基于改进的Harris角点的机载红外图像电子稳像

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

飞机飞行过程中产生的颠簸、晃动会使得采集到的红外图像序列存在抖动,影响后续对红外图像的观察,不利于对红外图像目标的识别、定位与跟踪。本文提出了一种基于改进的Harris角点法对机载红外图像进行实时电子稳像的方法。首先采用改进的Harris角点响应函数结合距离约束条件进行角点检测,该方法保证即使对图像质量较差的红外图像,也能检测出数量足够且分布均匀的角点;然后基于检测出的角点利用提出的关键帧参考方式结合多尺度的金字塔光流算法进行跟踪匹配,完成运动估计,进而实现对机载红外图像的电子稳像。用该方法对多组含有抖动的640×512尺寸的红外图像序列进行稳像处理,实验表明,本文提出的算法与传统的Harris角点检测算法相比有更好的检测效果,能够很好地去除红外机载图像序列的抖动,且能够满足50 fps采集速率下的实时处理。

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.