In traditional calibration methods for line-structured light vision sensors, the use of two-dimensional planar targets usually requires multiple placements of the target in the measurement space to capture more images to improve the final calibration accuracy, which complicates the calibration process. We propose a new calibration method for line structured light vision sensors (LSLVS) based on freely placeable circular targets. The circular target only needs to be placed freely twice, and the camera captures the images of the circular target and the optical strip. Based on the geometrical relationship of the circle under the camera projection and the nature of the vanishing line, the target plane is first reconstructed, and then multiple three-dimensional points on the target plane are obtained according to the perspective projection relationship, and finally the reconstruction of the optical plane is completed. Compared with the traditional calibration method, this method can be applied to measurements in a limited space, and it does not need sophisticated auxiliary equipment to control the target motion during the whole calibration process, which makes the calibration process more simplified. The final experimental results show that the method can reach an accuracy of 0.0523 mm when only two images are used and can be effectively applied to the line structured light measurement system.