A5101512187- Advanced Measurement and Detection Methods
- Robotics and Sensor-Based Localization
- 3D Surveying and Cultural Heritage
- Image Processing Techniques and Applications
- Robotic Path Planning Algorithms
- Image and Video Stabilization
- Advanced Image and Video Retrieval Techniques
- Video Surveillance and Tracking Methods
- Advanced Image Fusion Techniques
- Infrared Target Detection Methodologies
- Advanced Sensor and Control Systems
- Adaptive Control of Nonlinear Systems
- Greenhouse Technology and Climate Control
- Chaos control and synchronization
- Neural Networks Stability and Synchronization
- Digital Radiography and Breast Imaging
- Stability and Control of Uncertain Systems
- IoT-based Smart Home Systems
- Embedded Systems and FPGA Design
- Visual perception and processing mechanisms
- Dielectric materials and actuators
- Force Microscopy Techniques and Applications
- Image and Signal Denoising Methods
- Human Pose and Action Recognition
- Advanced Vision and Imaging
Chongqing University
2023-2025
Shanghai University
2005-2021
Anhui University
2020
Sony (Taiwan)
2014
Kunming University of Science and Technology
2014
University of Toronto
2013
Institute of Applied Technology
2011
In robotic micromanipulation, end-effector tips must be first located under microscopy imaging before manipulation is performed. The tip of micromanipulation tools typically a few micrometers in size and highly delicate. all existing systems, the process locating conducted by skilled operator, automation this task has not been attempted. This paper presents technique to automatically locate tips. consists programmed sweeping patterns, motion history image detection, active contour estimate...
In this paper, a fractional-order hyperchaotic system based on dual memristors is represented to analyze the nonlinear dynamic behaviors via replacing two coupled resistors with memristors. It worthy note that has zero eigenvalues and three nonzero eigenvalues, therefore equilibrium plane can be separated into diverse areas which indicates stable or chaotic. The simulation results illustrate initial states have significant impact behaviors, mirrored by phase portraits, bifurcation diagrams,...
Adaptively reconstruction of the skin surface covering skeletons aircrafts, ships, high-speed trains, buildings with steel frames, etc., based on on-site measured point clouds is an important issue in both industry and construction. However, skeleton has characteristics long narrow shape variable structure, resulting a sparse distribution cloud its different areas, which brings great challenges to surface. In this article, method for accurately reconstructing structures arbitrary distributed...
This paper proposes an adaptive linear parameter varying (LPV) fault tolerant control approach with application to unmanned surface vehicle (USV) steering control. First, LPV baseline controller is designed guarantee the acceptable tracking performance and stability of USV under time-varying operating states (distinct surge speed). Then, estimation compensating mechanism combining strategy are synthesized asymptotically stabilize closed-loop system in presence actuator faults. Ultimately,...
In video action recognition, effective spatiotemporal modeling is crucial. However, traditional two-stream methods face challenges in integrating spatial information from RGB images and temporary optical flow without long-range temporal modelling. To address these limitations, we propose the Deep Fusion Module (DFM), which focuses on deep fusion of consists two components. First, an Attention (AFM) to effectively fuse shallow features obtained a network, thereby facilitating integration...
ABSTRACT Currently, the majority of robots equipped with visual‐based simultaneous mapping and localization (SLAM) systems exhibit good performance in static environments. However, practical scenarios often present dynamic objects, rendering environment less than entirely “static.” Diverse objects within pose substantial challenges to precision visual SLAM system. To address this challenge, we propose a real‐time inertial system that extensively leverages environment. First, reject regions...
This letter presents a novel method to estimate the relative poses between RGB-D cameras with minimal overlapping fields of view in panoramic camera system. calibration problem is relevant applications such as indoor 3D mapping and robot navigation that can benefit from 360$^\circ$ field using cameras. The proposed approach relies on descriptor-based patterns provide well-matched 2D keypoints case Integrating matched corresponding depth values, set are constructed calibrate multiple...
A new kind of embedded system for acoustic detection based on wireless transmission GPRS is designed, and the prototype this principle developed. Hardware environment consists processor, transceiver module correlated interface circuits. Supported software platform includes programs data acquisition safety transmission. By functional performance tests, complete, stable, safe reliable, it effectively achieves expected purpose real-time
An attitude estimation of Pan & Tilt system for the small unmanned aerial vehile (SUAV) was gotten by extended Kalman arithmetic. In order to improve available payload SUAV, The MEMS sensors are selected design system. Firstly, magnetometer compensated with environment disturb magnetic field. Secondly, a new KALMAN arithmetic put forward basing on analyzing several kinds sensors. Furthermore, Extended through linearization and discretion. Finally, testings verify has good performance in...
Creating 3D plant models is often a difficult and laborious task. To make it easier more natural, the integration of digital images unorganized point clouds from digitizer provides promising approach for rapeseed model generation. In present study, were incorporated in generation complex rapeseeds at maturity stage. Unorganized image sequences taken different viewpoints using digitizer. The points used automated registration all data sets viewpoints, which pair-wise registration. Later,...
Based on the characteristics of binocular vergence eye movements, A new type method to reduce blind areas caused by double cameras in motion platform was put forward, and a model for control movements established according neural pathways human oculomotor system. With model, it is guaranteed that target can remain public vision field all time during moving. This feature very important moving system, since 3D reconstruction needs have an overlapping vision. To evaluate effectiveness proposed...
Multi-focus image fusion is a technique for obtaining an all-in-focus in which all objects are focus to extend the limited depth of field (DoF) imaging system. Different from traditional RGB-based methods, this paper presents new multi-focus method assisted by sensing. In work, sensor used together with color camera capture images scene. A graph-based segmentation algorithm segment map sensor, and segmented regions guide locate in-focus blocks among source construct reference image. Five...