Small moving targets detection in infrared image using fuzzy cognitive map

Abstract: Detection of small moving targets is of utmost importance in defense systems. For this purpose, using Infrared (IR) images is preferred to visible images for the reason that in circumstances such as marine navigation in fog and dust conditions, visible images are coming with false alarms that render the detection of small target difficult. So far different methods presented for IR small targets detection and false alarm reduction. In this study, first of all the image was enhanced with a morphological operation called Top-Hat. Then, the IR small target detection problem was considered as a tow-class pattern recognition problem and Fuzzy Cognitive Map (FCM) was applied for the classification. FCM is a simple, but also powerful intelligent method which is used for modeling and analysis of information. This paper is the first attempt to present FCM for IR small target detection. The obtained results indicate that FCM can detect the IR small targets with higher accuracy than any other pattern recognition methods.