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

Picture classification method and device and electronic equipment

A picture classification and picture technology, applied in the computer field, can solve the problems of low reference value for users, inability to obtain business information, cluttered display of pictures, etc., and achieve the effect of accurate classification

Pending Publication Date: 2021-01-05
BEIJING SANKUAI ONLINE TECH CO LTD
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in the prior art, the above-mentioned pictures of users or merchants on the network platform are not classified, stored, managed or used, resulting in, for example, messy display of pictures, which is of little reference value to users
The platform is also unable to obtain accurate business information from the above pictures

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
  • Picture classification method and device and electronic equipment
  • Picture classification method and device and electronic equipment
  • Picture classification method and device and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027] A picture classification method disclosed in the embodiment of the present application, such as figure 1 As shown, the method includes: Step 110 to Step 130.

[0028] Step 110, acquire the target picture, and the label text and description text of the target picture.

[0029] The picture classification method disclosed in the embodiment of the present application is applicable to the classification scene of pictures uploaded to the network platform by merchants or users, such as environmental pictures, product pictures, license pictures, etc. pictures etc.

[0030] The online platforms of different business lines will have different contents of uploaded pictures, and the tags and descriptions set by users or merchants for pictures will also be different. For example, for the training business line, the pictures uploaded to the network platform usually include: classroom records, homework results, classroom environment, etc.; for the catering business line, the picture...

Embodiment 2

[0071] A picture classification device disclosed in the embodiment of the present application, such as Figure 4 As shown, the device includes:

[0072] A picture information acquisition module 410, configured to acquire a target picture, as well as label text and description text of the target picture;

[0073] The feature encoding module 420 is configured to determine the vector representation of the picture content of the target picture based on a spatial attention mechanism, and determine the vector representations of the label text and the description text, respectively;

[0074] The classification processing module 430 is configured to perform classification processing on the target picture based on the vector representation of the picture content, the label text, and the vector representation of the description text.

[0075] When classifying pictures online, learn the importance weight of the region position in the picture through the spatial attention mechanism, enco...

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 picture classification method, belongs to the technical field of computers, and is helpful for accurately classifying pictures uploaded to a network platform. The picture classification method disclosed by the embodiment of the invention comprises the steps of obtaining a target picture, and a label text and a description text of the target picture; determining vector representation of picture content of the target picture based on a spatial attention mechanism, and determining vector representation of the label text and the description text respectively; and performing classification processing on the target picture based on the vector representation of the picture content, the label text and the vector representation of the description text. According to the method, the pictures are automatically classified by combining the picture content, the label text and the description text, so that the picture classification speed and accuracy can be improved; and according to the method, the important areas in the images are learned based on the spatial attention mechanism, the input image data are enhanced and then used for image classification, and the image classification accuracy is further improved.

Description

technical field [0001] The embodiments of the present application relate to the field of computer technologies, and in particular, relate to a picture classification method, device, electronic equipment, and computer-readable storage medium. Background technique [0002] In applications such as reviews or online shopping, merchants and users upload a large number of pictures to display products or reviews. And these comments contain a lot of useful information for merchants or users. For example, the platform collects information such as user satisfaction, business environment, and geographical location of a business by analyzing pictures uploaded by users about a certain business, so as to obtain more business data. For another example, the platform displays richer business information to the search terminal by analyzing the pictures uploaded by users about a certain business. However, in the prior art, the above-mentioned pictures of users or merchants in the network pla...

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): G06Q30/02G06Q30/06G06F16/55G06F16/58
CPCG06Q30/0201G06Q30/0643G06F16/55G06F16/5866
Inventor 李楠宋祺张睿
Owner BEIJING SANKUAI ONLINE TECH CO LTD