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

Deep convolutional neural network end-to-end based image retrieval method by layered deep searching

A convolutional neural network and neural network technology, applied in the field of image retrieval, can solve problems such as low level of automation and intelligence, difficulty in obtaining accurate search results, slow retrieval speed and image retrieval requirements

Active Publication Date: 2016-12-14
汤一平
View PDF8 Cites 98 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0013] In order to overcome the low level of automation and intelligence in the existing image search technology, the lack of deep learning, the difficulty in obtaining accurate search results, the large consumption of storage space in retrieval technology, and the slow retrieval speed, it is difficult to meet the image retrieval requirements in the era of big data, etc. Insufficient, the present invention provides an end-to-end image retrieval method through layered depth search based on a deep convolutional neural network, which can effectively improve the automation and intelligence level of image search, obtain accurate search results, and use relatively Less storage space, faster retrieval speed and slower to meet the image retrieval needs in the era of big data

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
  • Deep convolutional neural network end-to-end based image retrieval method by layered deep searching
  • Deep convolutional neural network end-to-end based image retrieval method by layered deep searching
  • Deep convolutional neural network end-to-end based image retrieval method by layered deep searching

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0143] refer to Figure 1-11 , the technical solution adopted by the present invention to solve its technical problems is:

[0144] An end-to-end approach to image retrieval via hierarchical deep search based on deep convolutional neural networks includes a convolutional neural network for deep learning and trained recognition, a fast visual segmentation algorithm for searching image objects, a method for coarse search A method for fast image alignment using hashing and Hamming distance and an exact alignment method for top-k ranked images based on images from candidate pool P;

[0145] (1) About designing a fast visual segmentation algorithm for searching image objects;

[0146] Since in most applications, the search image object is only a part of the whole image, especially in road monitoring and bayonet image comparison search, it is necessary to design a fast visual segmentation algorithm for search image objects to improve search efficiency ;

[0147] First, a fast vis...

Embodiment 2

[0254] The technology of image search in the present invention has universal applicability, and is suitable for network image search engine, video investigation and bayonet analysis and judgment, as long as the image data participating in the training is operated in the system developed by the present invention to learn, and obtain such objects (such as a specific person, a specific vehicle, etc.) after the characteristics of the search task can be realized.

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 deep convolutional neural network end-to-end based image retrieval method by layered deep searching. The method is characterized by mainly including a convolutional neural network used for deep learning, training and recognition, a rapid visual segmentation algorithm for searching image objects, a quick comparison method used for rough searching with a Hash method and Hamming distance fast images, and a precise comparison method for first k ranked images based on images from a candidate pool P. By the method, automation and intelligence level in searching images with images can be effectively heightened, search results can be acquired accurately, and demand of image retrieval in the big data era is satisfied with less storage space and high retrieval speed.

Description

technical field [0001] The present invention relates to the application of database management, computer vision, image processing, pattern recognition, information retrieval, deep neural network and deep learning technology in the field of image retrieval, especially to an end-to-end layered deep search based on deep convolutional neural network image retrieval method. Background technique [0002] Image retrieval, searching for images by image, is a technology for retrieving similar images by inputting images, and provides users with a search technology for retrieving related graphic and image data. The technology involves many disciplines such as database management, computer vision, image processing, pattern recognition, information retrieval and cognitive psychology. Its related technologies mainly include two key technologies: feature representation and similarity measurement. It is widely used in various fields such as big data graphic image retrieval, video detectio...

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): G06F17/30G06K9/46G06K9/62G06N3/08
CPCG06F16/5838G06F16/5862G06N3/084G06V10/44G06F18/22
Inventor 汤一平
Owner 汤一平
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