이미지 프로세싱을 이용한 토목 구조물의 정·동적 변위 계측시스템의 개발

Abstract

Civil engineering constructions need to measure their conditions and states using many kinds of contact sensors because they are so huge and change very slowly. The constructions are built for all the citizens, so their conditions and states must be measured carefully. Most of these measurement systems have limitation of various applications. There ate two parts in a measurement system depending on a period time to get the data. The one is a static measurement system and the other is a dynamic measurement system. The static measurement system needs a long period of every one month or half a year for measuring a movement and a coordinate of the constructions because the displacement and the coordinate are changed too small. And the dynamic measurement system obtains data from this system in an instant period time if the constructions shows many kinds of bad symptoms. This system is used for detecting the movements in several minutes and getting a proper vibration of the constructions on a short term because the displacement of huge civil constructions is changed too small. Experts can make an accurate estimate easily using the data with this system. The static measurement system needs just a high - resolution digital camera and measuring program. And the dynamic measurement system consists of a CMOS camera, a telephoto lens, Pan & Tilt equipment and measuring program. These systems are so easy to set up and save a lot of time. In this dissertation, camera calibration uses a well-established Tsai's methods. One of them that are popular among computer vision researchers is mono-view coplanar points which is used for a known target object. The basic geometry of a pin-hole camera model is used for these systems. Static and dynamic developed measurement systems using camera calibration are carried out for some experiments of these measurement systems on a real bridges. Static measurement system has a good experimental data in a suspension bridge. This system compares real displacements of several points at a real bridge with image displacements of the developed static measurement system. Dynamic measurement system has also good experimental results using vision system with one and two targets by long distance. The experimental results in both static measurement system and dynamic measurement system are achieved on the bridges to verify the effectiveness of and its applicability of the proposed vision system in the practical fields.