Dim and Small Target Detection Based on Spatio-Temporal Jitter Estimation

Abstract

To mitigate the impact of the halo effect on target detection in backgrounds of strong light, we propose a dim and small target detection algorithm based on spatio-temporal jitter estimation. First, a multidirectional spiral vignetting correction model algorithm, which introduces temporal and spatial information to represent the gradient distribution disparities between the target and its surrounding blocks, thus estimating the optimal background and negating the influence of light halo while realizing the background suppression. Subsequently, this article introduces a multiframe energy accumulation algorithm based on jitter estimation, this algorithm combines data from the 2-D Gaussian surface fitting and the optical flow method (OFM). Based on the jitter estimated, interframe registration is conducted, efficiently facilitating multiframe energy accumulation and enabling the detection of the target signal. Verification through experiments in varying scenes indicates that the detection performance of the algorithm proposed herein is significantly enhanced and capably detecting targets.