Research on Feature Point Measurement Technology Based on Stereo Vision

抄録

In view of the high similarity of feature points in low texture environment, this paper proposes an interactive method of manual selection of feature points based on stereo vision under the condition that automatic modeling cannot meet the requirement of 3D environment.Firstly, the model of binocular stereo vision camera with parallel optical axis is designed, then the camera calibration is carried out, and the 3D ranging system of interactive manual selection of feature point pairs is developed.In order to verify the effectiveness of the system, this paper uses corridor floor tiles with fewer texture features to carry out experimental tests. By verifying the three-dimensional coordinates of the measured feature point pairs, and comparing with the actual measured values, it is found that the measurement error is less than 1%.