A5101662204- Digital Media Forensic Detection
- Internet Traffic Analysis and Secure E-voting
- Advanced Steganography and Watermarking Techniques
- Spam and Phishing Detection
- Heart Rate Variability and Autonomic Control
- Advanced Vision and Imaging
- Seismic Imaging and Inversion Techniques
- Advanced Image and Video Retrieval Techniques
- Video Surveillance and Tracking Methods
- Stroke Rehabilitation and Recovery
- Organizational Leadership and Management Strategies
- 3D Surveying and Cultural Heritage
- Hydraulic Fracturing and Reservoir Analysis
- EEG and Brain-Computer Interfaces
- Transcranial Magnetic Stimulation Studies
- Generative Adversarial Networks and Image Synthesis
- Advanced Malware Detection Techniques
- Image Retrieval and Classification Techniques
- Hydrocarbon exploration and reservoir analysis
- E-commerce and Technology Innovations
- Complex Systems and Time Series Analysis
- Image Processing and 3D Reconstruction
- Cardiovascular Health and Risk Factors
- Non-Invasive Vital Sign Monitoring
- Image Enhancement Techniques
Hunan University of Science and Technology
2024
Tsinghua University
2013-2023
Chongqing University
2004-2020
Beijing Institute of Technology
2020
Chongqing University of Technology
2020
Chinese Academy of Medical Sciences & Peking Union Medical College
2019
Research Institute of Petroleum Exploration and Development
2009-2019
China-Japan Friendship Hospital
2019
Chengdu Second People's Hospital
2019
Chongqing Medical University
2019
We propose a method to recover the shape of 3D room from full-view indoor panorama. Our algorithm can automatically infer collection partially oriented superpixel facets and line segments. The core part is constraint graph, which includes lines superpixels as vertices, encodes their geometric relations edges. A novel approach proposed perform reconstruction based on graph by solving all constraints constrained linear least-squares. selected used for are identified using an occlusion...
Online underground economy is an important channel that connects the merchants of illegal products and their buyers, which also constantly monitored by legal authorities. As one common way for evasion, buyers together create a vocabulary jargons (called "black keywords" in this paper) to disguise transaction (e.g., "smack" street name "heroin" [1]). Black keywords are often "unfriendly" outsiders, created either distorting original meaning words or tweaking other black keywords....
Promotional infection is an attack in which the adversary exploits a website's weakness to inject illicit advertising content. Detection of such challenging due its similarity legitimate activities. An interesting observation we make our research that almost always incurs great semantic gap between infected domain (e.g., university site) and content it promotes selling cheap viagra). Exploiting this gap, developed semantic-based technique, called Semantic Inconsistency Search (SEISE), for...
The popularity of online gambling could bring negative social impact, and many countries ban or restrict gambling. Taking China for example, violates Chinese laws hence is illegal. However, illegal websites are still thriving despite strict restrictions, since they able to make tremendous illicit profits by trapping cheating players. In this paper, we conduct the first deep analysis on targeting unveil its profit chain. After successfully identifying more than 967,954 suspicious websites,...
With the growing attention on multimedia security and rapid development of natural language processing technologies, various linguistic steganographic algorithms based automatic text generation technology have been proposed increasingly, which brings great challenges in maintaining cyberspace. The prevailing steganalysis methods neural networks only conduct with feature vectors from last layer network, may be insufficient for steganalysis. In this paper, we propose a scheme densely connected...
Objectives: To investigate the feasibility of Electroencephalogram (EEG) power spectrum analysis as a noninvasive method for monitoring intracranial pressure (ICP). Methods: The EEG signals were recorded in 62 patients (70 cases) with central nervous system (CNS) disorders our hospital. By using self-designed software, was conducted and index (PI) calculated automatically. Intracranial measured by lumbar puncture (LP). Results: We found significant negative correlation between PI ICP ( r =...
