Morphology-based building target detection from forward-looking infrared imagery in dense urban areas

In this article, we have presented a morphology-based method for building recognition from FLIR image sequences captured by the midwave calibrated camera in dense urban areas. Based on the imaging parameters (the horizontal and vertical angle of field of view, the imaging distance, and the elevation), the imaging size of the building target can be determined in real time, and then the SE, which has a high correlation with the target's imaging size, has been generated. Additionally, to improve the performance of target detection and straight-line extraction, the line segment grouping method and the gradient-based straight-line extraction method are adopted. Experimental results show the algorithm can recognize the prominent building object with higher local contrast in FLIR image sequences in dense urban areas and maintain higher performance.