De-noising and welding corner extraction method of point cloud from large-scale hull section

Reverse Engineering is one method for digital product development, which is more and more widely used in the measurement of big scale accessory and complicated curve surface reverse, for example, large-scale measurement in the shipbuilding industry, with the development of modern shipbuilding technology, this technology is highly emphasized. Data de-noising and corner extraction methods of point cloud from measured hull section are the key technology of reverse engineering, the result will affect the quality of model reconstruction directly. So in this paper, firstly present a comprehensive data de-noising method to the hull section and get better data points, make use of the edge extraction and sorting method to obtain ordered edge points, secondly propose the adaptive method to calculate the support region for the straight line fitting, finally calculate the intersection of the straight lines to get corners and borders of point cloud. Through experiments we get the results, verify that data de-noising and corner extraction methods are feasible and have good anti-interference ability, and good ability to detect and locate the corner points.