Industry Insights

Five Key developments in GeoAI as per the industry experts

This blog post on expert forecast on AI/ML is part of the GeoAI Series, inspired by the WGIC GeoAI Report.

Harsha Vardhan Madiraju July 15, 2021
GeoAI Key Trends and geospatial industry expert forecast

For developing the report “Geospatial AI/ML Applications and Policies: A Global Perspective,” WGIC conducted one-to-one interviews with more than thirty AI/ML experts in the geospatial industry, including WGIC Members. Here is the list of five key developments in GeoAI that we will witness as per the experts.

1. Increased automation

Many of the current geospatial AI/ML tasks require some aspects of human intervention to get adequate results. Advances in AI/ML techniques will allow for full automation in areas like mapping, object identification, feature/attributes extraction in the objects identified e.g. number of lanes in a road, or condition/ damages in the road or building.

2. Better natural resource management

With advances in deep learning techniques and easy access to satellite imagery and remote sensing data, geospatial AI/ML will find greater adoption in industries like agriculture, forestry, climate change, etc., e.g., those that involve tracking and managing natural resources.

3. Real-time applications

Several of the current geospatial AI/ML techniques require days, weeks or even months to turn geospatial data into actionable results. Increasing computing power, edge computing, better algorithms and ML support in field equipment will allow for the development of real-time and near real-time geospatial AI/ML applications.

4. More data for AI/ML

With the advancement of highly-capable miniaturized remote sensing, imagery, and LIDAR systems and satellites, businesses will have access to a greater collection of real-time geospatial and remote sensing data with increasingly higher resolution/quality that will allow for new geospatial AI/ML use-cases currently not possible.

Several of the current geospatial AI/ML techniques require days, weeks, or even months to turn geospatial data into actionable results. Increasing computing power, edge computing, better algorithms, and ML support in-field equipment will allow for the development of real-time and near real-time geospatial AI/ML applications.

5. Accessibility of AI/ML applications

The application of geospatial AI/ML techniques to business problems currently requires trained data scientists and machine learning engineers. With pre-trained ML models integrated with GIS software, geospatial AI/ML will become accessible for all, from a small farmer to an executive at a large enterprise.


Also Read: