The small target detection based on maximum likelihood estimation and spot detection operator

Abstract

In this paper, aiming at the small target detection problem in the infrared image sequence, we propose a small target detection method based on maximum likelihood estimation and NNLoG spot detection operator. Compared with the traditional method, our proposed method can partially solve the nonlinear motion of the small target in image sequence. The real target trajectory is approximated by polynomial to enhance the signal to noise ratio of target. To validate the proposed method, we create eight experiments to simulate. The experiment result shows that our method is very valuable for small target detection.

Small Target Detection Based on Squared Cross Entropy and Dense Feature Pyramid Networks

“…Compared with cascaded networks, this type of structure can extract and fuse multi-scale features of targets better, and extract directional features. Hou and Mei [13] proposed a weak target detection method MLE-NNLoG based on nonlinear maximum likelihood estimation, which can effectively detect linear or nonlinear moving targets. But for non-linear moving targets, this method needs to perform multi-dimensional search, and when the target movement is more complex, the number of dimensions to search is more, and the real-time performance of target detection is also worse.…”