Research

Research Interests

My research goal is to develop AI and computer vision systems for real-world applications, with a focus on:

• Computer Vision & Image Processing: Plant growth analysis and phenotyping using multispectral imaging
• Machine Learning: Graph neural networks for spatio-temporal modeling and pedestrian intention prediction
• Generative AI: Anomaly detection and geometric shape generation

Current Projects

Agricultural Imaging and Phenotyping
Developing computer vision and AI-based methods for analyzing multispectral imagery of horticultural crops. This work involves integrating multispectral imaging with low-field magnetic resonance imaging for comprehensive plant phenotyping.

Traffic-Aware Pedestrian Intention Prediction
Introduced a traffic-aware graph neural network architecture (TA-STGCN) for predicting pedestrian intentions in dynamic traffic environments. This work improves performance on the PIE dataset and has applications in autonomous systems and smart cities.

Past Projects

GAN-Based Anomaly Detection
Designed generative adversarial networks with custom loss functions for detecting anomalies in time-series data, with applications in smart meter systems and power grids.

Graph Neural Networks for Spatio-Temporal Modeling
Applied graph neural networks to model and predict human behavior patterns in complex environments.