Abstract: The accurate extraction of the centerline of light stripes is a key issue in achieving three-dimensional measurement in line-structured light three-dimensional measurement systems. This paper starts from the working principle and system composition of line-structured light, introducing the development history of current linestructured light. Secondly, it outlines the existing methods for extracting the centerline of light stripes, and analyzes and compares their principles and key technologies. Finally, it analyzes the advantages and disadvantages of traditional centerline extraction algorithms, and concludes that deep learning is more suitable for industrial inspection applications due to its strong adaptability and high accuracy performance. This paper studies the centerline extraction algorithm of line-structured light stripes, conducts in-depth analysis of existing problems, and prospects for future research directions.