Swimming Tracker - Motion Recognition
Download as PPTX, PDF1 like1,256 views
This document summarizes a swimming tracker that uses MEMS sensors and machine learning algorithms to detect and analyze swimming motions. It discusses (1) introducing objectives like detecting swimming and resting, counting laps, and recognizing styles while minimizing RAM usage; (2) processing raw sensor data like filtering gravity force from accelerometer data and integrating gyroscope data; and (3) analyzing swimming data by classifying orientation, chains of motion, and inputting data into probabilistic classifiers like expectation-maximization to determine probability distributions of activities.
![Raw data processing
Chains of motion subactions
Dictionary of motion subactions: [A, B, C, D,...]
● Typical for swimming: A, B,...
● Typical for resting: C, D,...
Temporal chains of moves:
● Swimming: AABAAAABABAAABABBAA
● Swimming: ACACCACAACCACACCACC
● Resting: ADADDADAADDADADDADD
● Resting: ACACCCCACCCACCCCACC](https://image.slidesharecdn.com/mems-embedded-techtalk-2015-150619121056-lva1-app6892/85/Swimming-Tracker-Motion-Recognition-21-320.jpg){kind=tracking}
Swimming Tracker - Motion Recognition
- 1. ©2015 GlobalLogic Inc. Swimming Tracker: Motion Recognition Orest Hera
- 2. ©2015 GlobalLogic Inc. Introduction Raw data processing Swimming data analysis 01 02 03
- 3. Introduction: ● Objectives ● MEMS sensors ● Classification problems ● Sensor-based activity recognition 01
- 4. Introduction Objectives ● Swimming and resting detection ● Lap counting ● Swim style recognition ● Minimal usage of RAM of microcontroller
- 5. Introduction MEMS (MicroElectroMechanical System) sensors ● 3D Accelerometer ● 3D Gyroscope ● 3D Magnetometer
- 6. Introduction Applications of MEMS sensors ● Mobile devices (Android, iOS) ○ Display/map orientation ○ Step counter, Compass applications ○ Augmented reality ● Small custom devices ○ Small vehicle navigation and stabilization (quadcopter) ○ Industrial automation ○ Innovative smart systems
- 7. Introduction Pattern recognition tools ● Image recognition open source software: ○ OpenCV library ○ Many specialized tools (face, poses, hands tracking) ● Speech recognition open source software: ○ CMU Sphinx (HMM) ○ Julius (HMM 3-gram) ○ Kaldi (Deep neural network) ● Sensor-based activity recognition: ○ Custom classifiers
- 8. Introduction Sensor-based activity recognition ● Accelerometer “Activity recognition from accelerometer data” / N.Ravi, N.Dandekar, P.Mysore, M.L.Littman, IAAI’05 Proceedings, Vol. 3, 1541-1546, 2005 ● Gyroscope and Accelerometer A Public Domain Dataset for Human Activity Recognition Using Smartphones / D.Anguita, A.Ghio, L.Oneto, X.Parra and J.L.Reyes-Ortiz, ESANN 2013. Bruges, Belgium 24-26 April 2013 (UCI Machine Learning Repository) ○ 30 subjects performing activities of daily living ○ 561-feature vector with time and frequency domain variables for fixed-width window ○ 10299 labeled instances ● Techniques of Natural Language Processing
- 9. Introduction Activity recognition from accelerometer data Classifiers: ● Decision Trees ● K-nearest neighbors ● SVM ● Naive Bayes Accuracy: ● Multiple subjects cross-validated: 92 - 99% ● One subject training, same subject another data for testing: 70 - 90% ● One subject training, another subject for testing: 46 - 73%
- 10. Raw data processing: ● Gravity force detection by accelerometer ● Rotation speed by gyroscope ● Complementary filter ● Compass 02
- 11. Raw data processing Raw sensor data processing
- 12. Angle by vector of gravity force Raw data processing Advantages: ● Direct measurement without error accumulation Disadvantages: ● System own acceleration should be filtered ● Relaxation time due to Low-pass filter
- 13. Raw data processing Angle by rotation speed integration Advantages: ● Can be used during accelerated motion Disadvantages: ● Integration error accumulation
- 14. Raw data processing Complementary filter Advantages: ● High frequency by gyroscope ● Low frequency by gravity force vector Disadvantages: ● Cannot compensate error accumulation drift of rotations around vector of gravity force
- 15. Expected Magnetometer raw data Raw data processing Reality
- 16. Raw data processing Magnetic distortions Distortion types: ● Hard iron (permanent magnet) ● Soft iron (easily magnetized and demagnetized) Types of compensating methods: ● Offline (least squares methods) ● Real-time adaptive (Kalman filter, neural networks,...)
- 17. Raw data processing Magnetometer calibration Ellipsoid equation: The least-squares problem Pseudo-inverse matrix
- 18. Swimming data analysis: ● Device orientation ● Chains of motion subactions ● Probabilistic classification 04
- 19. Swimming data analysis X-,Z-axes horizontal direction (yaw)
- 20. Swimming data analysis X-,Y-,Z-axes vertical direction (pitch)
- 21. Raw data processing Chains of motion subactions Dictionary of motion subactions: [A, B, C, D,...] ● Typical for swimming: A, B,... ● Typical for resting: C, D,... Temporal chains of moves: ● Swimming: AABAAAABABAAABABBAA ● Swimming: ACACCACAACCACACCACC ● Resting: ADADDADAADDADADDADD ● Resting: ACACCCCACCCACCCCACC
- 22. Probabilistic classification Swimming data analysis ● Input data points
- 23. Probabilistic classification Swimming data analysis ● Input data points ● Expectation-maximization
- 24. Probabilistic classification Swimming data analysis ● Input data points ● Expectation-maximization
- 25. Probabilistic classification Swimming data analysis ● Input data points ● Expectation-maximization ● Probability distribution
- 27. ©2015 GlobalLogic Inc. Thank You