公路结构光视觉检测系统中标定技术的研究仪器科学与技术专业论文.docx

哈尔滨工业大学工学硕士学位论文 哈尔滨工业大学工学硕士学位论文 - - II - Abstract The detection of pavement surface cracking is crucial for maintenance of pavements. Traditional manpower-based methods have many defects, such as time consuming, non-precision and so on. They cannot fulfill the requirements of the large quantity of pavements nowadays. With the developments of optical, computer and sensor technologies as well as the theory of digital image processing, it is necessary to propose a fast method to improve the detection efficiency. Approaches based on computer vision have become the hotspot in this field all over the world. The main purpose of computer vision is to resume 2D image to 3D geometry structure. Structure-light calibration is the most important step in 3D- Reconstruction, which also is the basement of obtaining 3-D space information by computer vision, and it becomes a significant field of computer vision. Based on the Marr’s computer vision principle, the main work of this paper is to research the technology and the method of the Structure-Light calibration, and then obtain the parameters of camera, followed by establish the relationship between 3D space and 2D image provide reliable data and finally lay a good foundation for next step of computer vision. Tasks which had been done in this article are as follows: Firstly, research image feature point extraction issues by the numbers. In the process of corner detection, the element, the methods and the capability of Harris operator are analyzed and contrasted, and then using VC++6.0 to complete different images’ Harris corner detection. The detection results turns out that the speed of Harris corner detection method is fast and this method can get even and accurate corners. So this algorithm is stable. Secondly, the principle and the state-of-art of calibration technology based on computer vision at home and abroad are elaborated. After that, the existing calibration technology and the difficulties in calibration technology at