Embracing The GEO AI Revolution
Meet Anisha Pal
AI Engineer @ Hyperverge
Anisha Pal is an AI engineer in the geospatial division at Hyperverge Inc. Her work involves using computer vision to analyze geospatial data such as aerial and satellite imagery.
She wants to use her knowledge and expertise in the field of AI to create solutions that will help make the world be a greener place.
What inspired you to pursue a career in Artificial Intelligence?
When I was in school I spent the majority of my television time watching Discovery Channel and National Geographic.
There were a variety of shows that I found very intriguing, there were the ones where the scientists described a given scenery and how each animal in the surrounding thinks and reacts to the environment which I found very interesting, there was such a stark difference in how two animals react to similar stimuli.
There were also these shows which explored the human brain through different activities and ones where people were inspired by nature to design certain products.
All of this created a plethora of questions in my head and piqued my interest in problem-solving through inspiration from nature.
During my undergraduate studies, I was desperately looking for the bucket I will fit into and stumbled across Artificial Intelligence, it satisfied those curiosities that I had developed as a kid.
I found a wonderful peer group in college and we worked on various projects related to AI and explored different areas.
The more I dug into this field the more I realized its capability towards solving a wide array of problems.
At this point I realized the effect that AI would have on the world and how it is being harnessed to build long-lasting solutions in any field, it is not just humans who will benefit from AI but the planet in general.
I wanted to contribute towards this and that’s when I decided to pursue a career in AI.
How geospatial technology works?
Geospatial technology is a huge umbrella with several subdomains, there are GIS, remote sensing, and GPS.
Each of these has very different applications, GPS is quite common and our lives are dependent on it now, we cannot imagine a world without GPS, it’s everywhere from helping us decide which route to take for work to aiding marine biologists with their research, we will be lost without it.
GIS data generally involves imagery from different sources such as aerial imagery captured by drones, satellite imagery, thermal imagery, etc. The imagery is analyzed using computer vision techniques to come up with different reasons based on the problem that can be used further in the system.
HD map creation for autonomous vehicles is one technology that requires both GPS data for tracking the position of the vehicle as well as important data points such as lane width, lane type, road signs, etc that are detected using satellite imagery.
Do you have any AI projects that want to share your experience?
So, currently, at Hyperverge we are working on detecting illegal construction for aiding property tax collection using AI-based infrastructure change detection.
The property tax collection process currently involves capturing aerial or satellite imagery and comparing the images from two different years manually to flag changes in the property which are then visited by property tax assessors for further investigation.
This method has a lot of issues ranging from time-consuming and slow to costly and is highly dependent on human error.
AI-driven feature extraction and change detection automate analytics on aerial imagery. By comparing images from two different times, land parcels where there are changes are flagged. These land parcels can then be classified as per the change detected (new building, expansion, swimming pool, etc).
The end product is a map layer that can be imported into the county’s existing software and can be used to plan on-ground assessment visits. Further, land parcels where changes are found can be classified and prioritized as per estimated tax value.
The software can check whether the identified land parcels have the required permits. Priority lists of land parcels where assessors should focus their attention can be auto-generated
Mention any industry where AI could have a big impact? and Why?
AI is already transforming the practice of medicine and I think it will continue to make significant contributions in the medical field.
Apart from assisting doctors in medical diagnosis AI systems can also cure untreatable situations like amputated limbs, prosthetics is a growing market.
Recent projects such as brain implants by Neuralink further go on to prove that the role of AI in the medical field is bigger than we have imagined or are imagining, it doesn’t stop at just general medical diagnosis but reaches into the sci-fi space.
Whether it is a boon or a bane is something we need to think about but either way, AI will create a big impact in this field.
Where do you see yourself in 5 years in the field of AI?
I am very passionate about the conservation of nature and the recent events involving Amazon rainforest fires, Australian bushfires, so many floods and cyclones in India have left me very disturbed.
So I hope in the future I can create a product using AI that can give a very accurate prediction of forest fires or predict oncoming forest fires so that it can be prevented. I am sure forest fire prevention techniques using AI already exist.
I just hope to make one which will require us to never worry about them anymore.
Forest fires are just one thing that is a threat to the environment, I hope AI can help us preserve and restore nature in all ways possible.