1 of 1

Smart Pedestrian Counter : low-cost foot traffic counter

Sonoma State University Science Symposium 2020

Sarah Chesbrough and Logan Lawrence

Faculty Advisor: Dr. Farid Farahmand | Industry Advisor: Neil Hancock

lawrencl@sonoma.edu chesbrou@sonoma.edu

Website: https://www.smartpedestriancounter.com/

System Diagrams

The city of Rohnert Park wants a cost-effective, modular, non-invasive, and easily removable way to monitor traffic on the city trails

​

Major Requirements

  • Be low-cost
  • Ensure privacy
  • Have no permanent installations
  • No monthly fees for data transmission
  • Be Expandable
  • Send data to private database

Technologies Used

LoRaWAN : A low power, low packet, long range, messaging protocol that uses 915 MHz.

​

OpenCV: Open computer vision, used for image recognition for counting number of people and bikes in a given image��HB100- Doppler effect sensor, used to determine when there is motion, so we know when to take pictures.

Conclusion

Problem Statement

System overview: node

System overview

Results

Acknowledgments

We would like to thank our advisors for this project, Dr. Farid Farahmand and Neil Hancock. We also would like to thank Darren Jenkins, the city manager for Rohnert Park, for allowing us to interview him for the project. We extend our thanks to OSRP for providing the funding for this project. Lastly we would like to thank the SSU Engineering faculty, staff, and students who have helped us get this far.

  • Shows the data visualization of our sent data, which can be seen on our website�
  • The data that is processed by OpenCV gets sent from node via Lora������
  • Shows the number of pedestrians and bikes detected versus number present�
  • Only starts to miss some bikes once there are 3 people and 3 bikes present at once

Future Works

  • Created a system that takes pictures when motion is detected

​

  • Has the ability to distinguish between bicyclists and pedestrians

​

  • Can send data using LoRa

​

  • Visualizes the data on a website for the user

​

  • Transition to a lower-power embedded system�
  • Expanding to even more nodes�
  • Combining all parts into one setup�
  • Testing complete setup outdoors