Remote three-dimensional scanning system and method based on machine vision

Abstract

The invention discloses a remote three-dimensional scanning system and method based on machine vision. The system comprises two parts, namely a hardware equipment part and a software system part. The hardware equipment includes image acquisition equipment, an optical filter, a light source, camera calibrating equipment, etc. In order to solve the problem of rapidly acquiring an object three-dimensional point cloud under a natural environment, the invention designs a set of three-dimensional scanning hardware equipment and further has a matching software system developed. Furthermore, by referring to an image acquisition method, a camera focal length calibrating method, a system calibrating method, an image processing method and a three-dimensional point cloud acquisition method, the system can rapidly achieve three-dimensional scanning on objects and further realize robustness of equipment on ambient light and also keep the three-dimensional point cloud calculation precision. Compared with the prior art, the system which is not limited by ambient light conditions during a three-dimensional point cloud acquiring process achieves a relatively high scanning speed and relatively high scanning precision. The system characterized by simple equipment structure and low cost is not limited by time and place when in use.

Description

technical field [0001] The invention belongs to the field of machine vision, and in particular relates to a remote three-dimensional scanning method and system based on machine vision. Background technique [0002] Three-dimensional laser scanning technology is a new technology that has appeared in recent years, and it has attracted more and more attention in the research field at home and abroad. With the development of computer graphics, some related technologies have appeared one after another or have made great progress on the original basis. New terms such as virtual reality, digital city, digital museum, and reverse engineering have emerged one after another and have begun to enter the application stage. These technologies are indispensable for 3D geometric modeling, and traditional modeling tools such as AutoCAD, 3DMax, and MAYA have gradually been unable to meet people's modeling needs for complex curved objects, so 3D laser scanning technology emerged as the times ...

AI Technical Summary

Problems solved by technology

[0006] 1. The scanning conditions are harsh. Many 3D scanning devices use the method of active vision to scan, use the geometric information in the lighting source to help extract the geometric information in the scene, and restore the depth information of the object according to the degree of distortion of the light. It is highly dependent on and easily affected by ambient light. It is necessary to strictly control the light environment during scanning to improve scanning quality. For example: the Coment optical camera system produced by Steinbichler, Germany, which projects a series of rasterized light onto the scanned image. To measure the surface of the object, it is necessary to work in a dark environment

[0007] 2. The calculation speed of 3D points is slow. The 3D scanning equipment based on a single image or stereo vision often requires a lot of calculation work. The multiple sets of corresponding object image pairs are obtained, and the disparity of the corresponding points is used to calculate the three-dimensional information of the object. These methods all use parallel binoculars to obtain the three-dimensional information of the object, but it is difficult to ensure the parallelism of the binoculars in practice, and the calculation Algorithms for feature points and matching points are often complex and will take up a lot of time

[0008] 3. The scanning accuracy is low. Most devices are not robust to ambient light, and the devices used also have certain errors. If no hardware or software measures are taken for the system, it is difficult to guarantee the scanning accuracy

[0009] 4. The structure of the device is complex and expensive. Devices with lower environmental requirements and better scanning performance often use additional hardware and software to weaken the influence of ambient light and improve calculation accuracy, which will inevitably make the structure of the 3D scanner complex and expensive. Expensive, such as the 3D scanning equipment proposed by JensSchlemper and others that can be used outdoors, the volume is about 1m*0.5m*0.4m, and the cost is about 800,000 yuan

Images