Application of Artificial Intelligence-Based UAV Photogrammetry Technology in Electric Power Surveying and Mapping Engineering

With the rapid development of science and technology and the rapid development of artificial intelligence, UAV aerial survey technology has also developed rapidly. UAV technology and image fusion and many other technologies, not only to FIG faster, but also a high accuracy mapping, can be applied to a variety of complex terrain. In this paper, a small UAV aerial view of discrete image, change the viewing angle characteristics of the SIFT algorithm has been improved in order to improve image registration rate. In this paper, Harris feature points and SIFT feature vectors are used to extract image features, PCA is used to reduce the dimension of feature vectors, BBF algorithm improves the matching speed, and PROSAC algorithm refines the feature points to achieve automatic image registration. Through the detection of 12 aerial images of unmanned aerial vehicles, the matching rate of the algorithm proposed in this paper reaches 98.55%; A topographic map in a range of plant 1.2 km² the accuracy of analysis, the method of the large scale topographic map has good effect in complex terrain conditions. The research in this paper effectively improves the overall level of application of UAV aerial survey technology in power surveying and mapping.