A5024158572- Robotics and Sensor-Based Localization
- Topological and Geometric Data Analysis
- Advanced Vision and Imaging
- Muscle activation and electromyography studies
- Smart Agriculture and AI
- Advanced Image and Video Retrieval Techniques
- Non-Invasive Vital Sign Monitoring
- Respiratory and Cough-Related Research
- Remote Sensing in Agriculture
- Prosthetics and Rehabilitation Robotics
- Leaf Properties and Growth Measurement
- 3D Surveying and Cultural Heritage
- Insect and Arachnid Ecology and Behavior
- Voice and Speech Disorders
- Video Surveillance and Tracking Methods
- Advanced Sensor and Energy Harvesting Materials
- Plant and Biological Electrophysiology Studies
- Context-Aware Activity Recognition Systems
- Energy Efficient Wireless Sensor Networks
- Speech Recognition and Synthesis
- Micro and Nano Robotics
- Image Retrieval and Classification Techniques
- Image Processing and 3D Reconstruction
- Greenhouse Technology and Climate Control
- ECG Monitoring and Analysis
North Carolina State University
2016-2025
North Central State College
2019-2024
Institute of Arctic and Alpine Research
2017
University of Colorado Boulder
2017
National Institute of Neurological Disorders and Stroke
2016
Duke University
2016
Rice University
2016
University of Pennsylvania
2016
Indiana University Bloomington
2016
University of North Carolina at Chapel Hill
2010-2011
In this work, dynamically tunable, superlyophobic surfaces capable of undergoing a transition from profound behavior to almost complete wetting have been demonstrated for the first time. initial state, with no voltage applied, these exhibit contact angles as high 150 degrees wide variety liquids surface tensions ranging 21.8 mN/m (ethanol) 72.0 (water). Upon application an electrical voltage, state is observed. We examined experimentally and theoretically nature transitions. The reported...
In this paper, we propose and demonstrate a novel wireless camera network system, called CITRIC. The core component of system is new hardware platform that integrates camera, frequency-scalable (up to 624 MHz) CPU, 16MB FLASH, 64MB RAM onto single device. device then connects with standard sensor mote form mote. design enables in-network processing images reduce communication requirements, which has traditionally been high in existing networks centralized processing. We also back-end...
Reliable environmental context prediction is critical for wearable robots (e.g., prostheses and exoskeletons) to assist terrain-adaptive locomotion. This article proposed a novel vision-based framework lower limb simultaneously predict human's multiple forecast windows. By leveraging the Bayesian neural networks (BNNs), our can quantify uncertainty caused by different factors observation noise, insufficient or biased training) produce calibrated predicted probability online decision-making....
We present a new method for continuously and accurately estimating the shape of continuum robot during medical procedure using small number X-ray projection images (e.g., radiographs or fluoroscopy images). Continuum robots have curvilinear structure, enabling them to maneuver through constrained spaces by bending around obstacles. Accurately robot's over time is crucial success procedures that require avoidance anatomical obstacles sensitive tissues. Online estimation complicated...
Environmental context prediction is important for wearable robotic applications, such as terrain-adaptive control. System efficiency critical robots, in which system resources (e.g., processors and memory) are highly constrained. This article aims to address the of real-time environmental lower limb prostheses. First, we develop an uncertainty-aware frame selection strategy that can dynamically select frames according motion uncertainty captured by Bayesian neural networks (BNNs) environment...
Accurate prediction of sweet potato yield is crucial for effective crop management. This study investigates the use vegetation indices (VIs) extracted from multispectral images acquired by a small unmanned aerial vehicle (UAV) throughout growing season, along with in situ-measured plant physiological parameters, to predict yield. The data acquisition process through UAV field imaging discussed detail extraction bands that we as features. experiment designed combination different nitrogen...
