Multi-Index Image Retrieval Method Combined with Gist Features
A technology for indexing images and images, applied in digital data information retrieval, special data processing applications, instruments, etc., can solve the problems of low image retrieval accuracy, ignoring the global distribution information of feature points, and single local information.
Inactive Publication Date: 2020-03-31
XIAN UNIV OF TECH
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, as far as the existing technology is concerned, many approximate duplicate image retrieval methods are based on local features and BOF retrieval models, but these methods only use a single local information, ignoring the global distribution information of feature points, resulting in poor image retrieval. Not very accurate
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0063] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0064] The frame diagram of the multi-index image retrieval method combined with GIST features, such as figure 1 As shown, it can be divided into two major systems, namely: offline system and online system; the offline system processes the target image library and generates the two-dimensional inverted table database required for online system query; the online system mainly It is to complete the query process of the query image in the target image library.
[0065] The processing object of the offline system is the reference image library, which performs SIFT feature extraction, GIST feature extraction, feature clustering, and quantization from feature vectors to visual vocabulary for images in the reference image library, and generates visual vocabulary, feature Hamming quantization, and features The inverted index table for use in the onl...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The multi-index image retrieval method combined with GIST features disclosed in the present invention: extract local SIFT features from images in the reference image library, and establish GIST features according to the coordinate information in the local SIFT features; use the local description information in the local SIFT features and the local Establish a two-dimensional BOF feature model based on the description information; establish a two-dimensional BOF-based voting retrieval model based on the two-dimensional BOF feature model; combine the SIFT feature and GIST feature with the two-dimensional inverted list index based on the BOF feature, and compare the SIFT feature and GIST Brightness quantization processing, the local SIFT feature Hamming quantization information obtained by applying the information fusion strategy and the GIST feature Hamming quantization information are fused into the BOF-based voting retrieval model to accurately retrieve approximate duplicate images. The multi-index image retrieval method of the present invention realizes the retrieval of approximately repeated images in large-scale data images.
Description
technical field [0001] The invention belongs to the technical field of image analysis and retrieval methods, and in particular relates to a multi-index image retrieval method combined with GIST features. Background technique [0002] In recent years, technologies such as communication multimedia technology, computer network, large-capacity memory and digital image equipment have been developed and applied rapidly; among them, the application of digital image involves various aspects such as national defense and military, industrial manufacturing, news media and mass entertainment, etc., resulting in A variety of image databases have been established, and the amount of information has grown exponentially. [0003] Nowadays, a large number of images are uploaded and downloaded on the Internet every day, and a large number of near-duplicate images will be stored in the network database. In order to use these large numbers of near-duplicated images in an orderly and rigorous ma...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/583G06K9/46G06K9/62
CPCG06F16/5838G06V10/462G06F18/23213
Inventor 廖开阳汤梓伟郑元林赵凡曹从军蔺广逢袁晖
Owner XIAN UNIV OF TECH