Image text semantic extraction method based on GPU (Graphics Processing Unit)
An extraction method and image technology, which are applied in the fields of instrumentation, calculation, electrical digital data processing, etc., can solve the problems of incompatibility of efficiency and accuracy, long data update cycle of image search system, and low accuracy of textual description of images, etc. Achieve the effect of avoiding redundant parsing, improving scalability, and improving execution efficiency
Inactive Publication Date: 2013-01-02
HUAZHONG UNIV OF SCI & TECH
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] In view of this, the object of the present invention is to provide a GPU-based image text semantic extraction method, which solves the incompatibility of efficiency and precision in existing image text semantic extraction methods, and the low accuracy of literal description of images , Image search system background data update cycle is long
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
[0075] The present invention will be further described below in conjunction with the accompanying drawings.
[0076] Such as figure 1 Shown, the image text semantic extraction method based on GPU of the present invention comprises the following steps:
[0077] (1) Obtain a collection of web pages;
[0078] (2) Establishing a document object model (document object model, DOM for short) collection according to the webpage collection, which includes a plurality of text nodes and image link nodes;
[0079] (3) determine the collection of image link nodes and the quantity of image link nodes according to the label of image link nodes;
[0080] (4) judge whether the quantity of image link node is 0; If not 0, then proceed to step (5), if it is 0, then proceed to step (17);
[0081] (5) Take out an image link node, and set counter i=0;
[0082] (6) Find whether there is a text node around the image link node according to empirical rules; if it does not exist, then proceed to step...
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
An image text semantic extraction method based on a GPU (Graphics Processing Unit) comprises the following steps of: acquiring a webpage collection; establishing a document object model collection according to the webpage collection, wherein the document object model collection comprises a plurality of text nodes and image link nodes; determining a collection of the image link nodes and a number of the image link nodes according to a label of the image link nodes; judging whether the number of the image link nodes is 0; if the number of the image link nodes is not 0, taking out an image link node, and setting a counter to be i=0; and searching whether the text nodes exist around the image link nodes according to thumb rules. The invention solves the problems that in an existing image textsemantic extraction method, efficiency and precision cannot be both obtained, the text description precision of an image is low, and an update cycle of background data of an image searching system islong.
Description
technical field [0001] The invention relates to a GPU-based image text semantic extraction method. Background technique [0002] With the enrichment of network multimedia, a large amount of image data and video information are generated on the network every day, and image, as one of the most basic and important forms of multimedia information, has become more and more It is more and more widely used in various fields. The image search engine collects, organizes, and indexes images in the network, and the user enters keywords or sample pictures, and the system returns the images that the user is interested in in the form of a result set. [0003] At present, the mainstream commercial image search engines are mainly based on text, and the text-based image retrieval technology mainly uses manual annotation or extracts text keywords from webpages to realize text description of images. Image text semantic extraction in image search system is a method to extract textual semantic...
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): G06F17/30
Inventor 金海郑然江武邓巍
Owner HUAZHONG UNIV OF SCI & TECH