×

img Accessibility Controls

Search

Filters
Loading...

Loading filters...

8846 result(s) found
Enhancing Head and Neck Cancer Treatment: Deep Learning for Efficient Tumor Contour Detection from Limited CT Scan Data.

We plan to address the problem of the tumor volume and contour detection from the CTimages as a semantic segmentation problem. In this project, first, we utilize deep semantic segmentation models such as UNet [1] and SegResnet for tumor volume segmen

Section: Academic Innovations (Innovations)
Large-Scale Manually Annotated LiDAR Datasets from Indian cities to Facilitate Semantic Segmentation

The research methodology consists of a series of sequential steps to achieve the objectives. The key stages of the methodology include data collection, manual annotation, training of segmentation models, and development of a web-based system to host

Section: Academic Innovations (Innovations)
Drone Enabled Deep Learning Analytics for Monitoring and Yield Assessment of Commercial Crops.

Data Construction Phase: Capturing and Constructing Panorama Images using Drone Imagery

Section: Academic Innovations (Innovations)
Machine-Learning based Battery Cell Balancing Mechanism for Two-Wheeled Electric Bikes for Improving Lifespan of battery pack

Two-wheeled electric bikes are becoming increasingly popular due to their environmental friendly, energy efficiency, and low operating costs. The battery pack is the most costly equipment that accounts for about 40%-50% of electric bike cost Therefor

Section: Academic Innovations (Innovations)
Development of a High-frequency Inductor based Power Module to Improve the eBike Battery Life – A low cost and reliable solution

A CFPM with controlled inductor current can supply or absorb current during driving and during regenerative braking respectively in eBike applications. A CFPM connected in cascade configuration with eBike battery is presented

Section: Academic Innovations (Innovations)
ARAS in unstructured environments: Understanding the driver behaviour and driving style of a two-wheeler ride

The proposed methodology involves the integration of multimodal sensing technologies, including GPS, accelerometers, gyroscopes, and cameras, into a mobile application platform. Machine/deep learning algorithms will be employed to analyze sensor data

Section: Academic Innovations (Innovations)
Enhancing Road Safety and Congestion-free Traffic Routing through Reinforcement Learning based Multi-Agent Planning.

Optimizing traffic flow by adjusting traffic lights at junctions is crucial for reducing queue length and minimizing vehicle delay time, thus improving traffic management. Action, environment conditions, reward patterns and state transitions forms th

Section: Academic Innovations (Innovations)
Simulation specific Scenario Aware Datawith Multi-modal Analytics for decisionmaking using hybrid complex reasoning

Safety and improved drivability is always of high priority in Autonomous vehicles or bringing assisted technologies in AVs. It is highly challenging as it involves multiple considerations like driver’s drivability patterns, speed, direction of trav

Section: Academic Innovations (Innovations)
Behavior-Conditioned Safety-Aware​ Synthetic Scenario Generation for the​ Indian Traffic

The idea of the proposed work can be summarized in Figure 1. First, we need to collect data on how traffic flows in the real world. This data helps us understand how diverse agent types move and interact with each other. The drone image dataset may b

Section: Academic Innovations (Innovations)
Integrate and test system-on-chip UMA (Universal Multifunction Accelerator) board for a proprietary pupillometry device used in clinics for pupil examination.

This project focuses on integrating and validating a System-on-Chip UMA board within a clinical pupillometry device used for precise pupil examination. It involves hardware–software interfacing, sensor calibration, communication protocol design, an

Section: Academic Innovations (Innovations)
arrowtop
Latest Updates
Loading…