Effective gesture recognition method and device, control method and device and electronic device

A recognition method and gesture technology, applied in neural learning methods, character and pattern recognition, instruments, etc., can solve the problems of unsmooth system operation, slow algorithm operation, long response time, etc., to meet real-time detection requirements and ensure stability The effect of fast performance and computing resources

Active Publication Date: 2019-11-15
RECONOVA TECH CO LTD
View PDF6 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, the difficulty in applying deep neural networks to embedded devices is that the network is huge and complex, the computing power of embedded devices is insufficient, the algorithm runs slowly, the system does not run smoothly, and the response time is long, which brings inconvenience to users. good experience

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
  • Effective gesture recognition method and device, control method and device and electronic device
  • Effective gesture recognition method and device, control method and device and electronic device
  • Effective gesture recognition method and device, control method and device and electronic device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0067] mainly as figure 1 Shown, be the embodiment of gesture recognition method, it comprises the following steps:

[0068] 11. Obtain the image data of the current frame from the camera, and convert the image data into a three-channel RGB image format.

[0069] 12. To preprocess the collected images, firstly, normalize the images. Generally, the normalization process can be summarized as the following formula:

[0070]

[0071] where min is x i (i=1, 2...n) min, max is x i The maximum value of (i=1, 2...n).

[0072] Analyze the gesture detection and recognition results of the previous frame of images to determine whether a valid gesture is detected and perform corresponding processing. If no effective gesture is detected in the previous frame of image, the normalized image size is scaled to the input size of the first neural network model. If a valid gesture is detected and recognized in the previous frame image, map the position of the gesture in the previous frame ...

Embodiment 2

[0077] mainly as figure 2 Shown is an embodiment of the gesture control method, which includes the following steps:

[0078] 21. Perform statistics and analysis on the gesture recognition results of all detection frames within a fixed time interval before the current frame, and judge whether there are continuous and stable valid gestures within the fixed time interval.

[0079] Continuous and stable valid gestures are defined as: in the specified number of consecutive frames, the proportion of frames in which valid gestures are detected is greater than the specified threshold, the fluctuation range of the gesture area is small, and the gesture category does not change. The number of consecutive frames and the ratio threshold are specified by those skilled in the art according to the performance of the model and the actual situation of the product, and the fluctuation of the gesture area is measured by the relative position of the effective gesture area detected in two adjacen...

Embodiment 3

[0085] mainly as image 3 Shown, be the embodiment of the neural network model training procedure in effective gesture recognition method, it comprises:

[0086] 31. Acquire training images and gesture annotation information including required gestures. Among them, the training images are all images containing the gesture category to be recognized, and the situation of no gesture is no longer a separate category. The types of gestures to be recognized can be flexibly specified by those skilled in the art according to actual needs, and are not limited to a certain type or several types. The gesture annotation information includes two aspects: (1) frame information of all gestures to be recognized in the image, the frame information includes the center point x value of the gesture frame, the center point y value of the gesture frame, the width of the gesture frame, and the height of the gesture frame; 2) Encoding of all gesture categories to be recognized in the image. The ge...

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 provides an effective gesture recognition method and device, an effective gesture control method and device and an electronic device, and the method comprises the steps: S11, obtaining acurrent frame image collected by a camera; S12, performing gesture detection and recognition on the current frame image according to a preset recognition algorithm to obtain a possible region, a gesture category and confidence of a gesture in the current frame image; S13, sequentially carrying out gesture detection and recognition on all image frames of the video within a fixed time interval after the current frame to obtain a possible region, a gesture category and a confidence coefficient of a gesture in the image; and S14, judging whether the proportion of the image frames with the same gesture in the image frames within the fixed time interval is greater than a preset proportion threshold value or not, and if so, considering the gesture as a valid gesture. Gesture detection and recognition can be effectively and rapidly carried out on the embedded terminal, and convenient and rapid man-machine interaction is carried out.

Description

technical field [0001] The invention relates to a real-time gesture detection and judgment method, device and electronic equipment of computer vision based on artificial intelligence deep learning technology. Background technique [0002] With the rapid development of computer technology, there are more and more applications of deep learning in the field of computer vision. Among them, using gestures for human-computer interaction is a very convenient method and has great application value. Gesture recognition and control technology can provide a remote non-contact human-computer interaction method, so fast and accurate gesture recognition algorithms can bring users a convenient and friendly experience. At present, the difficulty in applying deep neural networks to embedded devices is that the network is huge and complex, the computing power of embedded devices is insufficient, the algorithm runs slowly, the system does not run smoothly, and the response time is long, which...

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
CPCG06N3/084G06V40/107G06N3/045
Inventor 徐绍凯贾宝芝
Owner RECONOVA TECH 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