Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
MASK RECOGNITION MODEL TRAINING METHOD AND APPARATUS, DEVICE, AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2023/221608
Kind Code:
A1
Abstract:
The present application provides a mask recognition model training method and apparatus, a device, and a storage medium. The method comprises: respectively cutting each picture in a first training set into a plurality of image blocks having the same size, and performing tag labeling on the image blocks to obtain a second training set; respectively cutting each picture in a third training set into a plurality of image blocks having the same size, and performing tag labeling on the image blocks to obtain a fourth training set; pre-training a first preset model by using the second training set to obtain a pre-trained model; and performing formal training by using the fourth training set and a second preset model to obtain a mask recognition model, wherein the first preset model and the second preset model can use a YOLOV4-tiny network model, the YOLOV4-tiny network model comprises a backbone network and a remaining part of networks, and parameters of the backbone network of the models in the pre-training and formal training are frozen. According to the method, an operation speed of the model and the accuracy of recognition of a small target at a relatively long distance are improved.

Inventors:
MENG HAIXIU (CN)
WAN YECONG (CN)
CHEN LUCHENG (CN)
SHI SENMIN (CN)
ZHENG XUDONG (CN)
Application Number:
PCT/CN2023/080248
Publication Date:
November 23, 2023
Filing Date:
March 08, 2023
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
COSMO INST OF INDUSTRIAL INTELLIGENCE QINGDAO CO LTD (CN)
COSMOPLAT IOT TECH CO LTD (CN)
HAIER DIGITAL TECH QINGDAO CO LTD (CN)
International Classes:
G06V40/16
Foreign References:
CN114898434A2022-08-12
CN114283469A2022-04-05
CN113989179A2022-01-28
CN114283462A2022-04-05
Attorney, Agent or Firm:
LEADER PATENT & TRADEMARK FIRM (CN)
Download PDF: