Grey correlation analysis method for monocular vision pose measurement error

Aiming at the impact of multiple factors on the accuracy of monocular visual pose measurement system, grey correlation method is used to analyze the pose measurement errors of 11 independent variable parameters, including radial distortion error, tangential distortion error, image point error, normalized focal length error and center point error. The main factors affecting the pose accuracy are obtained, which proves the effectiveness of the grey correlation method for visual system error analysis. The calculation results show that the comprehensive correlation degree between absolute error sequence and zero sequence caused by five types of parameters of radial distortion, tangential distortion, image point coordinate error, normalized focal length error and center point error is less than 0.56, which is significantly lower than other parameters, indicating that its influence on measurement accuracy is significantly higher than other parameters.