Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Eye height measurement method based on deep learning

A technology of height measurement and deep learning, applied in the field of artificial intelligence, can solve problems such as harsh measurement conditions and affecting the accuracy of measurement results, and achieve the effect of long time-consuming improvement, simple process and accurate results

Pending Publication Date: 2021-02-05
HANGZHOU DIANZI UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, during the measurement process, the measured person is required to sit still for a long time, and the slight movement of the measured person will seriously affect the accuracy of the measurement results. For younger measured persons, the measurement conditions are relatively harsh

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
  • Eye height measurement method based on deep learning
  • Eye height measurement method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] Below in conjunction with accompanying drawing, the present invention will be further explained;

[0021] The hardware environment in this embodiment is: 8vCPU / 64G memory, the GPU is V100, and the software operating environment is: CUDA:9.2.148, python3.7, pytorch 1.0.1.post2.

[0022] Such as figure 1 As shown, it specifically includes the following steps:

[0023] Step 1. Dataset division

[0024] Divide the captured user original pictures into training set, verification set and test set;

[0025] Step 2. Data preprocessing

[0026] Reset the picture pixels in the original data set to two pixel sizes of 320*480 and 224*224 to obtain training set 1, verification set 1, test set 1 and training set 2, verification set 2, and test set 2 respectively.

[0027] Step 3. Neural Network Training

[0028] Input the training set 1 into the pre-trained FPN neural network, and input the training set 2 into the pre-trained UNET neural network. The pre-training weights of the t...

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 an eye height measurement method based on deep learning, and the method specifically comprises the steps: carrying out the data preprocessing, and resetting the pixels of a picture in a data set according to different neural networks; carrying out image segmentation: inputting the picture into a neural network, and segmenting the picture to obtain an image only comprising an eye part; processing the eye image, detecting and extracting an eye contour line in the eye image, and obtaining the eye height through a method of making a minimum enclosing rectangle and a midperpendicular intersection distance; and finally, outputting a result, and solving the average value of the eye heights obtained by the two networks as a final result. Compared with the prior art, the method has the advantages of being convenient to operate, simple in process and accurate in result, and can be widely applied to preoperative data collection needing eye appearance plastic surgery, imagedata obtained after neural network segmentation only contain images of the eyes of a user, other facial features are removed, certain safety is achieved, and the user privacy is secured.

Description

technical field [0001] The invention belongs to the field of artificial intelligence, and in particular relates to a method for measuring eye height based on deep learning. Background technique [0002] Eye height refers to the distance between the upper and lower eyelids when the body is upright and the eyes are looking straight ahead. Both congenital drooping eyelids and drooping eyelids caused by weak eye muscles can reduce the height of a person's eyes. Eye height not only affects a person's spirit and appearance, but also affects sight. According to statistics, the number of people undergoing eye appearance correction surgery every year in the country reaches tens of thousands. In the process of data collection, users are required to provide data such as eye height and width. [0003] In the prior art, after the user arrives at a designated place, professionals usually use measuring tools such as a ruler to measure the eye data of the user. However, during the measu...

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): G06K9/00G06K9/46G06N3/04G06N3/08
CPCG06N3/08G06V40/193G06V40/19G06V10/44G06N3/045
Inventor 岳雪颖田泽坤孙玲玲
Owner HANGZHOU DIANZI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com