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

Monocular image-based glasses detection method and device, storage medium and equipment

A glasses detection, single-purpose technology, applied in the field of pattern recognition, can solve the problems of image gradient mutation, unfavorable glasses detection, judgment instability, etc., to achieve the effect of accelerating calculation, ensuring accuracy, and judging stability

Pending Publication Date: 2021-05-21
BEIJING TECHSHINO TECH +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] First of all, the method based on image processing is easily disturbed by the external environment such as illumination. When the local illumination is too strong, the gradient of the image will change suddenly, which is not conducive to the detection of glasses based on the edge of the image.
In addition, the method based on glasses edge detection only uses simple information such as edges and corners of the image, and does not use the texture features of the image, so the judgment is unstable and the effect is poor

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
  • Monocular image-based glasses detection method and device, storage medium and equipment
  • Monocular image-based glasses detection method and device, storage medium and equipment
  • Monocular image-based glasses detection method and device, storage medium and equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0040] The implementation of the present invention provides a method for detecting glasses based on monocular images. It should be noted that the detection of glasses in the present invention refers to the detection of transparent glasses, including nearsighted glasses, farsighted glasses, transparent plain glasses, etc. Sunglasses do not belong to The detection scope of the present invention.

[0041] Such as figure 1 As shown, the method includes:

[0042] Step S100: Obtain the left eye area or the right eye area from the face image to obtain a monocular image.

[0043] In this step, various face detection and positioning methods can be used to obtain the left eye area or the right eye area from the face image.

[0044] Step S200: Input the monocular image into the trained convolutional neural network to obtain the probability of wearing glasses and not wearing glasses, where:

[0045] In order to reduce the time complexity of the algorithm, the lightweight convolutional ...

Embodiment 2

[0066] An embodiment of the present invention provides a device for detecting glasses based on a monocular image, such as Figure 4 As shown, the device includes:

[0067] The preprocessing module 10 is used to obtain the left eye area or the right eye area from the face image to obtain a monocular image.

[0068] The detection module 20 is used to input the monocular image into the trained convolutional neural network to obtain the probability of wearing glasses and not wearing glasses, wherein:

[0069] The convolutional neural network includes the first convolutional layer, the first pooling layer, the second convolutional layer, the second pooling layer, the third convolutional layer, the third pooling layer, the first fully connected layer, The second fully connected layer, the third fully connected layer and the Softmax layer.

[0070] The convolutional neural network is trained through a training set, and the samples in the training set include left-eye image samples ...

Embodiment 3

[0087] The methods described in the above-mentioned embodiments provided in this specification can implement business logic through computer programs and record them on a storage medium, and the storage medium can be read and executed by a computer to achieve the effect of the solution described in Embodiment 1 of this specification. Therefore, the present invention also provides a computer-readable storage medium for glasses detection, including a memory for storing processor-executable instructions. When the instructions are executed by the processor, the method for glasses detection based on a monocular image in Embodiment 1 is implemented. step.

[0088]The present invention combines the monocular image with a deep convolutional neural network for glasses detection, and uses a lightweight convolutional neural network to perform binary classification on the acquired area of ​​one eye to determine whether a portrait wears glasses, which can be effective and fast It solves th...

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 monocular image-based glasses detection method and device, a storage medium and equipment, and belongs to the field of mode recognition. The method comprises the following steps: training a CNN through a training set comprising a left eye image sample and a right eye image sample, wherein the CNN comprises a plurality of convolutional layers, a pooling layer, a full connection layer and a Softmax layer; obtaining a left eye area or a right eye area from the face image to obtain a monocular image; inputting a monocular image into the trained CNN, obtaining eye features in the last full connection layer, enabling the Softmax layer to classify the eye features, and obtaining the probability of wearing glasses and the probability of not wearing glasses. According to the invention, the high symmetry of the face image is utilized, glasses detection is completed by only one eye image, and the judgment efficiency of an algorithm is improved while the accuracy of glasses detection is ensured.

Description

technical field [0001] The invention relates to the field of pattern recognition, in particular to a monocular image-based glasses detection method, device, computer-readable storage medium and equipment. Background technique [0002] The rise of deep learning has brought great development to biometric recognition, especially in face recognition, image understanding and other research directions. With the continuous improvement of the accuracy of face comparison algorithms, more and more researchers have begun to pay attention to the research on human behavior attributes. In the field of face recognition, research on human behavior includes research on human attributes, such as expressions, postures, etc., as well as external attributes, such as whether to wear glasses or sunglasses. [0003] At present, the research on whether to wear glasses mainly uses the image processing method, which is based on the information projection of the edge of the eye image to locate the bea...

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): G06K9/00G06N3/04G06N3/08
CPCG06N3/08G06V40/165G06V40/172G06N3/045
Inventor 周军王洋
Owner BEIJING TECHSHINO TECH
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