Wavelet-Based Contourlet Transform and Kurtosis Map for Infrared Small Target Detection in Complex Background

Abstract

Wavelet-based Contourlet transform (WBCT) is a typical Multi-scale Geometric Analysis (MGA) method, it is a powerful technique to suppress background and enhance the edge of target. However, in the small target detection with the complex background, WBCT always lead to a high false alarm rate. In this paper, we present an efficient and robust method which utilizes WBCT method in conjunction with kurtosis model for the infrared small target detection in complex background. We mainly made two contributions. The first, WBCT method is introduced as a preprocessing step, and meanwhile we present an adaptive threshold selection strategy for the selection of WBCT coefficients of different scales and different directions, as a result, the most background clutters are suppressed in this stage. The second, a kurtosis saliency map is obtained by using a local kurtosis operator. In the kurtosis saliency map, a slide window and its corresponding mean and variance is defined to locate the area where target exists, and subsequently an adaptive threshold segment mechanism is utilized to pick out the small target from the selected area. Extensive experimental results demonstrate that, compared with the contrast methods, the proposed method can achieve satisfactory performance, and it is superior in detection rate, false alarm rate and ROC curve especially in complex background.

Robust Unsupervised Multifeature Representation for Infrared Small Target Detection

“…Due to the limited feature representation of traditional Fourier transforms, Qi et al [17] utilize phase spectrum of quaternion Fourier transform to enhance Gaussian-like shape targets. In order to obtain more spectral information, many methods based on wavelet transform have been developed [18]- [22]. For example, Xin et al [21] use the Gabor wavelet transform to extract features from different scales and angles for better background suppression.…”

Fine-Grained Guided Model Fusion Network with Attention Mechanism for Infrared Small Target Segmentation

“…Despite the simple calculation of classic target segmentation methods, these methods are easily interfered by random noise. With the development of computer vision techniques, several hybrid segmentation methods, including morphological-based methods [35,36], genetic-based methods [37,38], and wavelet-based methods [39][40][41], have been proposed for better performance. Tese methods provide suitable detectors for specifc scenarios and increase their adaptability in complicated tasks.…”

Fast and robust star detection algorithm based on the dyadic wavelet transform

“…In the ROC curve [30], the vertical axis represents the detection rate $${P}_d$$ and the horizontal axis represents the false alarm rate $${P}_f$$, which can be expressed as $$$$\begin{equation}\left\{ { \def\eqcellsep{&}\begin{array}{@{}*{1}{c}@{}} {{P}_d = \dfrac{{{N}_C}}{{{N}_T}}}\\[12pt] {{P}_f = \dfrac{{{N}_e}}{{{N}_P}}} \end{array} } \right. \cdot \end{equation}$$$$where $${N}_T$$ represents the actual number of star points in the star image, $${N}_C$$ represents the number of correctly detected star points, $${N}_e$$ represents the number of star points for error detection, and $${N}_P$$ represents the total number of star points detected by the algorithm (including correct detections and false detections).…”

“…In the ROC curve [30], the vertical axis represents the detection rate P d and the horizontal axis represents the false alarm rate P f , which can be expressed as…”