Infrared Patch-Image Model for Small Target Detection in a Single Image

Abstract

The robust detection of small targets is one of the key techniques in infrared search and tracking applications. A novel small target detection method in a single infrared image is proposed in this paper. Initially, the traditional infrared image model is generalized to a new infrared patch-image model using local patch construction. Then, because of the non-local self-correlation property of the infrared background image, based on the new model small target detection is formulated as an optimization problem of recovering low-rank and sparse matrices, which is effectively solved using stable principle component pursuit. Finally, a simple adaptive segmentation method is used to segment the target image and the segmentation result can be refined by post-processing. Extensive synthetic and real data experiments show that under different clutter backgrounds the proposed method not only works more stably for different target sizes and signal-to-clutter ratio values, but also has better detection performance compared with conventional baseline methods.

Infrared small target detection based on directional zero-crossing measure

“…Predicting methods are used to estimate the background image, and then the original image is subtracted from the background image to get the pure target-like map. Based on the nonlocal self-correlation property of the infrared image, an infrared patch-image (IPI) model is presented to segment small targets by Gao et al [6]. In IPI model, a sparse matrix is used to represent the foreground image while a low-rank matrix is used to represent the background image.…”

Infrared Small Target Detection Using Modified Order Morphology and Weighted Local Entropy

“…In deal with the shortcomings of classical tophat operation, Bai et al [13] proposed a new tophat operation, which enhanced the effect of target detection, but still had a high false alarm rate when noises shape are similar to target's. Besides, there are large amounts of modified algorithms derived from mathematical morphology, including the toggle contrast operator [14], multiscale center-surround top-hat transform [15], hitormiss transform [16].In addition, there are methods based on vector machines [17], sparse decomposition [18], [19] and so on. However, some methods are computationally intensive and complicated, some algorithms have high false alarm rate when dealing with complex sky background, especially dense clouds, cloud edge and target-like interference.…”

Low-Rank and Sparse Decomposition Based Frame Difference Method for Small Infrared Target Detection in Coastal Surveillance

“…Recently, low-rank and sparse matrices recovery theory (LRMR) [5], [6] has been proposed and applied in target detection and tracking. And it has been proved more effective compared with conventional baseline methods in some situations [7]. These LRMR methods can be mainly classified into two kinds.…”

“…These LRMR methods can be mainly classified into two kinds. The first kind detects the small infrared target in a single frame [7]- [9], but its performance could degrade rapidly when the SCR is low. The other kind [5], [6], [10]- [12] makes use of the information of all the frames in a video sequence, the moving targets could then be more easily detected in a low-SCR scene.…”