Steganalysis of the Quantization Index Modulation (QIM) steganography in VoIP (Voice-over IP) stream is conducted this research. a popular media streaming and communication service on Internet. QIM makes it possible to hide secret information streams. Detecting short low embedding rates samples remains an unsolved challenge. Recently, neural network models have been demonstrated be capable achieving remarkable performances successfully applied many different tasks. The mainstream...
Extracting correlation features between codes-words with high computational efficiency is crucial to steganalysis of Voice over IP (VoIP) streams. In this paper, we utilized attention mechanisms, which have recently attracted enormous interests due their highly parallelizable computation and flexibility in modeling sequence, tackle problem Quantization Index Modulation (QIM) based steganography compressed VoIP stream. We design a light-weight neural network named Fast Correlation Extract...
In this letter, we present a novel and extremely fast steganalysis method of Voice over IP (VoIP) streams, driven by the need for quick accurate detection possible steganography in VoIP streams. We firstly analyzed correlations carriers. To better exploit correlation code-words, mapped vector quantization code-words into semantic space. order to achieve high efficiency, only one hidden layer is utilized extract between these code-words. Finally, based on extracted features, used softmax...
With the rapid development of Natural Language Processing (NLP) technologies, text steganography methods have been significantly innovated recently, which poses a great threat to cybersecurity. In this paper, we propose novel attentional LSTM-CNN model tackle steganalysis problem. The proposed method firstly maps words into semantic space for better exploitation feature in texts and then utilizes combination Convolutional Neural Networks (CNNs) Long Short-Term Memory (LSTM) recurrent neural...
The traditional Zernike moment-based method is often used for subpixel-level edge detection, but the algorithm still requires manual tuning on selecting threshold, not only inefficient, also leads to wrong detection of and affecting measurement accuracy. In order overcome this weakness, paper combines original with Otsu adaptive threshold a new fast proposed. This can directly obtain optimal step gray detect edge. time significantly shortened while ensuring results. calculates error caused...
Poststroke depression can lead to functional dependence, cognitive impairment and reduced quality of life. The aim this study was evaluate the effects a percutaneous mastoid electrical stimulator (PMES) plus antidepressants on poststroke function.This prospective, randomized, double-blind, sham-controlled study. A total 258 clinically depressed ischaemic stroke patients within 14 days index were randomly assigned PMES antidepressant (PMES group, N = 125) sham (sham 133) groups. All underwent...
With the rapid growth of panorama photography applications such as Android Photo Sphere developed by Google, convenience creating panoramic images has made it very popular in recording omni-directional information. In this paper we address problem room structure reconstruction from an indoor 360° panorama, and a system combining automatic reasoning semi-interactive refinement is proposed. The first calibrates camera performs perspective sampling to transform original into multiple projected...
Electroencephalogram (EEG) is generally used in Brain-Computer Interface (BCI) applications to measure the brain signals. However, multichannel EEG signals characterized by unrelated and redundant features will deteriorate classification accuracy. This paper presents a method based on common spatial pattern (CSP) for feature extraction support vector machine with genetic algorithm (SVM-GA) as classifier, GA optimize kernel parameters setting. The proposed performed data set Iva of BCI...
Stroke has been shown to cause cardiac autonomic dysfunction. Depression is common complication after acute ischemic stroke (AIS). The purpose of this study was investigate whether decreased heart rate variability (HRV) related poststroke depression (PSD) in patients with mild-moderate AIS.In study, we assessed function within 72 h from symptom onset by fractal dimension (FD). 503 (mean age 65.93 ± 10.19) AIS underwent FD test admission. Depressive symptoms were using 17-item Hamilton Rating...
Previous VoIP steganalysis methods face great challenges in detecting speech signals at low embedding rates, and they are also generally difficult to perform real-time detection, making them hard truly maintain cyberspace security. To solve these two challenges, this paper, combined with the sliding window detection algorithm Convolution Neural Network we propose a method which based on multi-channel convolution windows. In order analyze correlations between frames different neighborhood...