Unlock instant, AI-driven research and patent intelligence for your innovation.

Similar picture retrieval method and system

A similar image and retrieval system technology, applied in the field of similar image retrieval methods and systems, can solve the problems of inaccurate retrieval results, high efficiency, and difficult retrieval of images, etc., and achieve the effect of improving efficiency and accuracy

Inactive Publication Date: 2019-08-27
SINOSOFT
View PDF7 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] However, in dealing with complex picture scenes, due to the huge gap between the features and the upper-level semantic information contained in them, it is difficult to retrieve content-based images by applying the above existing methods, so the retrieval results are not accurate and efficient enough.

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
  • Similar picture retrieval method and system
  • Similar picture retrieval method and system
  • Similar picture retrieval method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0024] The following describes the specific implementation of a method for retrieving similar pictures provided by the embodiment of the present invention, see figure 1 , the retrieval method includes:

[0025] S100: collect the original picture, and put the original picture into the retrieval database;

[0026] Specifically, the camera or the network can be used to collect original pictures, and the original pictures can be put into the retrieval database.

[0027] S200: Preprocessing the original image to generate a preprocessed image;

[0028] Specifically, preprocessing the original pictures in the retrieval database to generate preprocessed pictures;

[0029] Wherein, the preprocessing of the original picture includes the following sub-steps S210-S220;

[0030] S210: Perform blurring and denoising on the original picture to obtain a denoised picture, so as to eliminate the interference of details of the original picture on picture retrieval; in this embodiment, Gaussia...

Embodiment 2

[0058] The following describes the specific implementation of a retrieval system for similar pictures provided by the embodiments of the present invention, see figure 2 , the retrieval system includes an acquisition module, a preprocessing module, a training module, a feature vector acquisition module, an index library establishment module and a retrieval module;

[0059] The collection module is used to collect the original picture, and put the original picture into the retrieval database;

[0060] The preprocessing module is used to preprocess the original picture to generate a preprocessed picture;

[0061] The training module trains the pre-processed pictures to obtain a picture classification model, extracts the features of the pre-processed pictures, and vectorizes them to obtain feature vectors; the index library building module imports the feature vectors into the retrieval database and creates Image index library;

[0062] The retrieval module uses the image classi...

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 similar picture retrieval method, which comprises the following steps: collecting an original picture, and putting the original picture into a retrieval database; preprocessing the original picture to generate a preprocessed picture; training the preprocessed picture to obtain a picture classification model, extracting features of the preprocessed picture according to thepicture classification model, and performing vectorization to obtain a feature vector; importing the feature vector into the retrieval database, and establishing a picture index library; and extracting features of a to-be-retrieved picture by using the picture classification model, and performing retrieving in the picture index library to generate a retrieval result. The invention further discloses a similar picture retrieval system. When the to-be-retrieved picture is retrieved, the picture classification model is firstly used for extracting the to-be-retrieved picture to perform feature extraction, so that the accuracy of the feature is improved; in addition, the image index library is used for further indexing the to-be-retrieved image, and the efficiency is higher than that of directly making a full-library feature matching party.

Description

technical field [0001] The invention belongs to the technical field of image retrieval, in particular to a retrieval method and system for similar pictures. Background technique [0002] Since the 1970s, research on image retrieval has begun. At that time, it was mainly based on text-based image retrieval technology (Text-based Image Retrieval, referred to as TBIR), which used text descriptions to describe the characteristics of images, such as paintings. Author, Year, Genre, Size, etc. After the 1990s, image retrieval technology that analyzes and retrieves the content semantics of images, such as image color, texture, layout, etc., has emerged, that is, Content-based Image Retrieval (CBIR) technology. [0003] Most of the existing similar image retrieval methods use traditional image texture, shape, edge and other features for image description. For example: [0004] Phash (perceptual hashing), which describes a class with comparable hash functions, image features are us...

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): G06F16/51G06F16/535G06F16/583G06F16/55
CPCG06F16/51G06F16/535G06F16/55G06F16/583
Inventor 杨浩陈宏江赵全军孙萍
Owner SINOSOFT