Chapter 9. Activity Recognition with Mobile Phone Sensors
While the previous chapter focused on pattern recognition in images, this chapter is all about recognizing patterns in sensor data, which, in contrast to images, has temporal dependencies. We will discuss how to recognize granular daily activities such as walking, sitting, and running using mobile phone inertial sensors. The chapter also provides references to related research and emphasizes best practices in the activity recognition community.
The topics covered in this chapter will include the following:
- Introducing activity recognition, covering mobile phone sensors and activity recognition pipeline
- Collecting sensor data from mobile devices
- Discussing activity classification and model evaluation
- Deploying activity recognition model
Introducing activity recognition
Activity recognition is an underpinning step in behavior analysis, addressing healthy lifestyle, fitness tracking, remote assistance, security applications, elderly...