We are very pleased to announce that Michael Bewley from nearmap will give a talk about geospatial analysis of tree canopy data at Sydney University this month.
Date: Thursday, 19th May 2022
Time: 6:30pm - 7:30pm (AEST)
Location: F07.01.173.Carslaw Building.Carslaw Lecture Theatre 173 (or register for Zoom link)
Michael Bewley - nearmap
Title: From AI to Action: Applying Geospatial AI to create Australia’s largest high resolution tree canopy study
Abstract:
Nearmap AI is one of the few very large scale geospatial machine learning based products in operation globally. Having been run at high resolution (7cm/pixel) on over 8 million square kilometres of historical imagery, we turn tens of petabytes of imagery into petabytes of AI data sets. Carefully constructed definitions of dozens of object types allow us to create a frequently updated vector map of urban areas in four countries. Instead of focussing on how to do Geospatial AI, we will cover a compelling example of what becomes possible with such a data set - national, high resolution vegetation management. Nearmap performed a study covering 5,000 suburbs worth of high resolution tree canopy data from 2021. In addition, a seasonally matched annual study over a decade in Adelaide showcases what is possible with annual monitoring of a city. The talk focusses on the methodology, findings, and implications of massive scale Geospatial AI. Quantifying tree cover is an important problem from a number of perspectives - urban heat islands, bushfire risk, liveability and other aspects of climate and the environment. It is important to be able to compare results in multiple geographic areas, and over time, in order to monitor change, and compare the impact of policy, climate and other effects between cities.
The analysis combines 7cm/pixel data on tree and building coverage from the Nearmap AI product suite, with spatial and population data from the Australian Bureau of Statistics on all Australian capital cities. It is the first of its kind to leverage high resolution tree data from a consistent methodology at a national scale, which naturally leads to new findings that were not previously quantifiable.
Biography
With degrees in electrical engineering and physics — and a passion for machine learning — Dr. Michael Bewley joined Nearmap in 2017 as our first data scientist. Now the Senior Director of AI Systems, he offers extensive machine learning experience from Cochlear, CommBank and the Australian Centre for Field Robotics.
Dr. Bewley leads the development of the technology behind Nearmap AI, quantifying the evolution of cities with the rich, accurate AI data sets.
Virtual attendance
This is an in person event event, and we hope you can join us at Sydney Uni. If you would like to attend virtually, the Zoom registration link is https://uni-sydney.zoom.us/j/89975275170?pwd=T3RtV1BxNnVSTkoyN0tUTE9LSXVrZz09. Please note for security reasons, you will need to register in advance for this meeting. After registering, you will receive a confirmation email containing information about joining the meeting.
Any questions, please feel free to contact: secretary.nswbranch@statsoc.org.au
@StatSocAus
