Object Detection Using OpenCV in Low-Visibility Environments
Loading...
Authors
Odegard, Cole
Date of Issue
2025
Type
Thesis
Language
en_US
Subject Keywords
Other Titles
Abstract
Object detection is crucial to autonomous vehicles’ ability to navigate environments and avoid collisions. In developing the “Snowbot” autonomous snow plow robot, we will employ the tools from Python’s OpenCV package to attempt to clean images from the live camera feed so that OpenCV models can correctly identify objects to avoid (OpenCV Documentation, n.d.). Snowbot is intended to operate in a low-visibility environment where snow, clouds, and other vision impairments will obscure objects, people, and animals. A successful Snowbot should be able to use OpenCV object detection models to detect common obstacles and interface with other programs to determine a path that avoids collisions with obstacles. In this thesis, we will lay out a workflow for image cleaning, object detection, and object tracking for Snowbot’s navigation system. Reliably identifying and tracking objects is critical to Snowbot's safety and functionality.