基于2D-TDI及运动补偿的红外图像增强算法

IR Image Enhancement Algorithm Based on 2D-TDI with Motion Compensation

针对红外图像的特点,提出了一种基于2D-TDI及运动补偿的新方法。将序列红外图像根据运动特性进行多分辨运动估计,然后根据得到的运动矢量进行2D-TDI。多分辨运动估计的引入,提高了运动估计对噪声的抗干扰能力,并有效地减小了运算量。该算法克服了传统图像直接叠加引入的图像边缘模糊的缺点,并能明显提高图像的信噪比。用真实的红外图像序列进行了处理,实验证明了算法的优越性,并取得了满意的效果。

A new IR image enhancement algorithm is proposed based on 2D-TDI with motion compensation on the basis of analyzing the features of IR image. Multi-resolution motion estimation is carried out on the IR image sequence according to the movement characteristics of the target, then 2D-TDI is applied to the image in line with the movement vector acquired by motion estimation. As a result of the introduction of multi-resolution motion estimation, anti- noise capability of motion estimation is improved with a small amount of computing. This method could improve the SNR of image significantly without the edge indistinctness induceed by direct image addition of common method. Experiments on real infrared image sequence prove the superiority of the algorithm, and have achieved satisfactory results.