Applying Computer Vision Techniques for Inclination Monitoring in Drainage Gallery

• Main Authors: Bo-Yan Wang, 汪柏諺
• Other Authors: 蘇苗彬
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/30421298975082369636

碩士 === 國立中興大學 === 土木工程學系所 === 105 === The research analyzes the relationship between the position of coordinate and the displacement by applying the coordinate of surveyor’s beacon of the image on the horizontal tilt meter OT-1401. In order to measure the displacement after the inclination of the horizontal tilt meter, the Raspberry Pi microcomputer camera and Open CV procedure as well as the Chessboard shoot were utilized in the study. The research made use of the Raspberry Pi microcomputer camera to take photo of the Chessboard’s beacon. In the image, there are nine coordinate points which will change if the horizontal tilt meter inclines. Using the relationship of the pixel and the distance, we found the difference between the angle of inclination read from the image and the actual one. At the same time, we verified the inclination deflection of the computer vision with the help of the SCA121T-D07 tilt meter. According to the laboratory experiment, the carrier and camera used can read the angle within two degrees and can reach 0.0333 degree by minimum. By analyzing the one-month spot data in the outdoor experiment, we found some small change which was tiny but still did not arrive at the minimum range of measurement (approximately 0.0333 degree). As a whole, the reading of both the X and Y coordinate were stable so the inclination deflection instrument of the computer vision can be applied to the deformation monitoring of the drainage gallery