Visual dictionary generation method and device and storage medium

Abstract

The invention discloses a visual dictionary generation method. The method comprises: acquiring at least one piece of feature information corresponding to an environment image in a target positioning area, wherein the feature information comprises related information of local elements in the environment image; constructing a target visual dictionary corresponding to the target positioning area based on at least one piece of feature information corresponding to the environment image, wherein the target visual dictionary at least comprises related information of local elements of a plurality of object images in the target positioning area, and the target visual dictionary is used for performing semantic representation on a to-be-positioned image in a target positioning area. The embodiment ofthe invention also discloses a visual dictionary generation device and a computer readable storage medium.

Description

technical field [0001] The present application relates to but not limited to the technical field of machine navigation, and in particular relates to a method and device for generating a visual dictionary, and a storage medium. Background technique [0002] At present, the relocation method based on Simultaneous localization and mapping (SLAM) uses the bag of words (BoW) model for image representation, and retrieves and analyzes the image based on the BoW model to obtain the accurate location and location of the machine. attitude. Usually, when using the BoW model to represent an image, it is first necessary to construct an image-based visual dictionary, which is composed of independent underlying image features; and then perform image analysis and understanding according to the different distributions of images in the visual dictionary. [0003] However, in the related art, the visual dictionary is generated by training a large number of image data sets, and the image featu...

AI Technical Summary

Problems solved by technology

[0003] However, in the related art, the visual dictionary is generated by training a large number of image data sets, and the image features in the visual dictionary are fixed and cannot change with the environment; therefore, the visual dictionary established in the prior art will cause Inconsistencies in image descriptions in different scenarios, for example, the accuracy of describing an image using the above-mentioned visual dictionary in one scenario may be 95%, but in another scenario the accuracy rate may only be 80%

Therefore, the visual dictionary in the prior art does not have the adaptability to different scenarios, and the consistency of descriptions in different scenarios will also be affected

Images