Locating anomalies in aerial multi-spectral imagery

AERIUM employs aerial drones to collect data and perform geospatial data analytics in real-time applicable for a variety of fields which include the oil sector, airport transportation, forestry, and smart wildlife management. This proposal focuses on leveraging the data characteristics available in multispectral imagery and that may be missing in standard RGB data to detect objects in vegetation or other target areas. The research will involve applying machine learning algorithms and computer vision techniques to multispectral aerial imagery collected using a drone in real-time. Object detection would benefit search and rescue operations as well as the agriculture industry among others. Our objective is to create a robust algorithm which can detect objects it has never seen before. Our proposed methodology is to leverage the information in multispectral data to provide a more concrete definition of what is known and unknown when locating anomalies.

Faculty Supervisor:

Irene Cheng

Student:

Partner:

Aerium Analytics

Discipline:

Computer science

Sector:

Agriculture; Professional, scientific and technical services

University:

University of Alberta

Program: