IIIT Naya Raipur harnesses drones to detect crop diseases, curb pesticide overuse

The institute claims that the forecasting solution, developed by the faculty of the Department of Electronics and Communication Engineering (ECE) and Computer Engineering (CE), can help farmers tackle crop diseases before they damage crops. It will also curb the overuse of pesticides, which is rampant among farmers due to lack of accurate information.

The forecasting solution was developed in collaboration with the Indira Gandhi Agricultural University in Raipur, which allows IIIT Raipur to use its agricultural fields for testing purposes.

Funding for the project was provided by the Seeds Division of the Department of Science and Technology, Government of India. The project was approved by the Department of Science and Technology in December 2020 and is expected to be completed this year.

Under the project, a drone can be deployed in the field and can capture images if it detects any insect or disease in the crops. The images are then sent directly to the institute’s servers where an image classification model based on convolutional neural networks is used to identify diseases and insects. A convolutional neural network is a deep learning algorithm that can read an image and assign weights to different objects in it so that it can distinguish one from the other.

A prediction analysis tool is used to analyze the problem and alert farmers to it, as well as recommend how much pesticide should be sprayed and on which part of the field. This will reduce the excessive use of pesticides among the farmers.

Anurag Singh, Assistant Professor of ECE at IIIT-Naya Raipur, said, “To train the AI ​​model, we have used drone and smartphone images captured by crops infected with different types of diseases and insects which are commonly found in India. affect crops.

According to Srivishal Tripathi, assistant professor of ECE at IIIT-Naya Raipur, the drones used for the project are internally designed and have in-built Internet of Things (IoT) sensors to capture time and position data. The idea is to get information about when the image was captured and what the GPS location was. “This will help us identify the owner of the land and suggest measures accordingly,” he said.

Singh explained, manually monitoring large tracts of land can be a challenge. Using drones, farmers can conduct inspections more frequently and in less time. Also, it is difficult to differentiate between insects that look similar but are actually not.

In the Union Budget for 2022, Finance Minister Nirmala Sitharaman had called for widespread application of drones in agriculture and other sectors through Drone-as-a-Service.