The Development of Gazing Algorithms for Tracking Oriented Recognition (GATOR)

ID Vision proposes to develop a computer vision system that hosts a suite of Gazing Algorithms for Tracking Oriented Recognition (GATOR), to accomplish very robust target recognition and tracking at a video rate of 30 Hz. The uniqueness of GATOR is that it employs tightly structured advanced Automatic Target Recognition gorithms to progressively increase the confidence level during recognition, which enables a rugged performance and real-time processing in complicated battlefield situations. The biologically inspired GATOR algorithms are mathematically sound and computationally efficient, which include: 1) a Morphological Shared Weight Neural Network to provide target recognition and target list updating; 2) a novel target wavelet filter to facilitate motion segmentation, target registration, and image alignment; and 3) a fuzzy logic data fusion scheme to integrate recognition results from multiple frames of images. The combination of three technologies could virtually eliminate the false alarm. During Phase I, the feasibility of the GATOR system will be proved by prototype demonstration, which includes motion segmentation, MSWNN target classification and updating, fuzzy logic data fusion, and target registration and tracking, During Phase II, the GATOR system will be further improved in terms of real-time processing, and directly using wavelet compressed data for recognition and tracking.