Infrared small target detection algorithm based on multi-directional derivative and local contrast
Citations Over TimeTop 24% of 2019 papers
Abstract
Infrared small target detection technology plays an extremely important role in military remote warning, satellite remote sensing technology, guidance and anti-mission, UAV detection and tracking, and other fields. Most traditional algorithms have high detection false alarm when there are thin and highlighted patterns in the background. To address the abovediscussed problem, this paper proposes an infrared small target detection algorithm based on multi-direction derivative and local contrast. The algorithm utilizes the two-dimensional Gaussian distribution of small targets to compute multidirectional derivative on each pixel of the image. Simultaneously, a sliding window is constructed to compute the local contrast. Finally, the derivative result and the local contrast is combined to get the target saliency map. Compared with traditional infrared small target detection algorithms in terms of background suppression factor (BSF) and signal-to-clutter ratio (SCR), our algorithms has better performance in both indicators. In addition, Receive Operating Characteristic (ROC) curve is introduced to evaluate the performance of the algorithm. The curve demonstrates that the proposed algorithm can achieve high detection rate with low false alarm rate preserved. The experimental results show that the proposed algorithm is simple, efficient and has high detection accuracy.
Related Papers
- → False-Alarm Regulation in Log-Normal and Weibull Clutter(1973)304 cited
- → An Anti-FOD Method Based on CA-CM-CFAR for MMW Radar in Complex Clutter Background(2020)10 cited
- → The improved constant false alarm rate detector based on multi‐frame integration for fluctuating target detection in heavy‐tailed clutter(2023)4 cited
- Anderson-Darling Test for CFAR Detection(2009)
- Performance Analysis of Mean Level CFAR Detector in Pearson Distribution Clutter(2009)