Simple and efficient registration of 3D point cloud and image data for an indoor mobile mapping system

Registration of 3D lidar point clouds with optical images is critical in the combination multisource data. Geometric misalignment originally exists pose data between and images. To improve accuracy initial applicability integration points image data, we develop a simple but efficient registration method. We first extract features from images: are extracted single-frame using classical Canny operator. The cost map subsequently built based on edge detection. optimization direction guided by map, where low represents desired direction, loss function also considered to robustness proposed Experiments show positive results.