Accessory identification method and device

An identification method and accessory technology, applied in the field of intelligent identification, can solve the problem of inability to save it, and achieve the effect of good performance and accurate results.

Inactive Publication Date: 2017-10-13
EMOTIBOT TECH LTD
View PDF5 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The core concept of this technology is to divide object detection and object recognition into two stages. The problem with the structure of this technology is that when the object detection in the first stage detects an error in the foreground object, it cannot be rescued by the second s

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
  • Accessory identification method and device
  • Accessory identification method and device

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0016] The embodiments of the technical solution of the present invention will be described in detail below in conjunction with the accompanying drawings. The following embodiments are only used to illustrate the technical solutions of the present invention more clearly, and therefore are only examples, and cannot be used to limit the protection scope of the present invention.

[0017] It should be noted that, unless otherwise specified, the technical or scientific terms used in this application shall have the usual meaning understood by those skilled in the art to which the present invention belongs.

[0018] Such as figure 1 As shown, an accessory identification method provided by an embodiment of the present invention includes:

[0019] Step S1, input a test image.

[0020] Step S2, using the feature extraction layer of the pre-trained deep learning model to extract the features of the test image, and generate a feature map corresponding to the test image.

[0021] Step S3: Use the ...

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 belongs to the technical field of intelligent identification, and provides an accessory identification method and a device. The accessory identification method of the invention includes the following steps: inputting a test image; using a feature extraction layer of a pre-trained deep learning model to extract the features of the test image, and generating a feature map corresponding to the test image; using an accessory position generation layer of the deep learning model to traverse the feature map, generating all possible accessory positions and the score of each accessory position, comparing the score of each accessory position with a preset score threshold, and selecting an accessory position; and using an accessory classification layer of the deep learning model to get an accessory category corresponding to the selected accessory position. According to the accessory identification method of the invention, two stages of model training are combined into the training and optimizing of one model, and the test speed and accuracy of accessory identification are improved.

Description

Technical field [0001] The invention relates to the field of intelligent recognition technology, in particular to an accessory recognition method and device. Background technique [0002] The existing accessory detection technology is divided into two stages. The first stage: search for the target foreground object (such as clothes) in the image, and find and filter the foreground object by using search boxes of different sizes; the second stage: use a classifier (Model) Identify the filtered foreground object, and determine which target object the foreground object belongs to (for example, a shirt or pants, etc.). The core concept of this technology is to divide object detection and object recognition into two stages. The problem with the structure of this technology is that when the object detection in the first stage is detected incorrectly in the foreground, it cannot be rescued in the second stage. come back. In addition, the detection in the first stage and the identifica...

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): G06K9/62G06N3/08
CPCG06N3/08G06F18/2415G06F18/214
Inventor 杨闵淳张为义张惠棠陈智凯
Owner EMOTIBOT TECH LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap