In the actual construction process of shield machine, in order to solve the problem that when the non-contact measurement is used to detect the segment state, the segment end face has more interference, and the traditional light strip center point coordinate extraction algorithm is easy to be disturbed and fluctuated, an optimization algorithm for light strip center point coordinate extraction is proposed. Firstly, the image is preprocessed by adaptive threshold binarization and morphological processing to eliminate some interference; secondly, the weighted gray barycenter method is used to extract the coordinates of the initial light stripe center point, which is taken as the optimization object. The distance threshold and cubic spline interpolation method are used to preliminarily optimize and secondary optimize the coordinates, and the final optimized line structured light center point is obtained.The test results show that the algorithm not only improves the extraction accuracy of the center point coordinate, but also meets the industrial requirements with high real-time performance, which lays a good foundation for subsequent detection of the ellipticity of the segment and capturing the relative shield tunneling attitude of the segment.