<title>New approach to laser image processing for dynamic scenes</title>

Abstract

This paper presents the signal processing method of laser images used for detection of small collinear obstacles in helicopter airborne applications. It is very difficult, if not almost impossible, for a regular passive imaging sensor based on CCD/FLIR detectors to detect such small, remote objects. It was shown in the literature that the fusion of range and intensity data is needed for such a task. The paper presents a new and improved algorithm for real time image processing which enable detection of small tiny objects, from a fast moving platform. Two issues related to 3D visual sensing from laser sensors are addressed here. First, concerning the filtering and reduction of the large amount of data to be coped with for enabling real-time detection with low false alarm rate, in any ambient light conditions. Second, the fusion of range and intensity data for detecting small objects form cluttered data with high probability of detection. The preliminary experimental results show the potential of the proposed algorithm, based on spatio-temporal integration, to detect with high probability such objects like electrical cables from ranges of up to 400 meters.© (1997) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.