Detecting Known Objects in a Noisy Scene Using Generalized Likelihood Ratio Test
Qi Chuan Tian, Jie Wen · Advanced materials research · 2012
According to the problem of the identification and localization of a known object in a scene, satisfied detection results can not be achieved using traditional detectors for images in photon-limited noise, an algorithm named Generalized Likelihood Ratio Test (GLRT) was derived for detecting known objects in a noisy scene. We used this algorithm to evaluate the existence of tiger in photons-limited images. Results show that the GLRT algorithm is effectiveness in detecting and localizing a known object embedded in a background image from photon-limited observations.