Study On Deformation Measurement Of Long Span Bridge With Self-calibration Image Technology

Deformation is one of the most important parameters describing bridge's safety and health. So deformation measurement is the key of bridges safety and health monitoring. Large deformation of long span bridges demands measurement technology with large range. GPS(Global Position System) technology, total station technology and image technology can be used to automatically measure the large deformation of long span bridges. But image technology with low cost is the most economical method. In this dissertation the importance of deformation measurement is analyzed. The principle of deformation measurement of long span bridge image technology is explained. A novel self-calibration method is proposed. In complex background, object recognition and location algorithm with high stability and high accuracy is the key of deformation measurement. Sub-pixel location technology is research. Mode recognition can't be used to recognize because of unstable shape parameters. The recognition and location algorithms integrating gray gravity center with template match is proposed.The software of deformation measurement system is developed to meet different demands. Function system is designed. Main function module and child function module is introduced. Based on direct memory access and asynchronous buffering method of Matrox imaging card, real-time grabbing and processing are realized. To test and verify the feasibility and performance of deformation measurement system with self-calibration image, static experiment and zero drift experiment are carried out indoor and outdoor. Error source is analyzed. Data are processed. Results show that system can dynamically measure the large deformation of long span bridges in low cost. The system can be used to measurement not only bridges deformation but also large structures' deformation.