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

Rapid image retrieval method based on supervised topology keeping hash

An image retrieval, supervised technology, applied in computer vision, pattern recognition, can solve problems such as limited intra-class differential expression ability

Active Publication Date: 2015-11-18
天津中科智能识别有限公司
View PDF1 Cites 58 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to propose a fast image retrieval method based on supervised topology preserving hashing to solve the problem that traditional supervised hashing methods have limited ability to express intra-class differences

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
  • Rapid image retrieval method based on supervised topology keeping hash
  • Rapid image retrieval method based on supervised topology keeping hash
  • Rapid image retrieval method based on supervised topology keeping hash

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0071] In order to describe the specific implementation of the present invention and verify the effectiveness of the present invention, we apply the method proposed in the present invention to a public image, that is, CIFAR-10. The database contains 60k 32*32 color images, a total of 10 categories, and each category has 6000 pictures. In our example, we randomly sample 100 images from each category as query set images, and another 500 randomly sampled images as training set images. figure 2 is a sample of the database image. We first extract the 512-dimensional GIST features of the training set and query set data, and randomly select 1000 anchor points to perform Gaussian kernel mapping on the features.

[0072] Following step S2 in the technical details presented earlier, after randomly initializing the binary code B, the model is trained. The weight parameters α, β, and γ are 1e-3, 1e-1, and 1, respectively, during model training. After the training is completed, the has...

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 a rapid image retrieval method based on supervised topology keeping hash. The method includes the steps of S1, extracting features of obtained training images and inquiry images, converting feature spaces into new nuclear spaces, and obtaining nuclear space representation of each image; S3, conducting binary coding on the training images and the inquiry images; S4, conducting image retrieval through binary codes. For solving the rapid image retrieval problem, hash coding is studied in the nuclear space with higher expression capacity, supervise information and topology keeping information are added in the hash mapping matrix studying process, a studied mapping matrix has higher semantic expression capacity and higher within-cluster variation expression capacity, and therefore the studied binary codes are more suitable for image retrieval tasks, retrieval accuracy is improved, and retrieval result sequencing is optimized.

Description

technical field [0001] The present invention relates to computer vision, pattern recognition, machine learning and other technical fields, in particular to a fast image retrieval method based on supervised topology preserving hashing (Supervised Topology Preserving Hashing, referred to as STPH). Background technique [0002] Image, text, video and audio data in the current network are emerging at an exponential rate, how to quickly and effectively find the information we need has become a problem that people are increasingly concerned about. Information retrieval, especially image retrieval is a relatively complicated process. Existing retrieval methods usually need to first express the image with high-level features, and then judge whether it is a similar image according to the similarity comparison of the feature vectors. How to perform effective feature expression and efficient similarity comparison are two research focuses of image retrieval. The invention mainly solve...

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
IPC IPC(8): G06F17/30G06K9/62
CPCG06F16/50G06F18/2413
Inventor 赫然谭铁牛孙哲南张树张曼
Owner 天津中科智能识别有限公司
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