Clothes identification method and system and clothes care device

An identification method and identification system technology, applied in the field of clothing identification method, system and clothing care device, can solve problems such as inappropriate nursing procedures and limited user identification ability

Pending Publication Date: 2021-05-11
QINGDAO HAIER SMART TECH R & D CO LTD +2
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] Existing clothes care machines have a variety of care programs to choose from according to different materials or types of clothes. When using them, users need to identify the materials or types of clothes, and then choose the appropriate program to run. This method requires users to Have the ability to distinguish the material or type, but usually the user's ability to distinguish is limited, resulting in an inappropriate selection of care procedures

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
  • Clothes identification method and system and clothes care device
  • Clothes identification method and system and clothes care device
  • Clothes identification method and system and clothes care device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] In this embodiment, the created neural network model is used to train the neural network model by using clothing images marked with M clothing attribute labels, and the trained neural network model is used to identify multi-attribute clothing.

[0036] Specifically, this embodiment includes two parts: clothing image detection model training and clothing recognition model training.

[0037] Such as figure 2 As shown, the training of the clothing image detection model includes:

[0038] S21: clothing image preprocessing.

[0039] Including but not limited to standardizing input image size, converting image format, image cleaning, labeling, data enhancement, data augmentation, etc.; and randomly divided into training set and verification set.

[0040] S22: Set training parameters, learning rate, optimizer, and loss function.

[0041] S23: Obtain the predicted value through rfcn network calculation.

[0042] Use the training set to calculate the rfcn network to get the p...

Embodiment 2

[0065] This embodiment proposes a clothing care device based on the neural network model created in Embodiment 1. The clothing care device, such as Figure 5 As shown, it includes a clothing detection system 3 , a clothing recognition system 4 , a clothing image acquisition module 51 , a cloud 6 and a clothing care module 52 .

[0066] The clothing detection system 3 is used to detect the target area from the clothing image as the input of the clothing recognition system 4 based on the clothing image detection model trained in the first embodiment before the clothing recognition system 4 recognizes the clothing image.

[0067] Clothing recognition system 4 such as Figure 4 As shown, it includes a creation module 41 , a training module 42 and a recognition module 43 , which are set on the cloud 6 . The clothing recognition system 4 is implemented based on the clothing recognition method proposed by the present invention, wherein the creation module 41 is used to create a neur...

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 clothes identification method and system and a clothes nursing device, and the method comprises the steps: creating a neural network model, enabling identification tasks of a plurality of clothes attribute labels including clothes types, clothes materials and the like to share a convolution layer, and only connecting a plurality of parallel output full connection layers after a final global pooling layer, wherein output of each full-connection layer corresponds to one clothes attribute category. the invention has the advantages that after a plurality of tasks are integrated into a multi-task problem, the semantic extraction capability of the model for correlation of the plurality of attributes of the clothes can be improved, meanwhile, the data utilization efficiency is enhanced, and when two clothes images have the same attribute, output of the full connection layer of the corresponding attribute is similar, so that only one network is used, the system consumption and the operation time are saved, the features of multiple dimensions are separated, and the accuracy of multiple tasks is improved.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and in particular relates to a clothing recognition method, system and clothing care device. Background technique [0002] Existing clothes care machines have a variety of care programs to choose from according to different materials or types of clothes. When using them, users need to identify the materials or types of clothes, and then choose the appropriate program to run. This method requires users to It has the ability to distinguish the material or type, but usually the user's ability to distinguish is limited, resulting in an inappropriate selection of care procedures. [0003] In order to reduce the difficulty for users to identify and select operations, there is a method in the prior art to automatically identify the type or material of clothing based on machine learning methods, and then automatically match the care program according to the identified type or material of clothi...

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): G06K9/62G06N3/04
CPCG06N3/045G06F18/214
Inventor 姜大鹏苏明月
Owner QINGDAO HAIER SMART TECH R & D CO 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