The image total station cannot achieve the fully automated measurement of the target point in the prism-free cooperative working mode. To address this issue, a fully automated measurement method of image total station based on the improved YOLOv5 algorithm is proposed. The YOLOv5 algorithm fused with the convolutional block attention module is used to realize the wide-angle camera identification and detection of the reflector. And the target automatic aiming algorithm is applied to realize the accurate aiming of the telephoto camera at the center of the reflector, which realizes the fully automated measurement of the position coordinates of the target point. With the help of the self-developed image total station, the identification and detection experiment of the reflector and the fully automated measurement experiment of the target point are carried out. Experimental results show that the accuracy of identifying and detecting reflector targets by the improved YOLOv5 algorithm can reach 98. 65% . Compared with manual photometric measurement method, the fully automated measurement method of target point has comparable measurement accuracy and increases the measurement efficiency by 1. 5 times. The proposed method has high measurement accuracy and measurement efficiency, which can be widely used in the unattended and fully automated measurement work occasions.