Interested region selection and super-resolution reconstruction of depth image for scanning lidar

Abstract

Scanning lidar scans the target region point-by-point and measures the time of flight (TOF) of laser signal at each point to obtain the 3D information of the target surface. By using fixed size of scanning spot, the resolution of reconstructed depth image is consistent with the number of scanning points. Therefore, traditional scanning lidar is hardly to achieve high resolution and scanning efficiency simultaneously. Aimed to address this issue, we propose a method of interested region selection and depth image super-resolution reconstruction. By constructing a simulation target region with 10 m × 10 m, the proposed method is used to scan this region. The position of the interested region is obtained by scanning the full field of view (FOV) with a large spot. Then the interested region with 4 m × 8 m is fine scanned with reduced scanning spot. By using the super-resolution reconstruction method of depth image, the resolution of the depth image obtained by fine scanning with 40 × 80 points is increased by two times. And the depth image of the interested region with 80 × 160 pixels is obtained. The simulation result shows that the lidar based on this method can give consideration to both high scanning efficiency and the resolution of reconstructed depth image.