- Video Surveillance and Tracking Methods
- Advanced Vision and Imaging
- Advanced Image and Video Retrieval Techniques
- Image Processing Techniques and Applications
- Image Enhancement Techniques
- Service-Oriented Architecture and Web Services
- Distributed Control Multi-Agent Systems
- Caching and Content Delivery
- EEG and Brain-Computer Interfaces
- Cloud Data Security Solutions
- Chaos-based Image/Signal Encryption
- Video Analysis and Summarization
- Human Pose and Action Recognition
- IoT-based Smart Home Systems
- Advanced Data Compression Techniques
- Energy Load and Power Forecasting
- IoT and Edge/Fog Computing
- Advanced Steganography and Watermarking Techniques
- Advanced Algorithms and Applications
- Neuroscience and Neural Engineering
- Robot Manipulation and Learning
- Control and Dynamics of Mobile Robots
- Functional Brain Connectivity Studies
- Access Control and Trust
- Advanced Image Fusion Techniques
Nanjing University
2024
Sanjiang University
2015-2024
China Three Gorges University
2010-2023
Nanchang University
2017-2019
First Affiliated Hospital of Nanchang University
2017-2019
Wuhan University of Technology
2004-2016
California State University, San Bernardino
2014-2016
China Energy Engineering Corporation (China)
2016
Nanjing University of Information Science and Technology
2015
University of Bridgeport
2010-2014
Triple-negative breast cancer (TNBC), defined as a tumor subtype that lacks ER, PR, and HER2, shows poor prognosis due to its aggressive biology limited treatment options. Deregulation of Aurora kinase A (Aur-A), member the mitotic serine/threonine family, overactivation mTOR pathway commonly occur in multiple types. We previously found Aur-A activated inhibited autophagy activity cell models. Whether how regulates TNBC are still unclear. Here, we p-mTOR highly expressed positively...
Images captured in foggy weather conditions often fade the colors and reduce contrast of observed objects. An efficient image fusion method is proposed to remove haze from a single input image. First, initial medium transmission estimated based on dark channel prior. Second, adopts an assumption that degradation level affected by each region same, which similar Retinex theory, uses simple Gaussian filter get coarse transmission. Then, pixel-level achieved between The can recover high-quality...
As the wind power dates have strong volatility and intermittency, it is difficult to meet safety stability of system operation. To make full use characteristics data further improve forecasting performance, an innovative hybrid framework proposed response above challenges, which non-stationary weakening technique, sample entropy (SE)-based prediction model allocation, optimized reduced kernel extreme learning machine (RKELM) a novel deep network are integrated befittingly. Initially,...
Aiming at the problem of tracking failure due to target deformation, flipping and occlusion in visual tracking, a template updating algorithm based on image structural similarity is proposed by dynamically adapt changes during specifically, queue used save recent N -frame results, decision made whether update or not score between current results image, if updated, matched from historical as new template. If optimal for subsequent tracking. The feature enhancement module segmentation are also...
With the popularity of mobile services, an effective context-aware service adaptation is becoming more and important for operators. In this paper, we propose a Co-evolution eXtended Classifier System CXCS to perform adaptation. Our key idea learn user context, match rule, provide best suitable services users. Different from previous schemes, our proposed can produce new user's initial classifier population quicken its converging speed. Moreover, it make current predict which should be...
With the popularity of mobile services, an effective context-aware service adaptation is becoming more and important for operators. In this paper, we propose a Co-evolution eXtended Classifier System (CXCS) to perform adaptation. Our key idea learn user context, match rule, provide best suitable services users. Different from previous schemes, our proposed CXCS can produce new user's initial classifier population quicken its converging speed. Moreover, it make current predict which should be...
Background/Aims: CyclinG1 (CycG1) is frequently overexpressed in solid tumors and overexpression of CycG1 promotes cell survival upon paclitaxel exposure by inducing polyploidy. Whether how regulates polyploidization caused small molecular targeted inhibitors remains unclear. Methods: Immunohistochemistry immunoblotting were utilized to examine protein expression. Cell proliferation was measured ATPlite assay, cycle distribution apoptosis flow cytometry and/or DNA fragmentation assays....
Web services provide a standard interaction interface for network-based services. With so many diverse web available over the Internet, can be composed together and delivered to users as package. The way combine is still an open problem, especially considering users' preferences. Users always have vague opinions on these preferences when they choose component In this paper, we propose user-preference-based selection engine compose services, which allows define non-quantifiable factors...
