A5100405123- Remote-Sensing Image Classification
- Advanced Image Fusion Techniques
- Infrared Target Detection Methodologies
- Image Enhancement Techniques
- Advanced Neural Network Applications
- Innovative Human-Technology Interaction
- Glaucoma and retinal disorders
- Mobile and Web Applications
- Usability and User Interface Design
- Currency Recognition and Detection
- Gaze Tracking and Assistive Technology
- Biometric Identification and Security
- Teaching and Learning Programming
- Child Development and Digital Technology
- Handwritten Text Recognition Techniques
- Advanced Authentication Protocols Security
- Advanced Computing and Algorithms
- Mobile Learning in Education
- User Authentication and Security Systems
- Advanced Image Processing Techniques
Zhejiang University
2022
Southeast University
2021
University of Colorado Boulder
2021
National Sun Yat-sen University
2019
State Key Laboratory of Digital Publishing Technology
2019
Ford Motor Company (United States)
2016
Due to various table layouts and styles, detection is always a difficult task in the field of document analysis. Inspired by great progress deep learning based methods on object detection, this paper, we present YOLO-based method for task. Considering large difference between objects natural objects, introduce some adaptive adjustments YOLOv3, including an anchor optimization strategy two post processing methods. For optimization, use k-means clustering find anchors which are more suitable...
When young children create, they are exploring their emerging skills. And when reflect, transforming learning experiences. Yet early childhood play environments often lack toys and tools to scaffold reflection. In this work, we design a stuffed animal robot converse with prompt creative reflection through open-ended storytelling. We also contribute six goals for child-robot interaction design. hybrid Wizard of Oz study, 33 ages 4-5 years old across 10 U.S. states engaged in then conversed...
Object detection is of wide application for its capability in recognizing and locating targets scenes. Under heavy hazy conditions, however, the performance on RGB images will greatly degrade since image contents are polluted. In this work, we propose an imaging system to acquire three-channel shortwave infrared (SWIR) spectrum facilitate object under conditions. The captures SWIR form pseudo color that most suitable detecting objects, such as pedestrians vehicles. Two different types...
We introduce our machine-learning method to remove the fog and haze in image. Our model is based on CycleGAN, an ingenious image-to-image translation model, which can be applied de-hazing task. The datasets that we used for training testing are creatd according atmospheric scattering model. With change of adversarial loss from cross-entropy hinge loss, reconstruction MAE perceptual improve performance measure SSIM value 0.828 0.841 NYU dataset. Middlebury stereo datasets, achieve 0.811,...
This paper presents the design, architecture, and implementation of “MyFord Touch Guide,” a novel, cross-platform mobile app that delivers an interactive experiential learning experience for Ford SYNCTM with MyFord TouchTM in-vehicle infotainment (IVI) system. incorporates production graphical user interface experience. Additionally, it integrates host video tutorials featuring computer-animated character, which offers insightful, self-guided tour essential features functions Guide is based...
Quadriplegia is an extremely serious disease. Due to the impaired function of patients' limbs, it difficult for them move freely by a wheelchair. This dilemma has become huge obstacle rehabilitation disabled, and meanwhile added extra burden society. To improve quality their lives, our team designs vision-controlled automatic wheelchair based on pupil center cornea reflection (PCCR) technique. By constructing multi-sensor real-time intelligent control system, realizes eye-controlled...