Machine Learning, Deep Learning and AI are increasingly being used along with GIS for a number of purposes. Integrating Machine Learning algorithms with ArcGIS provides better and more optimum results in less time.
Satellite images are of different resolution and implementing it successfully is not at all easy. Earlier it took months for reaching the final output. But now due to fast-paced innovations, it takes just one day. But for working with these solutions multiple libraries and project platforms are needed and not all of them support Machine Learning and Deep Learning algorithms and applications.
Esri platform allows a solution-centric approach. AI has seen considerable use in Computer vision and Language Processing. Esri cloud also provides the necessary infrastructure for Machine Learning and Deep Learning.
“Other than map data, there is also IoT. There are weather sensors and camera sensors. A lot of these challenges are solved by Machine Leaning. People nowadays are expecting unprecedented precision. All of this is converging towards GeoAI as location component is very smart”, said Abhay Swarup Mittal, CEO, Skymap Global, at Esri UC Delhi 2018.
In GeoAI, data capturing tool is enhanced by building footprint, road detection. Geospatial analysis and processing tools are enhanced by intelligent decision making support in presence of uncertainty and prediction models of sophisticated phenomenon.
“Pattern recognition is important as a lot of success would depend on it. So the focus should be on data recognition and Object Based Image Analysis which is also called Feature Extraction”, added Abhay Swarup Mittal.
Location is at the heart of merging GIS with Machine Learning and AI. The location component is indispensable and apps and services have to be customized factoring in location.
Cherie Muleh, Harris Geopatial, said, “Location expertise in Machine Learning includes visualization, location services, and movement optimization and location intelligence. Location services include geo-fencing and IoT. Movement optimization includes asset tracking, navigation, real-time movements.”
Daily datasets are available from multiple sources.
“Satellite images with deep learning help us in assessing the changes. Now that government of India has relaxed drone regulations, quality of data would improve”, said Cherie Muleh.
Nvidia and many other companies are doing a lot of research on processing of this data. And of course everything has to be processed using deep learning. More analytics on top of dashboards need to be done.
Deep Learning uses spectral data. Deep learning provides accuracy of more than 95%. With vast amounts of data, Deep Learning can search catalogues of data, which then populates GIS. Deep Learning overcomes the shortcoming with spectral data. It can be created using one dataset and applied to other with different spatial and spectral resolution. Spectral tools are usually pixel based while Deep Learning is object based.
Machine Learning and Deep Learning helps in efficient and faster decision making and better quality image extraction. These tools are available in ArcGIS pro and can be integrated smoothly.
Different demographics and require a particular model. There are multiple models that can be used. One size fit all approach definitely doesn’t work. Output that we have got should be properly structured.
Different modalities of data are used for finding more discrete features.
Titulaire d'un Doctorat en Géographie-Aménagement -Environnement, je travaille sur l'évaluation des politiques publiques de conservation dans les aires protégées au travers de la télédétection satelittale et les SIG. Je suis par ailleurs enseignant-chercheur et consultant en Géomatique. J'ai effectué plusieurs communications à des congrès scientifiques internationaux notamment aux USA, au Canada et en France. J'ai en outre rédigé plusieurs articles dans les revues scientifiques à comité de lecture et animé plusieurs séminaires.