A5100687501- Face and Expression Recognition
- Domain Adaptation and Few-Shot Learning
- Advanced Image and Video Retrieval Techniques
- Migration and Labor Dynamics
- Advanced Vision and Imaging
- 3D Surveying and Cultural Heritage
- Traditional Chinese Medicine Studies
- China's Socioeconomic Reforms and Governance
- Video Surveillance and Tracking Methods
- Image Retrieval and Classification Techniques
- Handwritten Text Recognition Techniques
- Remote Sensing and Land Use
- Advanced Neural Network Applications
- Robotics and Sensor-Based Localization
- Image Processing and 3D Reconstruction
- Multimodal Machine Learning Applications
- Face recognition and analysis
- Regional Development and Environment
- Image and Object Detection Techniques
- Sexual Assault and Victimization Studies
- Social Policy and Reform Studies
- Sentiment Analysis and Opinion Mining
- Satellite Image Processing and Photogrammetry
- Text and Document Classification Technologies
- Emotion and Mood Recognition
University of Chinese Academy of Sciences
2024
Xiangnan University
2017-2024
Shenzhen Institutes of Advanced Technology
2024
Chinese Academy of Sciences
2024
Hitotsubashi University
2022-2024
Hefei Institute of Technology Innovation
2024
Chongqing Medical University
2024
Children's Hospital of Chongqing Medical University
2024
Hunan University
2023
Southwestern University of Finance and Economics
2023
Abstract This study involved a meta‐analysis of 506 estimates extracted from 75 studies to estimate the effect size rural household registration ( hukou ) on wage levels. Our meta‐synthesis results indicated that negative wages is statistically significant; however, remains small in terms partial correlation coefficient. The meta‐regression analysis and test for publication selection bias differences among genders, corporate ownership sectors, periods are insignificant. We also found...
Invasive fungal infections (IFIs) pose a significant threat to immunocompromised individuals, leading considerable morbidity and mortality. Prompt accurate diagnosis is essential for effective treatment. Here we develop rapid molecular diagnostic method that involves three steps: enrichment using affinity-magnetic separation (AMS), genomic DNA extraction with silicon hydroxyl magnetic beads, detection through one-pot system. This method, optimized detect 30 CFU/mL of C. albicans in blood...
Background Violence against health professionals is a global public problem. In 2019, doctor was killed in Civil Aviation General Hospital (CAGH), which triggered national discussion about hospital violence. Sina Weibo, the Chinese version of Twitter, played an important role this discussion. The CAGH incident provides us with opportunity to explore how social media used on violence doctors. Methods Using built-in search engine data set containing 542 micro-blogs established. Three keywords:...
In current clinical practice, the localization of human acupuncture points relies extensively on subjective experience physicians. Therefore, despite being a crucial basic content traditional Chinese medicine (TCM), point has not been well expanded and promoted through intelligent means. Our goal is to explore an efficient reliable solution for recognition that addresses shortcomings subjectivity standardization in this task. We focus weak feature body surface back propose innovative...
Abstract As a common biometric recognition technology, face is also an important research direction in the field of computer. Although compared with initial research, current has made great progress, but there are still many difficulties practical application. In this paper, by extracting HOG features, after introducing detailed steps PCA and LDA subspace feature extraction methods, dimensionality reduction method combing applied to extract features. This first uses reduce dimension then for...
Abstract At present, facial expression recognition technology is widely used in artificial intelligence, transportation, medical and other aspects, so it has important research value. Traditional uses manual feature extraction method with low accuracy weak generalization ability, which difficult to be applied real life. With the development of deep learning, convolution neural network appears people’s vision. Different from traditional extraction, can learn image features independently, more...
Reconstruction of 3D structures from multiple 2D images has wide applications in such fields as computer vision, cultural heritage preservation, etc. This paper presents a novel multi-view stereo algorithm based on homogeneous direct spatial expansion (MVS-HDSE) with high reconstruction accuracy and completeness. It adopts many unique measures each step reconstruction, including initial seed point extraction using the DAISY descriptor to increase number sparse points, enhance efficiency,...
Discriminative subspace clustering (DSC) can make full use of linear discriminant analysis (LDA) to reduce the dimension data and achieve effective high-dimension by low-dimension in subspace. However, most existing DSC algorithms do not consider noise outliers that may be contained sets, when they are applied sets with or outliers, often obtain poor performance due influence outliers. In this paper, we address problem sensitivity outlier. Replacing Euclidean distance objective function LDA...
Favorite-longshot and reverse favorite-longshot biases have become widespread in various traditional sports betting markets recent years. However, there is a limited number of investigations that been conducted on the eSports market or bettors operate within it. In present research, we made efforts to re-examine bias inefficiency four typical eSport games: League Legends, Counter-Strike: Global Offensive, Dota 2, King Glory. Due natural characteristics e-sports, analyze reasons for from 4...
Correspondences matching is essential to image stitching and greatly influences the quality as homography matrix calculated from correspondences. Mismatching would be generated by using SIFT alone when dealing images with repeat similar structures. To improve accuracy, DAISY are combined extract match feature points. Then computed least square, RANSAC bundle adjust methods. Experiments show that accuracy improved results well robust.
Despite strong policy support for volunteerism, Chinese social service organizations require assistance in volunteer management, particularly concerning retaining volunteers and sustaining the supply of services. By interviewing from a successful organization analyzing collected data using constructive grounded theories methods, this study found that, contrary to conclusions previous studies, (1) sustainable volunteerism involves volunteers’ ability, motivation, resources; (2) ideal...
The arrival of big data era has affected the logic and trend social development. Big broadens coverage education, highlights subjectivity students, requires teaching content. enriches methods ideological political improves pertinence quality educators, promotes education.Big data’s core technology been generally accepted by colleges universities, application represented Hadoop system become a common tool for mining analysis education in universities. In data, we should pay attention to...
Text emotion analysis transforms a text sequence of indefinite length into category which is one the key research problems in field natural language processing. With wide application deep learning technology processing model based on has made new breakthrough. This paper builds basic framework and describes it from two aspects data preprocessing design network structure revolutionary neural network. Data mainly involves word segmentation embedding. Design (CNN) including input layer...
The electric field detector of the CSES satellite has captured a vast number lightning whistler events. To recognize them effectively from massive amount data, recognition algorithm based on speech technology attracted attention. However, this approach failed to events which are contaminated by other low-frequency electromagnetic disturbances. overcome limitation, we apply single-channel blind source separation method and audio develop novel model, consists two stages. (1) training stage:...
Abstract Taking MNIST data-set as research object, the HMM is introduced into Handwritten Number Recognition for first time. After implementation of classical training algorithm, some optimization methods are proposed problems existing in training. The general random initialization parameters lead to long time and unstable data. based on observations can speed up avoid data overflow. number iterations process not positively related output probability. In order obtain an optimal model, after...