eduzhai > Applied Sciences > Engineering >

CrossCount A Deep Learning System for Device-free Human Counting using WiFi

  • king
  • (0) Download
  • 20210505
  • Save

... pages left unread,continue reading

Document pages: 8 pages

Abstract: Counting humans is an essential part of many people-centric applications. Inthis paper, we propose CrossCount: an accurate deep-learning-based human countestimator that uses a single WiFi link to estimate the human count in an areaof interest. The main idea is to depend on the temporal link-blockage patternas a discriminant feature that is more robust to wireless channel noise thanthe signal strength, hence delivering a ubiquitous and accurate human countingsystem. As part of its design, CrossCount addresses a number of deep learningchallenges such as class imbalance and training data augmentation for enhancingthe model generalizability. Implementation and evaluation of CrossCount inmultiple testbeds show that it can achieve a human counting accuracy to withina maximum of 2 persons 100 of the time. This highlights the promise ofCrossCount as a ubiquitous crowd estimator with non-labour-intensive datacollection from off-the-shelf devices.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...