Font Size: a A A

Research On Gesture Recognition Method Of Shopping Cart

Posted on:2020-03-15Degree:MasterType:Thesis
Country:ChinaCandidate:X HeFull Text:PDF
GTID:2428330575491191Subject:Instrument Science and Technology
Abstract/Summary:PDF Full Text Request
Gesture is a natural and intuitive interpersonal communication mode in life,and human-computer interaction technology continues to advance with the development of computer technology.With the rapid development of Internet of Things technology and the improvement of people's living standards,people have put forward more intelligent and convenient requirements for traditional shopping methods.Based on this background,it is especially important to design a convenient and intelligent shopping cart,which has great practical application value.This paper focuses on gesture recognition technology,supplemented by wireless transmission technology,and designs a smart shopping cart based on gesture recognition.This design proposes a gesture segmentation method based on skin color segmentation and contour segmentation.The Raspberry Pi is used to complete gesture information collection and gesture contour extraction.The Hu distance value is selected as the feature vector,and the feature vector correction model is used to modify the feature vector.The BP neural network model is trained and the gesture contour information is identified by the Alibaba Cloud server.Hardware part: The hardware circuit design is carried out for the single chip module,the drive module,the power module and the WIFI module.Data transmission part: Controls the transmission of the shopping cart movement instruction by means of the Raspberry Pi communicating with the ESP8266.The image information is transmitted by connecting the Raspberry Pi to the server.At the end of the paper,experiments on gesture segmentation and gesture recognition were carried out.The experimental results show that the gesture segmentation design can extract the gesture information from the complex background better and the degree of reduction is higher.The gesture recognition model has a good recognition effect on the gesture image.The averagerecognition accuracy of the four gestures reaches 96.87%,which has certain practical significance and application prospects.
Keywords/Search Tags:Gesture recognition, WIFI module, momentum, BP neural network, Alibaba Cloud
PDF Full Text Request
Related items