A5014094093- Robotics and Sensor-Based Localization
- Advanced Vision and Imaging
- 3D Surveying and Cultural Heritage
- Advanced Image and Video Retrieval Techniques
- Robotic Path Planning Algorithms
- Soft Robotics and Applications
- Robotic Mechanisms and Dynamics
- Indoor and Outdoor Localization Technologies
- Image and Signal Denoising Methods
- Image and Object Detection Techniques
- Remote Sensing and LiDAR Applications
- Advanced Image Fusion Techniques
- Remote-Sensing Image Classification
- Anatomy and Medical Technology
- Augmented Reality Applications
- Sparse and Compressive Sensing Techniques
- Environmental Impact and Sustainability
- Complex Systems and Time Series Analysis
- Chemical Thermodynamics and Molecular Structure
- Fault Detection and Control Systems
- Advanced MRI Techniques and Applications
- Medical Image Segmentation Techniques
- Advanced Radiotherapy Techniques
- Mechanical Engineering and Vibrations Research
- Adsorption, diffusion, and thermodynamic properties of materials
Yanshan University
2020-2025
Shanxi University
2022-2025
United Imaging Healthcare (China)
2022-2024
Tsinghua University
2019-2024
National University of Defense Technology
2022-2024
University of Michigan
2020-2022
FX Palo Alto Laboratory
2020-2021
Nanjing Forestry University
2021
University of Technology Sydney
2017-2020
Institute of Remote Sensing and Digital Earth
2014-2015
Extracting road networks from very-high-resolution (VHR) aerial and satellite imagery has been a long-standing problem. In this article, neural-dynamic tracking framework is proposed to extract based on deep convolutional neural (DNN) finite state machine (FSM). Inspired by autonomous mobile systems, the authors train DNN recognize pattern of input data, which an image patch extracted in detection window centred at current location tracker. The predefined according environment associated...
In this letter, we investigate the convergence and consistency properties of an invariant-extended Kalman filter (RI-EKF) based simultaneous localization mapping (SLAM) algorithm. Basic algorithm are proven. These proofs do not require restrictive assumption that Jacobians motion observation models need to be evaluated at ground truth. It is also shown output RI-EKF invariant under any stochastic rigid body transformation in contrast SO (3) EKF SLAM (SO(3)-EKF) only deterministic...
Real-time simultaneous localization and dense mapping is very helpful for providing virtual reality augmented surgeons or even surgical robots. In this letter, we propose MIS-SLAM: A complete real-time large-scale deformable SLAM system with stereoscope in minimal invasive surgery (MIS) based on heterogeneous computing by making full use of CPU GPU. Idled used to perform ORB-SLAM robust global pose. Strategies are taken integrate modules from We solved the key problem raised previous work,...
Service robots should be able to operate autonomously in dynamic and daily changing environments over an extended period of time. While Simultaneous Localization And Mapping (SLAM) is one the most fundamental problems for robotic autonomy, existing SLAM works are evaluated with data sequences that recorded a short In real-world deployment, there can out-of-sight scene changes caused by both natural factors human activities. For example, home scenarios, objects may movable, replaceable or...
This letter proposes a new image-based localization framework that explicitly localizes the camera/robot by fusing Convolutional Neural Network (CNN) and sequential images' geometric constraints. The camera is localized using single or few observed images training with 6-degree-of-freedom pose labels. A Siamese network structure adopted to train an image descriptor network, visually similar candidate in set retrieved localize testing geometrically. Meanwhile, probabilistic motion model...
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the agent causing global pandemic of COVID-19. SARS-CoV-2 genome encodes a main protease (nsp5, also called Mpro) and papain-like (nsp3, PLpro), which are responsible for processing viral polyproteins to assemble functional replicase complex. In this study, we found that Mpro can cleave human MAGED2 other mammalian orthologs at Gln-263. Moreover, SARS-CoV MERS-CoV MAGED2, suggesting cleavage by an evolutionarily conserved...
