Under the Background of Artificial Intelligence, Image Local Feature Search, Matching and Tracking Algorithm

Xiaotao Ma

Hebei Software Institute, Baoding, 071000, China

Abstract: The traditional image local feature matching and tracking algorithm has poor robustness due to scale change and anti-noise. Therefore, this paper proposes an image local feature optimization matching and tracking algorithm under the background of artificial intelligence. A small number of feature points containing important structural information are used to determine the relationship between images. The algorithm extracts and describes the feature points of the two images, and then builds a Kd tree to retrieve the feature point data, and finally gets a more accurate feature point matching pair. Simulation results show that the algorithm can effectively reduce the computational cost of image processing and improve the anti-noise performance of the algorithm.

Keywords: Artificial intelligence; Image features; Tracking algorithm