IR small target detection based on human visual attention using pulsed discrete cosine transform

摘要: Detection of small targets in an infrared (IR) image with high reliability is very important for defence systems. Small IR are defined as salient features which attract the attention human visual system. In this study, a robust method detection proposed based on HV attention. method, first, Gaussian-like feature maps extracted from original image. Then, saliency (SMs) created pulsed discrete cosine transform, target and background clutter suppressed. Finally, to increase contrast between raise robustness against false alarms, SMs fused adaptively. Experiments carried out data set including real-life images well various complicated backgrounds. Qualitative quantitative assessments show that can detect more effective compared other methods Therefore, it be used many applications minimum alarms.