A Focal Length Calibration Method for Vision Measurement Systems Based on Multi-Feature Composite Variable Weighting

Existing focal length calibration methods rely on predefined calibration fields or control point networks, which are unsuitable for real-time applications with variable zoom in industrial and agricultural environments. This paper proposes a method based on global scanning principles and geometric constraints, eliminating control points and using symmetric features. A spatial weighting strategy optimizes redundant measurements by integrating optical distortion and the spatial distribution of measured points, enhancing accuracy. Experimental results show that the method achieves micron-level calibration precision, significantly improving visual measurement system accuracy under complex zoom conditions.