Madhya Pradesh has become the first state to implement an AI-based real-time alert system on a pilot basis for active forest management using satellite images, mobile feedback, and machine learning, an official said on Tuesday.
The Artificial Intelligence system will enable the forest department to detect land encroachment, land use change, and forest degradation.
"Madhya Pradesh has become the first state in the country to implement an AI-based real-time forest alert system. This historic step has been taken towards active forest management in the state. This system works with the help of satellite images, mobile feedback and machine learning," the official said.
This system is being implemented as a pilot project in five sensitive forest divisions including Shivpuri, Guna, Vidisha, Burhanpur and Khandwa, which have reported several encroachment and tree felling incidents.
It will be implemented at the state level, the official added.
He said that based on Google Earth Engine, the AI alert system analyses multi-temporal satellite data and identifies land use changes using a custom AI model.
Every possible change is sent to the field staff through a mobile app so that they can confirm it by visiting the site.
Guna Divisional Forest Officer (DFO) Akshay Rathore, who conceptualised the AI system for forest management, said this is the first time the forest department has combined satellite, AI, and field feedback in a continuous cycle, which improves itself over time.
The system is implemented under the leadership and institutional support of the Head of Forest Force (HoFF), Aseem Shrivastava, and Additional Principal Chief Forest Conservator of IT B.S. Annigeri.
"This system empowers forest staff to monitor and take immediate action. Alert generation and feedback process includes initial alert generation using Google Earth Engine to compare satellite images of three dates, identify changes in crops, wasteland, construction etc," Rathore said.
An official stated that each alert includes over 20 features, such as polygon alerts triggered by notable pixel changes, field verification for mobile app alerts, and uploads by field staff, including GPS-tagged photos, voice notes, and comments.
It also involves data enrichment with indexes like NDVI, SAVI, EVI, and SAR attributes.
The new process involves live monitoring on the Divisional Forest Officer's dashboard, showing real-time alerts categorised by beat and field posts, with filters for date, density, and area.
Alerts will be sent to a mobile app for field staff to take on-site action. The app will enable submission of survey data, including images, GPS, and voice recordings, and incorporate features like geo-fencing and distance measurement, the official said.
Except for the headline, this story has not been edited by The Telegraph Online staff and has been published from a syndicated feed.