Abstract: As a key technology in the field of intelligent drilling equipment, automatic drill pipe loading and unloading technology restricts the automation and intelligent development of underground drilling equipment in coal mines. The existing drill pipe automatic loading and unloading system mainly relies on the mechanical structure and proximity switches for positioning, which has the problem of poor positioning accuracy and low automation. To solve this problem, a drill pipe pose recognition algorithm based on monocular vision technology was proposed. The camera was used to capture the image containing the cooperative target, and the relative distance and posture between the camera and the cooperative target were calculated; the position and posture of the drill pipe relative to the manipulator was deduced through fixed coordinate transformation, and the manipulator was guided to automatically load and unload the drill pipe. First, we determine the overall scheme of the system and then establish a mathematical model of camera imaging using the principle of pinhole imaging and Zhang Zhengyou’s calibration method, so as to solve the internal and external parameters of the camera. Secondly, using the checkerboard calibration plate as the cooperation target of the drill pipe to be tested, a monocular ranging model of any plane in space was established according to the small hole imaging model and the spatial imaging relationship, and the distance between the optical center of the camera and the cooperative target point was calculated. Finally, the attitude matrix of the cooperative target was obtained through the camera imaging model. Combined with the internal and external parameters of the camera, the coordinate transformation was used to obtain the attitude matrix of the cooperative target in the world coordinate system, and then the position and attitude recognition of the drill pipe was completed through fixed coordinate transformation. To verify the accuracy of the algorithm, the drill pipe pose recognition test was carried out indoors. In the test, the repetitive distance measurement and attitude estimation were carried out for each on-site picture. The results show that the drill pipe distance recognition deviation is within 0.12%, and the drill pipe attitude recognition deviation is within 1.08%, which meets the precision requirements of automatic loading and unloading of drill pipes. The results also show that the drill pipe pose recognition algorithm based on monocular vision technology is real and effective. The algorithm can realize the intelligent recognition of drill pipe positioning, improve the automatic loading and unloading accuracy of drill pipe and the intelligent level of drilling equipment.