A5100439192- Remote-Sensing Image Classification
- Advanced Image and Video Retrieval Techniques
- Remote Sensing and Land Use
- Advanced Image Fusion Techniques
- Advanced Image Processing Techniques
- Advanced Vision and Imaging
- Recommender Systems and Techniques
- Infrared Target Detection Methodologies
- Image Processing Techniques and Applications
- Advanced Antenna and Metasurface Technologies
- Advanced Neural Network Applications
- Calibration and Measurement Techniques
- Image Retrieval and Classification Techniques
- Antenna Design and Analysis
- Atmospheric Ozone and Climate
- Metamaterials and Metasurfaces Applications
- Building Energy and Comfort Optimization
- Advanced Steganography and Watermarking Techniques
- Luminescence and Fluorescent Materials
- Image Enhancement Techniques
- Space Satellite Systems and Control
- Children's Physical and Motor Development
- Traffic Prediction and Management Techniques
- Electromagnetic wave absorption materials
- Face and Expression Recognition
Beijing Institute of Technology
2018-2024
Anhui Medical University
2024
Ministry of Education of the People's Republic of China
2024
University of Copenhagen
2024
Chengdu Sport University
2023-2024
University of Hong Kong
2022-2024
Ocean University of China
2023
Harbin Institute of Technology
2023
Yanshan University
2022
Guangzhou Research Institute of Non-ferrous Metals
2022
Accurate prediction of user behaviors is important for many social media applications, including marketing, personalization and recommendation, etc. A major challenge lies in that, the available behavior data or interactions between users items a given network are usually very limited sparse (e.g., >= 99.9% empty). Many previous works model from only historical logs. We observe that people members several networks same time, such as Facebook, Twitter Tencent's QQ. Importantly, their...
An aldehyde functionalized amino-borane has been found to respond multiple external stimuli such as temperature, pressure and solvents, producing distinct patterns colours.
This paper reviews the NTIRE 2021 challenge on learning super-Resolution space. It focuses participating methods and final results. The addresses problem of a model capable predicting space plausible super-resolution (SR) images, from single low-resolution image. must thus be sampling diverse outputs, rather than just generating SR goal is to spur research into developing formulations models better suited for highly ill-posed problem. And thereby advance state-of-the-art in broader field. In...
Target detection in hyperspectral images (HSIs) mainly relies on the spectral information of target prior. However, prior spectra with precise variation are often hard to obtain, and "different objects same spectrum" phenomena complex environments make it difficult detect variational targets. To mitigate this problem, self-supervised learning (SSL) is introduced mine HSIs supplement priors. The SSL-based detector, termed SSDCTD, optimized unlabeled using an unsupervised pretext task...
In this article, a novel method for developing dual-polarized broadband microwave absorber (MA) use at an extremely oblique incidence angle is presented. A detailed theoretical analysis performed and formulas are derived to obtain the required relative permittivity permeability tensors achieve absorption arbitrary angle. As proof of concept, MA operating <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math...
Object detection in remote sensing is a challenging task due to the arbitrary orientations of objects and vast variation number within single image. For instance, one image may contain hundreds small vehicles, while another only have football field. Recently, DEtection TRansformer (DETR) its variants achieved great success object by setting fixed queries using bipartite graph matching for one-to-one label assignment. However, we observed that can result relative redundancy when changes...
Abstract Multicolor emissive and responsive materials are highly attractive owing to their potential applications in various fields, polymers preferred for good processability high stability. Herein, we report a series of new based on methacrylate monomer containing switchable boron chromophore. In spite unconjugated nature, interestingly, the homopolymers from this display rare multicolor fluorescence solution that is dependent degree polymerization (DP). With an increasing DP, local...
Hyperspectral unmixing, which estimates end-members and their corresponding abundance fractions simultaneously, is an important task for hyperspectral applications. In this article, we propose a new autoencoder-based unmixing model with three novel components. First, sparse prior to maps. The proposed prior, called orthogonal (OSP), based on the observations that different maps are close because, generally, no more than two mixed within one pixel. As opposed conventional norm-based assumes...
