Research and implementation of dash cam installation and image quality automatic detection based on big data

Geotab‘s dash cams deliver a clear and complete picture of harsh driving events and provide crucial video evidence in the case of collisions and insurance disputes. It is essential to ensure the dashcams work well and capture high quality footage.
The object of this project is to automatically detect dash cam installation faulty and monitor the image quality. We will implement a streaming or batch-basis algorithm to send alerts to fleet managers that a camera needs to be fixed, reinstalled, etc. In addition, we will also quarantine the data coming from the dash cam so that that video data will not be used in subsequent training.
In addition, this project also involves building an ML model deployment and production pipeline steps and associated data pipeline steps to add additional valuable data to the products we provided.

Faculty Supervisor:

Marsha Chechik

Student:

Partner:

Geotab Inc

Discipline:

Computer science

Sector:

Information and cultural industries; Professional, scientific and technical services; Transportation and warehousing

University:

University of Toronto

Program: