Research Projects

Rubber plantations are a significant part of India's agricultural economy, especially in Kerala. One of the critical challenges in rubber production is optimizing the tapping process, which directly influences latex yield and tree health. Traditional tapping relies on manual expertise, often resulting in bark damage, inconsistent latex output, and early tree mortality. This proposal aims to develop an AI-powered, multimodal computer vision system that analyzes RGB, thermal, and 3D visual data to guide tapping decisions intelligently. The system will utilize deep learning and transformer-based models to detect bark stress, assess tapping scar quality, estimate latex flow potential, and suggest optimal tapping depth, frequency, and location. The approach will be implemented through a decision support system (DSS) deployable on mobile devices and drones. Kerala, contributing over 75% of India’s rubber, offers an ideal testbed for field validation. The project will be developed at IIIT Kottayam, located in the heart of Kerala’s rubber belt, in collaboration with NIT Puducherry, Kerala Agricultural University (KAU), and the Rubber Board of India. Over a 3-year period, this project will integrate advanced AI with real-world agronomy to empower farmers and estate managers with a sustainable, data-driven tapping advisory tool. The outcomes will significantly improve latex yield, reduce tree stress, and introduce scalable precision agriculture tools in plantation ecosystems.