Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Image retrieval method based on edge direction difference characteristic bag

An edge direction and image retrieval technology, applied in special data processing applications, instruments, electrical digital data processing, etc., can solve the problem of image retrieval accuracy, low callback rate and speed, poor robustness of retrieval efficiency, image retrieval speed and efficiency Lowering and other issues

Active Publication Date: 2013-11-20
XIDIAN UNIV
View PDF2 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although this method uses the visual word co-occurrence table and accurate mapping to extract the optimal visual word representation image, which improves the accuracy of image retrieval, but when applied to a large image database, the number of visual words increases rapidly, and then the application of high-order probability predictor It will greatly increase the computational complexity of time, resulting in a decrease in the speed and efficiency of image retrieval
[0005] The above two documents are similar to the present invention in the technical field, but the common problem is that the accuracy rate, callback rate and speed of image retrieval are low, and the robustness of retrieval efficiency is poor when applied to large image databases

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Image retrieval method based on edge direction difference characteristic bag
  • Image retrieval method based on edge direction difference characteristic bag
  • Image retrieval method based on edge direction difference characteristic bag

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] The invention is an image retrieval method based on edge direction difference feature bags. With the continuous development of multimedia technology and the continuous popularization of the Internet, the application of image information is becoming more and more extensive. The application of image information often requires image retrieval, such as digital library, public safety and crime investigation, through image retrieval, people can quickly and effectively query the information they need. The application of image retrieval is integrated into people's daily life, which brings convenience to people. Image retrieval has more and more broad application prospects.

[0047] Reference to the implementation of the image retrieval method based on edge direction difference feature bags of the present invention figure 1 , give the following specific examples:

[0048] Step 1: Input the retrieved color image, which is the image to be queried. The retrieved image can be an im...

Embodiment 2

[0084] Embodiment 2 The image retrieval method based on edge direction difference feature bag is the same as embodiment 1

[0085] Step 1, input the retrieved color image;

[0086] This example inputs a search image randomly selected in the Corel-1000 image database, see image 3 , it is necessary to retrieve images of the same type in the Corel-1000 image database, which includes 10 types of images, see figure 2 , each category includes 100 images, and some examples of each category such as figure 2 As shown, the retrieved image used in this example is image 3 shown.

[0087] Step 2: Perform grayscale transformation on the input retrieved image, process it through a direction-tunable filter, select a two-dimensional Gaussian function as the filter kernel function, and calculate the directional derivative of the image in the X and Y directions with the first-order Gaussian kernel function The convolution of each pixel gets the energy function W in 2L directions σ (x, y...

Embodiment 3

[0099] Embodiment 3 The image retrieval method based on edge direction difference feature bags is the same as that in Embodiment 1-2

[0100] This example also selects the Corel-1000 image database, which includes 10 types of images, and some examples of each type are as follows figure 2 As shown, each class includes 100 images, and the same retrieval process of embodiment 1 is performed on each image in the database, and when the value n of the number of retrieved images returned is 10, 20, ..., 100, all images in the image database are calculated. The average retrieval accuracy rate and the average retrieval recall rate of 1000 images were drawn and compared with several well-known retrieval methods in the prior art such as the methods proposed by Elami, Jhanwar, Hung, Chuen and the CSD method , the comparison curve of the average retrieval accuracy is shown in Figure 6 As shown, the comparison curve of the average retrieval callback rate is as follows Figure 7 shown. ...

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

No PUM Login to View More

Abstract

The invention discloses an image retrieval method based on an edge direction difference characteristic bag. Images to be retrieved are input at first; the images are processed by a direction adjustable filter; image edge pixels are extracted combined with the result of the direction filter; direction difference characteristics of all the edge pixels of the images are extracted; a training image is selected from the images to be retrieved randomly and the edge pixel direction difference characteristics are extracted; a characteristic bag dictionary is constructed with the training image direction difference characteristic in a clustered mode; edge pixel point direction difference characteristics of the image to be retrieved are also extracted; the retrieval image and the image to be retrieved are both based on the characteristic dictionary to extract encoding histogram characteristics; encoding histogram similarity matching is carried out on the retrieval image and the image to be retrieved; an image retrieved result is displayed according to the similarity matching value. The image retrieval method based on the edge direction difference characteristic bag is high in retrieving speed, high in accuracy and call-back rate, is superior in retrieval of large-sized image data, and can be applied to image retrieval of real-time human-machine interaction and a large-sized image database.

Description

technical field [0001] The invention belongs to the technical field of image processing, and relates to a content-based image retrieval method, in particular to an image retrieval method based on an edge direction difference feature bag, which is applied to the fields of real-time human-computer interaction, image retrieval and classification, and the like. Background technique [0002] In recent years, the development of computer multimedia technology has advanced by leaps and bounds. The multimedia information system has surpassed the traditional database system. It integrates various non-text data, such as digitized sound, images and video images, into the system, processes them with computers, and Through computer network transmission, it greatly facilitates people's demand for multi-directional and multi-level information. Then, how to perform more direct and fast retrieval based on the visual features of images has become an important issue in the information field. T...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/30
Inventor 田小林焦李成刘宪龙王爽马文萍马晶晶刘燕张小华
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com