Multi-scale analysis based image feature bag constructing method

Abstract

The invention provides a multi-scale analysis based image feature bag constructing method through introducing a multi-scale analysis concept of images into a feature bag model. The method comprises the steps of firstly carrying out decomposition on an image by using wavelet transformation, then respectively extracting local area features of a high-frequency sub-band and a low-frequency sub-band of the image, constructing a high-frequency visual dictionary and a low-frequency visual dictionary respectively, then describing the image by using the visual dictionaries, and generating an image feature bag. The method provided by the invention focuses on the level of multi-scale feature extraction and semantic description of the image, detail information in the image can be better captured so as to generate visual feature vocabularies, and the new feature bag model can be specifically applied to classification, retrieval and the like of digital image data such as medical images, remote sensing images, network images and the like.

Description

technical field [0001] The invention belongs to the field of computer vision, and more specifically relates to a feature bag construction method based on multi-scale analysis. Background technique [0002] With the progress of the information age, digital images, as an important carrier of information, have shown explosive growth in number. While massive digital image data brings great convenience to people's information collection, transmission and acquisition, it also makes the classification, storage and retrieval of information face many new difficulties. How to analyze, identify and obtain useful information more quickly and accurately from the vast image data is one of the most important research topics in the field of computer vision. [0003] The early image classification and identification were mainly done manually, using the text information attached to the image for classification and retrieval. However, with the geometric increase in the number of images and th...

AI Technical Summary

Problems solved by technology

[0004] Local feature extraction and visual dictionary establishment are the two most critical steps in the feature bag model. Currently, the most commonly used local features include features based on image blocks (Patches) and features based on key points (Key Points). Among them, SIFT The local features based on key points are more popular, but recently, with the improvement of computer performance, new research believes that the features based on image blocks can have the same classification effect as SIFT, and there is more room for further improvement.

The problem with image block-based features is mainly the lack of representation of image details and textures

Images