Wireless Binocular Stereovision Measurement System Based on Improved Coarse-to-Fine Matching Algorithm
Cheng Wang,Teng Xiao,Zhiqun Gong,Shiting Yang,Dongliang Zhang,Fei Deng
DOI: https://doi.org/10.1155/2023/1132569
2023-02-09
Structural Control and Health Monitoring
Abstract:To cope with the challenges some commonly used displacement sensors may face in structural health monitoring, we proposed a wireless binocular stereovision measurement system based on the improved coarse-to-fine matching algorithm. The vision measurement system can perceive multipoint three-dimensional displacement and consists of two cameras, one remote control unit, and a solar panel. The improved coarse-to-fine matching algorithm only requires a flat surface with some painted textures instead of designed markers with refined patterns (circle, square, and chessboard). Omitting the textures keeps a balance between measurement accuracy and the burden of the target installation in the actual field of structural health monitoring. The proposed coarse-to-fine matching algorithm utilizes scale-invariant feature transform (SIFT) points to generate a robust initial guess and combines improved least squares matching (LSM) algorithm for advanced corresponding point matching. A precalculated look-up table for interpolation and parallel computation is used for higher matching precision and computational efficiency. Several experiments were conducted to present the accuracy and feasibility of the proposed metering system. A simulation test, an indoor shaking table test with several commonly used displacement sensors, and an outdoor experiment consider the natural condition. Experimental results verified that the proposed matching methodology could achieve 0.2-pixel matching accuracy, 0.1 mm measuring precision in a laboratory, and millimeter-level precision in real conditions.
construction & building technology,engineering, civil,instruments & instrumentation