AN ORIENTATION SELECTIVE NEURAL-NETWORK AND ITS APPLICATION TO COSMICMUON IDENTIFICATION

We propose a novel method for identification of a linear pattern of pi xels on a two-dimensional grid. Following principles employed by the v isual cortex, we employ orientation selective neurons in a neural netw ork which performs this task. The method is then applied to a sample o f data collected with the ZEUS detector at HERA in order to identify c osmic muons which leave a linear pattern of signals in the segmented u ranium-scintillator calorimeter. A two dimensional representation of t he relevant part of the detector is used. The results compared with a visual scan point to a very satisfactory cosmic muon identification. T he algorithm performs well in the presence of noise and pixels with li mited efficiency. Given its architecture, this system becomes a good c andidate for fast pattern recognition in parallel processing devices.