A5100339552- Quantum Computing Algorithms and Architecture
- Advanced Neural Network Applications
- Image Retrieval and Classification Techniques
- Advanced Graph Neural Networks
- Quantum Information and Cryptography
- Neural Networks and Applications
- Advanced Vision and Imaging
- Human Pose and Action Recognition
- Medical Image Segmentation Techniques
- Quantum-Dot Cellular Automata
- Rough Sets and Fuzzy Logic
- Advanced Image Processing Techniques
- Video Surveillance and Tracking Methods
- Adversarial Robustness in Machine Learning
- Generative Adversarial Networks and Image Synthesis
- Higher Education and Teaching Methods
- Face and Expression Recognition
- Fuzzy Logic and Control Systems
- Video Analysis and Summarization
- Digital Media Forensic Detection
- AI in cancer detection
- Image Processing Techniques and Applications
- Biometric Identification and Security
- Anomaly Detection Techniques and Applications
- Time Series Analysis and Forecasting
Fujian Agriculture and Forestry University
2021-2025
Fudan University
2008-2025
Sichuan University
2025
Yale University
2025
Hubei University of Technology
2023-2024
Shanghai Fudan Microelectronics (China)
2019-2023
Novatek Microelectronics (Taiwan)
2023
China Electronics Technology Group Corporation
2022
Anhui University
2022
Southern University of Science and Technology
2022
In this correspondence, we have presented a rotation and gray scale transform invariant texture recognition scheme using the combination of quadrature mirror filter (QMF) bank hidden Markov model (HMM). first stage, QMF is used as wavelet to decompose image into subbands. The features derived from statistics based on first-order distribution levels are then extracted each subband image. second sequence subbands modeled (HMM), one HMM designed for class textures. exploit dependence among...
Robotic soccer games, which have become popular, require timely and precise decisionmaking in a dynamic environment.To address the problems of complexity critical situation, policy improvement robotic games must occur.This paper proposes an adaptive method that uses reinforcement learning (RL), decision-making system for game is composed two subsystems.The first subsystem architecture proposed criticizes second implements policy.Inspired by support vector machine (SVM), situation...
To tackle the problem on trajectory planning or design of control law, this paper introduces a visual servoing system for manipulator with redundant joints that approaching target is determined spontaneously by law. The proposed method resolves joint solution and obstacle avoidance. work comprises two procedures, feature extraction position-based (PBVS) collision avoidance within working envelope. In PBVS control, pose must be reconstructed respect to robot results in Cartesian...
Abstract Background Both contextual and cancer-related factors could be identified as causes of the interindividual variability observed for symptoms experienced during breast cancer treatment with chemotherapy. Understanding age differences predictors latent class memberships symptom heterogeneity contribute to personalized interventions. This study aimed identify role on in women undergoing chemotherapy China. Methods A cross‑sectional survey was conducted among patients three tertiary...
This brief presents a reconfigurable VLSI architecture which is designed for multi-transform codec in several video coding standards of MPEG-2/4, VC-1, H.264/AVC and AVS. The multiple constant multiplication algorithm with two fusing strategies provided to generate multipliers the matrix calculation blocks. Additionally, adder-sharing strategy adopted unified preprocessing/postprocessing block save circuit areas. proposed can support different through static reconfiguration forward/inverse...
An adversary can fool deep neural network object detectors by generating adversarial noises. Most of the existing works focus on learning local visible noises in an "patch" fashion. However, 2D patch attached to a 3D tends suffer from inevitable reduction attack performance as viewpoint changes. To remedy this issue, work proposes Coated Adversarial Camouflage (CAC) arbitrary viewpoints. Unlike trained space, our camouflage generated conceptually different training framework consists...
With the continuous development of landscape restoration technology, how to use modern technology efficiently reconstruct degraded and damaged historical gardens help them restore protect has become an important topic. Traditional 3D reconstruction methods often face challenges in accuracy efficiency when facing complex garden geometry ecological environment. To this end, paper proposes a hybrid model DGA-Net that combines deep convolutional network (DCN), graph (GCN) attention mechanism...
This study aimed to identify latent classes of online health information seeking (OHIS) behaviors among young women diagnosed with breast cancer in China and examine associated personal characteristics support tailored education strategies. Young were recruited from a center between April September 2024. Participants completed questionnaires on demographic clinical characteristics, OHIS behaviors, psychosocial cognitive factors, trust, social norms, communication, experience. Latent class...
Time series forecasting has attracted significant attention in recent decades. Previous studies have demonstrated that the Channel-Independent (CI) strategy improves performance by treating different channels individually, while it leads to poor generalization on unseen instances and ignores potentially necessary interactions between channels. Conversely, Channel-Dependent (CD) mixes all with even irrelevant indiscriminate information, which, however, results oversmoothing issues limits...
Fatigue is the most frequent and distressing symptom affecting physical, cognitive, affective domains of breast cancer patients. The Functional Assessment Chronic Illness Therapy-Fatigue (FACIT-F) has been widely used in patients with chronic diseases shown satisfactory reliability validity. This study aimed to examine psychometric properties FACIT-F among Chinese cancer.Using a convenience sampling method, cross-sectional survey (January 2020 September 2022) was recruited from two tertiary...
Fingerprint recognition is widely used in daily life. However, the physiological phenomena of human can still fail under dry and low temperature conditions.We present two main approaches for enhancing accuracy fingerprint systems cases fingerprints. The first approach an image enhancement algorithm with multiple training stages (MTS) fingerprints at temperatures, which aims to restore quality normal quality. second method a Cycled Contrastive Ridge Generative Adversarial Network (CCRGAN)...
Graph convolutions have been a pivotal element in learning graph representations. However, recursively aggregating neighboring information with leads to indistinguishable node features deep layers, which is known as the over-smoothing issue. The performance of neural networks decays fast number stacked layers increases, and Dirichlet energy associated decreases zero well. In this work, we introduce framelet system into analysis take multi-scale perspective leverage alleviate Specifically,...
This paper presents a novel learning algorithm of fuzzy perceptron neural networks (FPNNs) for classifiers that utilize expert knowledge represented by IF-THEN rules as well numerical data inputs. The conventional linear network is extended to second-order one, which much more flexible defining discriminant function. In order handle numbers in networks, level sets input vectors are incorporated into learning. At different levels the numbers, updating weight vector depends on minimum output...