In order to improve the encoding capacity and decoding accuracy of encoding marker point in largescale photogrammetry

a large-capacity encoding marker point is designed which is composed of a positioning circle

a starting circle and an encoding circle. During decoding

the circles are extracted and distinguished according to roundness and area

and perspective transformation of the encoding marker point is performed to correct the position of the encoding circle. The corrected coded marker point are decoded. Experiments are performed on the object and image plane angles of 0°

45°

60° and outdoor complex scenes respectively

and the decoding accuracy rate can reach to 100 %. The experimental results show that the encoding marker point has the advantages of convenient extraction and recognition

large coding capacity and strong robustness

and provide a new encoding marker point for data splicing in large scale photogrammetry.