Abstract
Real-time small infrared (IR) target detection is critical to the performance of the situational awareness system in high-altitude aircraft. However, current IR target detection systems are generally hardware-unfriendly and have difficulty in achieving a robust performance in datasets with clouds occupying a large proportion of the image background. In this paper, we present new results by using an efficient method that extracts the candidate targets in the pre-processing stage and fuses the local scale, blob-based contrast map and gradient map in the detection stage. We also developed mid-wave infrared (MWIR) and long-wave infrared (LWIR) cameras for data collection experiments and algorithm evaluations. Experimental results using both publicly available datasets and image sequences acquired by our cameras clearly demonstrated that the proposed method achieves high detection accuracy with the mean AUC being at least 22.3% higher than comparable methods, and the computational cost beating the other methods by a large margin.
Highlights
Small and Dim Infrared TargetHigh-altitude aircraft has great potential in early warning and detection, space offense and defense, and electronic countermeasures, which become increasingly significant in a modern battlefield
Since the previous application of the Laplacian of Gaussian (LoG) filter [16] and the difference of Gaussian (DoG) filter [17], which are widely used in blob-target detection, difference of Gaussian (DoG) filter [17], which are widely used in blob-target detection, we propose to apply the DoG filter to the input image because the DoG is more hardwarewe propose to apply the DoG filter to the input image because the DoG is more hardwarefriendly than the LoG filter [24]
[22], get image; (b–h) normalized 3-D mesh obtained by the local contrast measure (LCM) [18], infrared patch-image (IPI) [10], MCPM [19], local intensity and gradient (LIG) [22], partial sum of the tensor nuclear norm (PSTNN) [12], facet kernel and random walker (FKRW) [15], and the proposed method; and (i) precision–recall curve (PRC) comparison
Summary
High-altitude aircraft has great potential in early warning and detection, space offense and defense, and electronic countermeasures, which become increasingly significant in a modern battlefield. IR target characteristics and proposed a new detection algorithm for In this paper, we present small and dim target detection.new results of applying the efficient and effective detection method combining local contrast, scale and and effective gradient detection features. The key contributions of the proposed method are described as follows: method combining local contrast, blob-like morphology and scale and gradient features further improve theofdetection performance imaging in [4].To. The key contributions the proposed method in arereal described as system, follows:the earlier framework for small and dim target detection method in was fine-tuned. To further improve the detection performance in real imaging system, the earlier the computational time is a little more than the earlier version, the background clutter is framework for small and dim target detection method in [23] was fine-tuned.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.