A FPGA and Zernike Moments Based Near-Field Laser Imaging Detector Multi-scale Real-Time Target Recognition Algorithm

Abstract

Laser imaging Detector is one direction of Detectors' development. The encounter attitude of the detector and aircraft can be arbitrary, and there may be clouds and smoke, thereby causing interference to imaging. This paper presents a stable algorithm based on Zernike moments which can effectively recognize flying plane and against the cloud and smoke interference. The paper attempts to calculate Zernike moments of an image by using FPGA, and treat them as some kinds of feature to identify target by using the Fisher classifier which can achieve a good classification result.