Recent years have seen significant technological advancements in precision farming and plant phenotyping. Remote sensing along with deep learning (DL) techniques can increase phenotyping efficiency help on-farm decision making rapid stress detection. In this work, we use these to evaluate drought soybean plants, a crop whose yield is significantly affected by water availability. Images were taken from high vantage the field at various times throughout day. Each image given wilting score...
The growing popularity of 3-D movies has led to the rapid development numerous affordable consumer displays. In contrast, technology generate content lagged behind considerably. spite significant improvements quality imaging devices, accuracy algorithms that data, and hardware available render such calibrate, reconstruct, then visualize data remain difficult use, extremely noise sensitive, unreasonably slow. this paper, we present a multi-camera system creates highly accurate (on order...
Mapping and exploration are essential tasks for swarm robotic systems. These become extremely challenging when localization information is not available. In this paper, we explore how stochastic motion models weak encounter can be exploited to learn topological about an unknown environment. Our system behavior mimics a probabilistic model of cockroaches, as it inspired by current biobotic (cyborg insect) We employ tools from algebraic topology extract spatial the environment based on...
The real-time detection of drought stress has major implications for preventing cash crop yield loss due to variable weather conditions and ongoing climate change. most widely used indicator sensitivity/tolerance in corn soybean is the presence or absence leaf wilting during periods water stress. We develop a low-cost automated system using computer vision coupled with machine learning (ML) algorithms that document response soybeans field crops. Using ML, we predict status plants more than...
Computer vision has shown promising potential in wearable robotics applications (e.g., human grasping target prediction and context understanding). However, practice, the performance of computer algorithms is challenged by insufficient or biased training, observation noise, cluttered background, etc. By leveraging Bayesian deep learning (BDL), we have developed a novel, reliable vision-based framework to assist upper limb prosthesis during arm reaching. This can measure different types...
Physiological and kinematic signals from humans are often used for monitoring health. Several processes of interest (e.g., cardiac respiratory processes, locomotion) demonstrate periodicity. Training models inference on these detection anomalies, extraction biomarkers) require large amounts data to capture their variability, which not readily available. This hinders the performance complex models. In this work, we introduce a methodology improving such by incorporating phase-based...
Abstract Physical inspection and sorting of foraminifera is a necessity in many research labs, as serve paleoenvironmental chronostratigraphic indicators. In order to gain counts species from samples, analyze chemical compositions, or extract morphological properties foraminifera, labs require human time effort handling these microscopic fossils. The presented work describes Forabot, an open‐source system which can physically manipulate individual for imaging isolation with minimal...
Utilizing the latest neural engineering developments, researchers have enabled biobotic insects that function as search-and-rescue agents to help map under-rubble environments and locate survivors hazardous conditions. The Web extra at http://youtu.be/oJXEPcv-FMw is a video in which authors Alper Bozkurt, Edgar Lobaton, Mihail Sichitiu demonstrate acoustic steering of roach biobots search for disaster victims trapped under rubble.
Lower-limb robotic prosthetics can benefit from context awareness to provide comfort and safety the amputee. In this work, we developed a terrain identification surface inclination estimation system for prosthetic leg using visual inertial sensors. We built dataset which images with high sharpness are selected IMU signal. The used is also computed simultaneously. With such information, control of robotized be adapted changes in its surrounding.
Predicting the user's intended locomotion mode is critical for wearable robot control to assist seamless transitions when walking on changing terrains.Although machine vision has recently proven be a promising tool in identifying upcoming terrains travel path, existing approaches are limited environment perception rather than human intent recognition that essential coordinated operation.Hence, this study, we aim develop novel system fuses gaze (representing user intent) and (capturing...
Smart camera networks have recently emerged as a new class of sensor network infrastructure that is capable supporting high-power in-network signal processing and enabling wide range applications. In this article, we provide an exposition our efforts to build low-bandwidth wireless platform, called CITRIC, its applications in smart networks. The platform integrates camera, microphone, frequency-scalable (up 624 MHz) CPU, 16 MB FLASH, 64 RAM onto single device. device then connects with...