Electronic Image Stabilization of Airborne Infrared Images Based on Improved Harris Corner Detection

WANG Shiman,XU Wenhai,DONG Lili,XU Zhou
DOI: https://doi.org/10.3724/sp.j.7102068595
2020-01-01
Abstract:Bumps and sloshing generated during aircraft flights disrupt the capture of a sequence of infrared images. This affects the subsequent observation of the images, as well as the identification, location, and tracking of the target. In this paper, a method for real-time electronic image stabilization of airborne infrared images is proposed based on an improved Harris corner method. First, an improved Harris corner response function was combined with a distance constraint to perform corner detection. This method ensures that a sufficient number of corner points with uniform distribution can be detected, even for infrared images with poor image quality. The proposed key-frame reference method was then combined with the multi-scale pyramid optical flow algorithm;furthermore, the detected corner points were applied to the aforementioned combination, thereby achieving tracking matching and motion estimation. Consequently, electronic image stabilization of the airborne infrared images was realized. Using this method, multiple sets of jittery infrared image sequences were processed;the image size was 640×512. Experimental results show that the proposed algorithm exhibits a better corner detection effect than the traditional Harris corner detection algorithm. It can remove jitter from the airborne infrared image sequence and ensure real-time processing at an acquisition rate of 50 fps.
What problem does this paper attempt to address?