Correcting Lidar Intensity Signal for Target Detection Applications

Abstract

Laser scanning or as literary referred to as Lidar is the digital technique of obtaining

information about physical objects using laser light. Airborne Lidar has shown increasing utility for feature extraction applications through enhancing the physical objects recognition process. Airborne and terrestrial Lidar provide a radiometric information alongside standard geometric information to the end user such as intensity. Intensity can be considered one of the most important parameter to identify ground features towards accurate object recognition application. This value cannot be used directly because the signal may influence by many variables during the travel between the source light (sensor) and the target. These including atmospheric, incidence angle and the target properties effects, for this reason this value should be corrected before using. This research will focus on incidence angle effect to correct Lidar intensity signal. Incidence angle is a function of the illumination direction between the sensor and the target and the point cloud orientation (normal vector). This paper, investigates the possibility of improving an approach to compute the normal vector for individual point clouds following 3D moment invariant theory. This was applied by using the commercial (OPALS), this software contain tools to compute normal vector value for individual points which were used later to compute the required incidence angle value. This leads to correct the intensity value and deliver normalized value toward improvements of automatic feature recognition applications