A new binocular camera calibration method based on a directional target and multi-constraint optimization is proposed. The novel directional planar target is equipped for determining the target rotation direction and assigning a code number to each calibration corner to ensure the matching of homonymous corners even in the case that the target is partially occluded and thereby achieve binocular camera calibration. After building a model of binocular camera calibration with a directional target, we introduce zenith and azimuth angles to describe the orientation of the planar target and select images in which the target has markedly different orientations at different positions as calibration images, so as to raise the stability of the binocular calibration results. To improve the accuracy of the binocular calibration results, we make full use of the 3D geometric information on the target and propose a multi-dimensional constraint optimization model for binocular parameters. The experimental results show that our method can effectively improve the stability and accuracy of the calibration results compared with Zhang’s calibration method. Multiple measurements of standard gauge blocks further verify the effectiveness of the proposed method.