Creating a scalable model for optimizing terrestrial & mangrove tree counts and survival estimates using computer vision

In efforts to offset carbon emissions, many companies have recently invested in tree planting efforts; planted
trees absorb carbon dioxide from the atmosphere and store that carbon as they grow. The amount of carbon a
tree can store depends on a variety of factors, including tree size, species, and whether the tree is alive or dead.
In large scale tree-planting efforts, like those performed by Veritree, estimating carbon offset accurately requires
these factors to be recorded for a vast number of trees. This time-intensive process can be shortened by
developing “computer vision” algorithms which can automatically detect trees and extract their features from
smartphone images. Using computer vision, this project aims to increase the rate at which Veritree can survey
their tree populations, ultimately providing a more comprehensive assessment of their planting efforts.

Faculty Supervisor:

Jess McIver

Student:

Partner:

Veritree

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

The University of British Columbia

Program: