Font Size: a A A

Design Improvement And Implementation Of Gesture Recognition Based On OpenCV

Posted on:2019-09-21Degree:MasterType:Thesis
Country:ChinaCandidate:H HeFull Text:PDF
GTID:2428330566987222Subject:Computer technology
Abstract/Summary:PDF Full Text Request
Nowadays,the concept of Intelligent Electronics has gradually deepened the production and life of people,and the demand for the convenience of electronic operation is increasing.Using gesture to do command operation of electronic products has become a key technology of convenience,such as gesture manipulation,UAV,video game or other smart home with camera.At the same time,gesture recognition can also combine the new technology such as AR to realize many functions such as virtual operation and so on,and greatly improve the sense of human-computer interaction.Therefore,it is of great significance to study the gesture recognition.This technology involves a variety of aspects,such as image pattern recognition and analysis,video image processing and computer vision.This paper introduces the development background and significance of gesture recognition project,and analyzes the steps and algorithms involved in the whole process.The gesture recognition system is designed to facilitate human-machine interaction.The template is ten digital gesture images from 0 to 10.This system is based on OpenCV function library to study gesture recognition in C++ development environment.The algorithm is based on HU invariant moments as image features.In this paper,the system can dynamically identify the numbers expressed by the gestures and display the results by recording the operator gestures through the camera,and can also identify the moving direction of the gestures,so as to choose the function of the machine end option box or the digital option key by using gestures.The number of frames per second is 27.On the whole,it can be divided into four modules: image preprocessing,feature extraction,recognition and matching and target tracking,including sharpening processing,Gauss denoising,HSV color extraction,expansion corrosion,edge detection,contour extraction,template matching and other steps.At present,the system has been developed,and the experimental results can basically achieve the tracking of gestures and the recognition of the numbers,and the results are displayed.
Keywords/Search Tags:Image processing, pattern recognition, gesture recognition, opencv
PDF Full Text Request
Related items