Despite the promising performance on benchmark datasets that deep convolutional neural networks have exhibited in single image super-resolution (SISR), there are two underlying limitations to existing methods. First, current supervised learning-based SISR methods for remote sensing satellite imagery do not use paired real sensor data, instead operating simulated high-resolution (HR) and low-resolution (LR) image-pairs (typically HR images with their bicubic-degraded LR counterparts), which...
Abstract Background As life improves and sedentary time increases, young children's physical fitness gradually declines. Methods Multi-stage stratified whole cluster sampling was utilized to sample 5584 preschoolers. Young infants' morphology, function, quality were revealed using factor analysis. Results The analysis separated 3–6-year-olds into two genders: 1,551 men in group A "high fitness" 1,499 B "low fitness"; 1,213 women 1,321 B. measured by standing long jump(1.00), weight(1.00),...
Abstract Heterogeneity of consumer preferences is a crucial feature the platform economy, and service price charging strategies crucially influence customer choice behavior. To enhance experience boost product value, video platforms offer value‐added services, also known as premium membership features. For instance, Youku's benefits include seamless playback, high‐definition picture quality, exclusive episodes. This study constructs two‐stage Hotelling model from monopoly to duopoly...
Pseudo-invariant calibration sites (PICS) have been used for the radiometric and stability monitoring of satellite optical sensors. Several stable PICS, such as those in Sahara Desert North Africa, were selected vicarious earth remote sensing satellites. However, selection procedure PICSs whole Northwest China has not fully explored before. This paper presents a novel technique selecting PICS by combined using coefficient variation (CV) iteratively reweighted multivariate alteration...
Due to the discreteness of integer data, there are a large number gaps and continuous columns in histogram based on data. Aiming at characteristics, this paper presents robust reversible watermarking algorithm for relational database histogram. Firstly, it groups tuples according watermark length grouping key. Secondly, calculates prediction errors uses absolute values construct Thirdly, traverses find all turn, computes sum height each column selects group that has largest as positions...
Abstract Online product pricing is an important factor that affects the purchase of customers and earnings e‐commerce platform enterprises. It has increasingly prominent influence on consumers’ purchasing decisions. To study customers’ anchoring psychology behaviors, this paper proposes a model considers online consumers to be anchored by price sales. The cognitive bias described utility function based anchoring‐adjustment heuristic, consumer choice behavior strategy are studied with...
Remote sensing image scene classification (RSISC) plays a vital role in remote applications. Recent methods based on convolutional neural networks (CNNs) have driven the development of RSISC. However, these approaches are not adequate considering contributions different features to global decision. In this paper, triplet-metric-guided multi-scale attention (TMGMA) is proposed enhance task-related salient and suppress task-unrelated redundant features. Firstly, we design module (MAM) guided...
Conformal array antennas (CAAs) can effectively broaden the beam scanning range and improve radiation flexibility. Unfortunately, confirmation will increase difficulty of analysis synthesis, stealth CAAs is arduous. In this article, we propose an innovative optimization algorithm for stealthy based on twin delayed deep deterministic policy gradient (TD3) algorithm, which combines advantages learning reinforcement learning. Furthermore, characterize CAA as electromagnetic model (EMM) embed it...
In relational databases, embedding watermarks in integer data using traditional histogram shifting method has the problem of large distortion. To solve this problem, a reversible database watermarking without redundant distortion is proposed, taking advantage number gaps histogram. This embeds watermark bit by on basis grouping. First, an constructed with absolute value prediction error as variable. Second, positional relationship between each column and gap analyzed to find out all columns...
A hybrid illumination method using both daylight and light-emitting diodes (LEDs) for indoor lighting is presented in this study. The can be introduced into the space by a panel-integration system. part LEDs are combined within specific luminaire that provide uniform illumination. turned on dimmed through closed-loop control when illuminance inadequate. We simulated calculated efficiency of our LED system, compared with fluorescent systems. Simulation results show daylight/LED better than...