A5101911937- Advanced Neural Network Applications
- Drilling and Well Engineering
- Tunneling and Rock Mechanics
- CCD and CMOS Imaging Sensors
- Industrial Vision Systems and Defect Detection
- Advanced Image and Video Retrieval Techniques
- COVID-19 diagnosis using AI
- Advanced machining processes and optimization
- Context-Aware Activity Recognition Systems
- Underwater Vehicles and Communication Systems
- Environmental and Industrial Safety
- Rock Mechanics and Modeling
- Domain Adaptation and Few-Shot Learning
- Advanced Battery Technologies Research
- Diverse Industrial Engineering Technologies
- Neural Networks and Applications
- Advanced materials and composites
- Gait Recognition and Analysis
- Advanced Power Generation Technologies
- Robotics and Sensor-Based Localization
- Electric Vehicles and Infrastructure
- Inertial Sensor and Navigation
- Electric and Hybrid Vehicle Technologies
- Indoor and Outdoor Localization Technologies
Shijiazhuang Tiedao University
2020-2024
National Changhua University of Education
2022
University of California, San Diego
2021
University of Science and Technology of China
2019
Shanghai Jiao Tong University
2008
Computer vision has achieved impressive progress in recent years. Meanwhile, mobile phones have become the primary computing platforms for millions of people. In addition to phones, many autonomous systems rely on visual data making decisions, and some these limited energy (such as unmanned aerial vehicles also called drones robots). These batteries, efficiency is critical. This paper serves following two main purposes. First, examine state art low-power solutions detect objects images....
IMU based inertial tracking plays an indispensable role in many mobility centric tasks, such as robotic control, indoor navigation and virtual reality gaming. Despite its mature application rigid machine (e.g., robot aircraft), human users via mobile devices remains a fundamental challenge due to the intractable gait/posture patterns. Recent data-driven models have tackled sensor drifting, one key issue that plagues tracking. However, these systems still assume are held or attached user body...
Abstract Considering the serious wear of disk cutters in tunnel project and low prediction accuracy existing methods, a new method for from an energy perspective was proposed. Based on contact mechanics, rock breaking process analyzed, combined action sliding rolling friction considered. Furthermore, model established, which realized predicted under different geological conditions throughout construction line. In addition, considering influence penetration model, values penetrations were...
Neural networks have shown great performance in cognitive tasks. When deploying network models on mobile devices with limited resources, weight quantization has been widely adopted. Binary obtains the highest compression but usually results big accuracy drop. In practice, 8-bit or 16-bit is often used aiming at maintaining same as original 32-bit precision. We observe different layers sensitivity of quantization. Thus judiciously selecting precision for layers/structures can potentially...
Computer vision has achieved impressive progress in recent years. Meanwhile, mobile phones have become the primary computing platforms for millions of people. In addition to phones, many autonomous systems rely on visual data making decisions and some these limited energy (such as unmanned aerial vehicles also called drones robots). These batteries efficiency is critical. This article serves two main purposes: (1) Examine state-of-the-art low-power solutions detect objects images. Since...
As the main rock-breaking tool of tunnel boring machine, wear disc cutter is affected by geological conditions, equipment factors, and tunneling parameters when it interacts with rock. Because complex factors affecting wear, difficult to accurately predict cutter. In this study, mechanism force were analyzed, a theoretical prediction model was established based on friction work principle. The in determined simulation, method proposed. Finally, verified field data. results show that average...
Neural networks have shown great performance in cognitive tasks. When deploying network models on mobile devices with limited computation and storage resources, the weight quantization technique has been widely adopted. In practice, 8-bit or 16-bit is mostly likely to be selected order maintain accuracy at same level as 32-bit floating-point precision. Binary quantization, contrary, aims obtain highest compression cost of much bigger drop. Applying different precision layers/structures can...
The Low-Power Image Recognition Challenge (LPIRC, https://rebootingcomputing.ieee.org/lpirc) is an annual competition started in 2015. identifies the best technologies that can classify and detect objects images efficiently (short execution time low energy consumption) accurately (high precision). Over four years, winners' scores have improved more than 24 times. As computer vision widely used many battery-powered systems (such as drones mobile phones), need for low-power will become...
Abstract The purpose of this research is to develop a cheap, accurate and autonomous quadcopter. This quadcopter contains microcontroller, gyroscope, accelerometer, barometer, electronic governor brushless motor. At present, the noise removal filter sold on market complementary filter, its disadvantage that it will produce distortion when attitude changes at large angle. Therefore, in order improve shortcomings quadcopter, we use Kalman replace filters achieve goal. same time, through...