Abstract: The artificial measurement of elevator guide rails is labor intensity,low efficiency and low precision.In order to solve the problem,designed a set of elevator guide rails automatic measurement system based on binocular vision.The system consists of the hardware system and the software system.Aiming at the problems of low contrast and less feature points of the surface of the elevator guide rails,the laser grid is projected onto the surface of the workpiece,so that the surface can be determined with identifying characteristics.It uses improved RANSAC algorithm to extract the laser grid.And then it uses SURF algorithm to extract and match the image feature points,and calculates the spatial coordinates of the feature points based on the camera calibration result.Finally,it calculates the flatness,parallelism and perpendicularity of the elevator guide rails according to the spatial coordinates of the feature points.Experimental results show that the system is reasonable,good software stability and high efficiency.It improves the efficiency and accuracy of the measurement of elevator guide rails and has a very good practical value.

Key words: binocular vision, elevator guide rail, feature extraction, image matching