Feature Point Identification of Hydropower Plant Based on Improved ORB Algorithm

• Published in: 2023 5th International Conference on Intelligent Control, Measurement and Signal Processing (ICMSP) pp. 1108 - 1112
• Main Authors: Xiu, Ji, Hou, Faming, Li, Dexin, Guo, Shuanghao, Bai, Yang, Zhang, Jiajun
• Format: Conference Proceeding
• Language: English
• Published: IEEE 19.05.2023
• Subjects: Cosine similarity; Feature extraction; Hamming distances; Hydroelectric power generation; Hydropower station; Improve ORB; Intelligent control; Mismatching rate; Signal processing; Signal processing algorithms
• DOI: 10.1109/ICMSP58539.2023.10169144

The traditional ORB algorithm has some problems such as poor feature point recognition, difficult to eliminate mismatching and scale invariance. This paper presents an improved ORB algorithm to identify and match feature points of hydropower plant. First of all, FAST algorithm uses rBRIEF descriptor to describe the detected feature points in the images of hydropower station equipment; Secondly, the feature points extracted by ORB were constructed by Hessian matrix to extract the feature points of hydropower equipment images. Finally, the direction consistency of feature points is used to eliminate the feature points that are not within the threshold range by cosine similarity invariance. The algorithm improves the accuracy of feature point matching, eliminates mismatching feature points and has the advantages of fast matching speed.