IIIT-Hyderabad researchers build tool to count roadside trees

Arpit Bahety, research fellow at the Centre for Visual Information Technology (CVIT), IIIT-Hyderabad under the guidance of Prof. CV Jawahar developed the Machine Learning (ML) algorithm that can automatically detect and count the number of trees abutting urban roads.

Hyderabad: Researchers at the International Institute of Information Technology (IIIT)-Hyderabad have come up with a computer vision-assisted tool to count roadside trees beside automatically generating maps showing the extent of urban tree coverage.

So far, trees censuses were done manually in a specified area or using satellite images or drones to get the big picture. This new computer vision-assisted tool could assist urban planners and communities to strategize and take steps on ways to increase tree canopy.

---

Arpit Bahety, research fellow at the Centre for Visual Information Technology (CVIT), IIIT-Hyderabad under the guidance of Prof. CV Jawahar developed the Machine Learning (ML) algorithm that can automatically detect and count the number of trees abutting urban roads. It also generates a colour-code map, detailing the extent of tree coverage on a particular route.

The IIIT-Hyderabad on Tuesday said with video footage of tree-lined streets obtained from Hyderabad and Delhi, the research team trained a Yolo (You Only Look Once) v4 model – a state-of-the-art ML model for real-time object detection.

The ML model was trained to recognise and detect trees based only on their trunks. Alongside tree detection, the algorithm also generates a tree count at regular time intervals while providing a final count at the end of given video footage, it said.

When the tree count is fed into a GPS extraction model, the system refers to an in-house scale developed by the team where > 50 trees per km refers to a very good tree count and < 20 represents very low tree count. These routes are then automatically displayed on a map with corresponding colours such as dark green for a very good tree count and black which is the lowest on the scale, it said.

“Our algorithm addresses the problem of double counting (of tree) that exists in traditional methods of census taking,” Arpit said.

The ML model which was tested in Hyderabad and Surat across various city streets and outskirts found to have an accuracy of 83 per cent.

---

Now you can get handpicked stories from Telangana Today on Telegram everyday. Click the link to subscribe.

Click to follow Telangana Today Facebook page and Twitter .