Font Size: a A A

Bone Conduction Sound Recognition Based Oral Motor Condition Monitoring And Automatic Analysis Of Dietary Laws

Posted on:2015-10-28Degree:MasterType:Thesis
Country:ChinaCandidate:S ChenFull Text:PDF
GTID:2298330452959559Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
In a medical resources limited social environment, mobile health monitoring,acting as a new thing, provides users with timely, effective and affordable health care.Human physiological information processing has become a new cutting-edge research.Technologies for mobile health applications and services will be the strategic directionof communications operators based on the development of The Internet of Things.This work achieves a basic mobile monitoring system for oral motor conditionmonitoring and automatic analysis of dietary laws, a kind of health service withphysiological information acquisition, processing, analysis, and results feedback.Through the data communication and co-processing of the three distributed parts,sensor node, mobile terminal and the server, the whole system is established.The monitoring and analysis for dietary laws is achieved by detecting the oralmotor conditions and the states of human movement. Oral motor conditions aredefined as: Chewing, Talking, Coughing, Silencing. Human movement states aredivided into: Immobile/Micro-Moving, Walking, Running. Due to the concealment ofthe oral motor conditions, the bone conduction microphone is used to capture the boneconduction sound. Human movement state is obtained by an accelerometer.The design of ARM embedded hardware considers the computing requirements,data storage space, size, CMU performance, power, chip compatibility. By a globaltradeoff on user information security, node performance, data scale and networktransmission, bone conduction sound signal processing, a method of two proceduresfor calculating MFCC parameters is proposed. Construct bone conduction sounddatabase and recognition system on server with HTK, the recognition rate is up tomore than90%.Android software is developed with Bluetooth, WIFI, data processing and storage,data statistical analysis and results visualization function. Statistical methods are uaedto analyze the oral motor conditions and the body motion, mine user’s eating laws.According to medical research on healthy diet, a diet scoring system is designed toshow the dietary laws in a graph form, provide reasonable guide to user.
Keywords/Search Tags:Physiological Sensor, Bone Conduction Sound Recognition, AndroidApplication, Statistical Analysis, Dietary Laws
PDF Full Text Request
Related items