In order to realize the high-precision calibration of the line-structured light vision measurement system, in view of the shortcomings of the traditional calibration method that the number of feature points is small and the calibration accuracy is affected by the grayscale extraction accuracy of the structured light to a certain extent, a two-dimensional plane-based method is designed. Target Improved Line Structured Light Algorithm. On the basis of the cross-ratio invariance theorem, the algorithm obtains a large number of characteristic data points through a simple checkerboard target, and then obtains the initial light plane equation through the least squares method. According to the characteristic that the center point of the structured light is on the light plane and on the plane of the checkerboard calibration board, the abnormal data points with large errors are screened out, and finally the light plane equation is optimized and fitted by the screened data points. In the general experimental environment, the results of the optical plane calibration comparison test show that the improved calibration method has a very significant improvement in measurement accuracy.