Smartphone based human activity recognition irrespective of usage behavior using deep learning technique

HIGHLIGHTS

• What: The authors focus on the problem of designing CNN based HAR models for accelerometer signals. This paper is arranged as follows: Sect 2 describes the related work; in Sect 3, proposed work is explained; Int. j. inf. tecnol. Recognizing transition and sequence of activities were the main objective, using a smartphone-based MultiInstance Multi-Label (MIML) ensemble model considering kNN distance metrics. This work aims to identify four human activities (sitting, standing, walking, jogging) based on inertial sensor reading of smartphones carried by users at Shirt Pocket (SPT), the right front pocket of the pant(RPT . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or Email

  Password

  Remember me

  Lost your password?

  If you don't have an account, you can create one here.