Study On The Design Of A Simple 3D Scanning System

Scanning and three-dimensional reconstruction of the three-dimensional object surface has important applications in many fields. The current scanning devices are bulky with high cost, which greatly limit their application. In this paper, we aim at constructing a simple 3D scanning system, propose the wa y to build a simple 3D scanning system, and design the relevant three-dimensional reconstruction algorithm. The main research contents include:(1) Through a detailed analysis of the operating principle of binocular stereo vision system and the three-dimensional reconstruction technology of two-dimensional images, we select two ordinary digital cameras, an ordinary computer(or smartphone) and other equipment to build a simple binocular 3D scanning system, greatly reduce the cost of the scanning system.(2) Based on silhouettes and orthogonal projection method, we use a common low-end smart phone to do the three-dimensional reconstruction test of standard material and the model photos, so as to reduce the cost effectively and facilitate the extraction of materials and relevant two-dimensional images timely.(3) For solving the problem of the target area identification in the collected photos, we establish multiple nonlinear regression models; convert the pixel row number to the corresponding coordinates and the discrete pixels to surface function as dimension reduction processing, thereby enhancing model applicability.(4) For the image distortion problem during the shooting process, we propose the calibration algorithm and reconstruct the 3D image directly o f the real object photos to overcome the angle error, which improve the adaptability simple 3D scanning system.According to the study, this paper presents the hardware and the supporting software of a low-cost simple 3D scanning system, and provides a lot of experiments with this system. It also conducts preliminary evaluation and further prospects based on the experimental data of the system.