Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Gesture data set acquisition method for YOLO network, and gesture recognition method and device

A technology of gesture recognition and acquisition method, which is applied in the direction of character and pattern recognition, instruments, computer parts, etc. It can solve the problems of poor ability to suppress complex background noise, and can not solve the problem of noise suppression well, and achieve the effect of strong suppression ability

Inactive Publication Date: 2018-10-02
深圳市智能机器人研究院
View PDF7 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The existing technology for photographing and recognizing user gestures is mainly a gesture segmentation method based on skin color segmentation, which separates the hand part from other parts in the picture by analyzing the difference between the skin color of the hand in the picture and the background color, and realizes Gesture recognition, however, this method has a poor ability to suppress complex background noise such as faces and skin-colored walls in the picture, especially for pictures centered on real scenes, traditional methods such as connected domain denoising can not Solve the problem of noise suppression very well, the above-mentioned gesture segmentation method based on skin color segmentation has great disadvantages

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
  • Gesture data set acquisition method for YOLO network, and gesture recognition method and device
  • Gesture data set acquisition method for YOLO network, and gesture recognition method and device
  • Gesture data set acquisition method for YOLO network, and gesture recognition method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] In this embodiment, the gesture data collection method used for YOLO network, such as figure 1 shown, including the following steps:

[0047] S1a. Obtain a picture containing a training gesture, so that the training gesture is included in the sliding window; the sliding window is a window that can move within the picture range, and the position and size of the sliding window can be adjusted;

[0048] S2a. Save the image acquisition window picture as a training gesture picture, and record the corresponding calibration frame position and size and gesture category label; the calibration frame position and size are the corresponding sliding window position and size when saving the training gesture picture, and the Gesture category tags include gesture type information for training gestures;

[0049] S3a. Use a simple threshold skin color segmentation algorithm based on the YCbCr color space to filter the background of the training gesture picture to form a gesture training...

Embodiment 2

[0072] The gesture recognition method of this embodiment can use the gesture data set collected in Embodiment 1, and use the YOLO network to train the gesture recognition model.

[0073] A gesture recognition method based on YOLO, comprising the following steps:

[0074] S1b. Utilize the steps S1a-S4a of the image collection method described in embodiment 1 to autonomously collect the gesture data set to be trained;

[0075] S2b. Using the data set to train the gesture recognition model of the YOLO network to obtain a gesture recognition model;

[0076] S3b. Capture the picture to be recognized by the camera device, and use a simple threshold skin color segmentation algorithm based on the YCbCr color space to filter the background of the picture to be recognized;

[0077] S4b. Use the trained gesture recognition model to perform real-time gesture category prediction and gesture positioning on the image to be recognized;

[0078] S5b. Smoothing and correcting the gesture reco...

Embodiment 3

[0096]A gesture recognition device based on YOLO, comprising:

[0097] memory for storing at least one program;

[0098] The processor is configured to load the at least one program to execute the YOLO-based gesture recognition method described in Embodiment 2.

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 gesture data set acquisition method for a YOLO network, and a gesture recognition method and device. The acquisition method comprises the steps that the frame including the training gesture is acquired, the image acquisition window frame is saved as the training gesture image, the training gesture image background is filtered by using a simple threshold skin color segmentation algorithm based on the YCbCr color space so as to obtain the training gesture image of diverse backgrounds, and the training gesture image of diverse backgrounds is applied to form the gesture data set. The gesture recognition method comprises the steps that YOLO network training and gesture recognition are performed by using the obtained gesture data set. The device comprises a memory and aprocessor. The YOLO network is enabled to have the recognition model for the gesture in the image or the frame, and the recognition process has high noise suppression capacity for the face, the skinlike wall and other complex backgrounds existing in the image. The gesture data set acquisition method for the YOLO network, and the gesture recognition method and device are applied to the technicalfield of image recognition processing.

Description

technical field [0001] The invention relates to the technical field of image recognition processing, in particular to a gesture data collection method, gesture recognition method and device for YOLO network. Background technique [0002] Terminology Explanation [0003] YOLO (YouOnlyLookOnce), is a target detection method based on deep learning, which can be used to train neural networks. The neural network trained by YOLO can be used to solve the regression problem of target area prediction and category prediction. Its advantage is that it can guarantee High detection speed and accuracy. [0004] Gesture commands are an efficient way of human-computer interaction. Computers, mobile phones or other smart terminals obtain user gestures, analyze and obtain corresponding gesture commands, and perform corresponding operations or feedback based on gesture commands. The smart terminal can obtain the user's gestures through buttons, touch screen, or camera shooting. Among them, t...

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): G06K9/00
CPCG06V40/28G06V40/107
Inventor 陈虎谷也盛卫华
Owner 深圳市智能机器人研究院
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
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