Unmanned aerial vehicle image matching pair selection method and system based on vocabulary tree retrieval

Abstract

The invention discloses an unmanned aerial vehicle image matching pair selection method and system based on vocabulary tree retrieval, and the method comprises the steps: firstly carrying out SIFT feature point extraction of each image in an image subset obtained through sampling through employing a random sampling strategy and an SIFT algorithm; secondly, for extracted SIFT feature points, constructing an initial feature set, sorting all SIFT feature points in the initial feature set from large to small according to scales, selecting the first h feature points in the sorted set, and constructing a feature subset about the h feature points; thirdly, for the obtained feature subset, constructing a vocabulary tree for unmanned aerial vehicle image retrieval by adopting a hierarchical K-meansclustering algorithm; wherein the vocabulary tree is used for establishing an unmanned aerial vehicle image index; and finally, carrying out unmanned aerial vehicle image matching pair selection through an adaptive similar image quantity selection algorithm by utilizing the established unmanned aerial vehicle image index.

Description

technical field [0001] The invention relates to the fields of photogrammetry and computer vision, and proposes a method and system for selecting matching pairs of UAV images based on adaptive threshold vocabulary tree retrieval. Background technique [0002] Image matching is a core research content in the field of photogrammetry and computer vision. For UAV images, due to the low flying height of the UAV platform, the small size of the non-measurement camera used, and the simultaneous shooting of multi-angle cameras, the UAV image has a large amount of data and high resolution. features. Exhaustive matching modes will result in high computational cost for image matching. Therefore, before the actual image matching, selecting overlapping image matching pairs is the key technology to achieve image matching acceleration. [0003] At present, the commonly used image matching pair selection methods can be divided into two categories. The first category is matching pair selec...

AI Technical Summary

Problems solved by technology

However, image retrieval based on vocabulary trees needs to consider the balance between accuracy and efficiency: if a pre-established vocabulary tree is used, an incomplete vocabulary tree will lead to a decrease in retrieval accuracy; if a corresponding vocabulary tree is established for each data set, a large amount of data will significantly increase the overall time consumption

This will result in too many or not enough "similar" images being retrieved

Images