This paper introduces Code-Vision, a benchmark designed to evaluate the logical understanding and code generation capabilities of Multimodal Large Language Models (MLLMs). It challenges MLLMs generate correct program that fulfills specific functionality requirements based on given flowchart, which visually represents desired algorithm or process. Code-Vision comprises three subsets: HumanEval-V, Algorithm, MATH, MLLMs' coding abilities across basic programming, algorithmic, mathematical...
In the context of sustainable development, water resources, energy, and carbon emissions are pivotal factors influencing rational planning economic development secure establishment ecological barriers. As a core food production area, how can Great River Basin balance pressure on “water–energy–carbon” system (WEC) to realize coordinated “nature–society–economy”? Taking Yellow in China as research object, this paper explores coupling characteristics virtual transfer trends WEC agricultural...
In minimal invasive surgery, it is important to rebuild and visualize the latest deformed shape of soft-tissue surfaces mitigate tissue damages. This letter proposes an innovative Simultaneous localization mapping (SLAM) algorithm for deformable dense reconstruction using a sequence images from stereoscope. We introduce warping field based on embedded deformation nodes with three-dimensional (3-D) shapes recovered consecutive pairs stereo images. The estimated by deforming last updated model...
In stereoscope-based minimally invasive surgeries (MISs), dense stereo matching plays an indispensable role in 3-D shape recovery, AR, VR, and navigation tasks. Although numerous deep neural network (DNN) approaches are proposed, the conventional prior-free still popular industry because of lack open-source annotated dataset limitation task-specific pretrained DNNs. Among algorithms, there is no successful real-time algorithm none GPU environment for MIS. This article proposes first...
In order to improve the motion stability of a pointing mechanism, trajectory planning and optimization were conducted. To obtain better stability, jerk is required continuous. Therefore, higher polynomial or B-spline are needed calculation amount increases in planning. A novel hybrid interpolation algorithm "7-order + improved quartic uniform curve 7-order polynomial" was proposed for planning, which cannot only ensure passes through all path points but also From first point second from...
The accelerating impact of climate change on giant panda (Ailuropoda melanoleuca) habitats have become an international research topic. Recently, many studies also focused medium-sized mountain ranges or entire to predict how will as the warms, but few say in detail what do where focus efforts. To fill this gap, paper presents a new method take comprehensive, fine-scale evaluations incorporating change, human disturbance, and current conservation networks translate them into practical...
A simulated annealing (SA) based variable weighted forecast model is proposed to combine and weigh local chaotic model, artificial neural network (ANN), partial least square support vector machine (PLS-SVM) build a more accurate model. The hybrid was built multistep ahead prediction ability tested on daily MSW generation data from Seattle, Washington, the United States. proved produce reliable results degrade less in longer predictions than three individual models. average one-week step has...
Sliding-window-based low-rank matrix approximation (LRMA) is a technique widely used in hyperspectral images (HSIs) denoising or completion. However, the uncertainty quantification of restored HSI has not been addressed to date. Accurate denoised facilitates applications such as multisource multiscale data fusion, assimilation, and product since these require an accurate approach describe statistical distributions input data. Therefore, we propose prior-free closed-form element-wise method...
In this study, a univariate local chaotic model is proposed to make one-step and multistep forecasts for daily municipal solid waste (MSW) generation in Seattle, Washington. For MSW prediction with long history data, forecasting was created based on nonlinear dynamic method called phase-space reconstruction. Compared other predictive models, such as artificial neural network (ANN) partial least square–support vector machine (PLS-SVM), commonly used linear seasonal autoregressive integrated...
This article reports on recent progress in robot perception and control methods developed by taking the symmetry of problem into account. Inspired existing mathematical tools for studying structures geometric spaces, sensor registration, state estimator, provide indispensable insights formulations generalization robotics algorithms to challenging unknown environments. When combined with computational learning hard-to-measure quantities, symmetry-preserving unleash tremendous performance. The...
This work reports a novel Bundle Adjustment (BA) formulation using Reproducing Kernel Hilbert Space (RKHS) representation called RKHS-BA. The proposed is correspondence-free, enables the BA to use RGB-D/LiDAR and semantic labels in optimization directly, provides generalization for photometric loss function commonly used direct methods. RKHS-BA can incorporate appearance within continuous spatial-semantic functional that does not require via image pyramids. We demonstrate its applications...