A5101997384- Aluminum Alloy Microstructure Properties
- QR Code Applications and Technologies
- Advanced Steganography and Watermarking Techniques
- Image and Video Quality Assessment
- Advanced Image and Video Retrieval Techniques
- Intermetallics and Advanced Alloy Properties
- Advanced Image Processing Techniques
- Solidification and crystal growth phenomena
- Metallurgy and Material Forming
- Digital Media Forensic Detection
- Image and Signal Denoising Methods
- Aluminum Alloys Composites Properties
- Visual Attention and Saliency Detection
- Image Processing Techniques and Applications
- Topic Modeling
- Advanced Vision and Imaging
- Image Retrieval and Classification Techniques
- Advanced Text Analysis Techniques
- Image Enhancement Techniques
- Semiconductor materials and interfaces
- Advanced Image Fusion Techniques
- Metallurgical Processes and Thermodynamics
- Natural Language Processing Techniques
- Geophysical Methods and Applications
- Advanced Materials Characterization Techniques
Shanghai Jiao Tong University
2012-2025
North China Electric Power University
2025
Mianyang Normal University
2024
Shandong Normal University
2024
Aviation Industry Corporation of China (China)
2023
Tianjin University
2023
University of Science and Technology Beijing
2023
Tianjin Hospital
2023
Southeast University
2020-2022
Beijing University of Posts and Telecommunications
2020
In the era of multimedia and Internet, quick response (QR) code helps people obtain information from offline to online quickly. However, QR is often limited in many scenarios because its random dull appearance. Therefore, this article proposes a novel approach embed hyperlinks into common images, making invisible for human eyes but detectable mobile devices equipped with camera. Our an end-to-end neural network encoder hide messages decoder extract messages. To maintain hidden message...
Video shakiness is an unpleasant distortion of User Generated Content (UGC) videos, which usually caused by the unstable hold cameras. In recent years, many video stabilization algorithms have been proposed, yet no specific and accurate metric enables comprehensively evaluating stability videos. Indeed, most existing quality assessment models evaluate as a whole without specifically taking subjective experience into consideration. Therefore, these cannot measure explicitly precisely when...
Computer graphics images (CGIs) are artificially generated by means of computer programs and widely perceived under various scenarios, such as games, streaming media, etc. In practice, the quality CGIs consistently suffers from poor rendering during production, inevitable compression artifacts transmission multimedia applications, low aesthetic resulting composition design. However, few works have been dedicated to dealing with challenge image assessment (CGIQA). Most (IQA) metrics developed...
In this paper, we present a simple but effective method to enhance blind video quality assessment (BVQA) models for social media videos. Motivated by previous researches that leverage pre-trained features extracted from various computer vision as the feature representation BVQA, further explore rich quality-aware image (BIQA) and BVQA auxiliary help model handle complex distortions diverse content of Specifically, use SimpleVQA, consists trainable Swin Transformer-B fixed SlowFast, our base...
UHD images, typically with resolutions equal to or higher than 4K, pose a significant challenge for efficient image quality assessment (IQA) algorithms, as adopting full-resolution images inputs leads overwhelming computational complexity and commonly used pre-processing methods like resizing cropping may cause substantial loss of detail. To address this problem, we design multi-branch deep neural network (DNN) assess the from three perspectives: global aesthetic characteristics, local...
Patients suffering from facial paralysis are on the hazard of disfigurement and loss vision due to blink function. Functional-electrical stimulation (FES) is one possible way restoring other functions in these patients. A restoration system for uni-lateral paralyzed patients described this paper. The achieves synchronized through processing myoelectric signal orbicularis oculi at normal side real-time as trigger stimulate eyelid. Design issues discussed, including EMG processing, stimulating...
Banding, also known as staircase-like contours, frequently occurs in flat areas of images/videos processed by compression or quantization algorithms. As undesirable artifacts, banding destroys the original image structure, thus inevitably degrading users' quality experience (QoE). In this paper, we systematically investigate assessment (IQA) problem, aiming to detect artifacts and evaluate their perceptual visual quality. Considering that existing databases only contain limited content...
Abstract Conventional superhydrophobic coatings frequently incorporate thermosetting resins as substrates, which are challenging to recycle. This study explores the application of ester‐exchange degradable 4,5‐Epoxyhexane‐1,2‐dicarboxylic glycidyl ester (DGEAC) combined with structurally engineered and perfluorooctyltriethoxysilane (POTS) modified carbon black@silica (CB@SiO 2 ‐POTS) particles fabricate recyclable (DGEAC/CB@SiO using a non‐solvent‐induced phase separation technique. The...
The overall scale of China’s logistics industry is growing rapidly, and the environment conditions for its development are constantly improving, which lays a solid foundation further accelerating industry. However, services characterized by subordination, immediacy, demand volatility, substitutability. Low-level integrated management seriously hinders service supply chain (LSSC) sustainable performance. Many studies have been limited to performance evaluation LSSCs, factors affecting LSSC...
QR (quick response) codes are widely used as an offline-to-online channel to convey information (e.g., links) from publicity materials display and print) mobile devices. However, not favorable for taking up valuable space of materials. Recent works propose invisible codes/hyperlinks that can hidden offline online. they require markers locate codes, which fails the purpose be visible because markers. This paper proposes a novel hiding architecture display/print-camera scenarios, consisting...
Quick response (QR) codes are widely used in offline to online channels transfer information from promotional materials mobile devices. Self-service medical equipment can record the data of each test, so use QR realize exchange between patients and doctors, institutions, self-service equipment, create a platform for health files. However, since anyone easily read code, it is not conducive protection patient privacy. Therefore, we propose code encryption decryption model based on robust...
experiments conducted on two AI-generated content datasets and traditional IQA show that MA-AGIQA achieves stateof-the-art performance, demonstrate its superior generalization capabilities assessing the quality of AGIs.
This paper presents a novel approach for accurate barcodes detection in real and challenging environments using compact deep neural networks. Our is based on Convolutional Neural Network (CNN) network compression, which can detect the four vertexes coordinates of barcode accurately quickly. consists stages: (i) feature extraction by base network, (ii) region proposal (RPN) training, (iii) classification regression, (iv) weights pruning recoding. The model trained first three stages then...
Analysis of functional connectivity networks (FCNs) derived from resting-state magnetic resonance imaging (rs-fMRI) has greatly advanced our understanding brain diseases, including Alzheimer's disease (AD) and attention deficit hyperactivity disorder (ADHD). Advanced machine learning techniques, such as convolutional neural (CNNs), have been used to learn high-level feature representations FCNs for automated classification. Even though convolution operations in CNNs are good at extracting...