by this telescope are processed in an analysis chain, which leads to a classification of the primary particle that induce a shower and to an estimation of its energy. One important step in this analysis chain is the parameter extraction from shower images. By the application of a watershed algorithm to the camera image, new parameters are computed. Perceiving the brightness of a pixel as height, a set of pixels can be seen as 'landscape' with hills and valleys. A watershed algorithm groups all pixels to a cluster that belongs to the same hill. From the emerging segmented image, one can find new parameters for later analysis steps, e.g. number of clusters, their shape and containing photon charge. For FACT data, the FellWalker algorithm was chosen from the class of watershed algorithms, because it was designed to work on discrete distributions, in this case the pixels of a camera image. The FellWalker algorithm is implemented in FACT-tools, which provides the low level analysis framework for FACT. This talk will focus on the computation of new, FellWalker based, image parameters, which can be used for the gamma-hadron separation. Additionally, their distributions concerning real and Monte Carlo Data are compared.