Cloud radiation properties and distribution significantly affect the forecasting accuracy, climate monitoring effectiveness global climate's change.A simple method was proposed to automatically recognize four different sky conditions (cirrus, cumulus, stratus clear sky) by means of extracting some features from visual images that can be used for training classifier.In this paper, texture features, color SIFT were extracted extreme learning machine cloud-type classification under experimental...
This paper presents a highly secure data hiding system (a.k.a. Steganography) in cover image using the low frequency Curvelet domain. The contribution of suggested technique is its high security, because it four secret keys (encryption key, two shuffling keys, key) and only component use transform steganography provides number advantages compared to other techniques such as: 1) Computation time reduction and, 2) are designed handle curves discontinuities small coefficients, so components...
Cloud services have been adopted by more and consumers different types of businesses. An intrinsically difficult problem faced service businesses is that no standard way to evaluate compare cloud services' trustworthiness, while trust has already become a critical factor for provider selection. In this paper, we propose trustworthiness evaluation comparison framework help make customized selection decisions based on trustworthiness. This utilizes fuzzy processing neural network handle...
A neurofeedback system adjusting an individual's attention is effective treatment for attention-deficit/hyperactivity disorder (ADHD). In current studies, accurate measure of the level human one key issues that arouse much interest. This paper proposes a novel optimized complex network method (OCNM) measuring using single-electrode electroencephalography (EEG) signals. time-delay embedding algorithm was used to reconstruct EEG data epochs into nodes OCNM network. Euclidean distances were...
Trustworthiness has become a critical factor for users to choose the most suitable computing services business or personal use. However, no standard exists evaluation and comparison of service trustworthiness. So we try propose trustworthiness framework that can make customized selection decisions based on common sense user-defined specific criteria. This utilizes neural network fuzzy logic handle inaccuracy subjectivity in evaluation. A feedback mechanism is also incorporated entire be...
The biggest challenge in developing an effective online learning environment is to map new technologies proven teaching pedagogies and objectives. Pedagogies of engagement, such as discussions, debates, peer interaction, laboratories other formats fostering student opinions, are often easily implemented the classroom environment, but interactive engagement techniques rarely courses. An e-learning platform should be able promote activities collaborative by engaging students process. In this...
The MPEG-4 video coding standard consists of object based schemes for multimedia and enables content functionalities. Video objects in still pictures or sequences should be first identified before the encoding process starts. An algorithm on information fusion, which can used extraction foreground streams real time is proposed this paper. method efficiently integrates image motion streams. thresholding technique operated HSV space provides a better use color than that traditional RGB space....
Multi-Agent Systems (MAS) are, nowadays, an important research area within Robotics and Artificial Intelligence. Motivated by recent advances in synchronization theory of complex dynamical network, this paper develops a general methods controller design for formation reaching group motion controlling the MAS. A novel model MAS is presented at first. Based on model, method decentralized each agent investigated. The stability properties system also analyzed detail. In control scheme topology...
In this paper, we develop an education-oriented portable brain-controlled robot system, consisting of a multi-parameter bioelectric signal acquisition device and Lego NXT Mindstorms robot. The system is only 45 cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> in volume weighs 70g. section wirelessly acquires brain signals the control remotely navigates via brainwaves through Bluetooth. Teenagers may use low-cost to enjoy pleasure...
The MPEG-4 standard consists of object based coding schemes for multimedia and enables content-based manipulation. An approach on information fusion, which can be used the detection location video objects (people) is proposed in this paper. method efficiently integrates color, edge motion streams. HSV space provides a better use color than that traditional RGB space. Using extracted from region contour adaptation an original to eliminate error. integrating various characteristic features...
Mobile social networks give online networking sites the abilities to extend their services mobile device users. Smart phones and tablets allow users interact with others when they are moving. Policy-based management simplifies of interaction functionalities by establishing policies control various activities involved in these functionalities. To detect resolve potential dynamic conflicts between rules configurations from different administrative domains, a knowledge-based policy analysis...
In this paper, we present a robust and fast object tracking algorithm based on sub-region classifiers compressive tracking. Compared with the original CT algorithm, tracker can improve robustness to occlusion, especially long-term occlusion. Firstly, target region is divided into four sub-regions in fixed mode. Then simple but feasible classification update strategy used for these classifiers. On assumption of rigidity, final location be evaluated by The experiments many challenging image...