Dharani D

Research Assistant with primary focus on using Machine Learning on multispectral, Hyperspectral and SAR geodatasets for various remote sensing applications and prototypes in emergency response, precision farming, ocean monitoring etc.,

• Using Node-RED as a Low-Code Approach to Model Interaction Logic of Machine-Learning Supported eHMIs for the Virtual Driving Simulator Carla, Automotive UI Conference, 2023, This publication presents a novel low-code methodology utilizing Node-RED for integrating Machine Learning-based VRU detection models with external Human-Machine Interface (eHMI) systems in the CARLA driving simulator, enhancing early integration and UX testing. In this study, we demonstrate the feasibility of using flow-based programming and YOLOv5 deep learning techniques for non-experts to develop and evaluate interactive eHMI scenarios, such as cornering lights and context-aware VRU warnings.
• Performance and Transferability Assessment of Convolutional Neural Network (CNN) Based Building Detection Models for Emergency Response, DLR e-library, 2022, Developed a Convolutional Neural Network (CNN) for the efficient detection of buildings in emergency situations using remote sensing data, significantly aiding rapid response teams through enhanced mapping capabilities demonstrated during the 2021 Germany floods. The work also evaluated the performance and transferability of the CNN model across diverse sensor platforms, proving its robustness and potential for improving real-time decision-making in disaster management and response.
• Explaining the Effects of Clouds on Remote Sensing Scene Classification, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022, Implemented the evaluation pipeline and visualized preliminary results for investigating the impact of clouds and atmospheric disturbances on the accuracy of machine learning-based remote sensing scene classification, enhancing the understanding of challenges in real-world satellite data analysis. The study identifies four distinct types of effects that cloud coverage has on scene classification predictions, providing insights essential for developing more resilient remote sensing applications.

Coursera: Neural Networks & Machine Learning