A5004436217- AI in cancer detection
- Retinal Imaging and Analysis
- Face and Expression Recognition
- Face recognition and analysis
- Advanced Neural Network Applications
- Advanced Image and Video Retrieval Techniques
- Domain Adaptation and Few-Shot Learning
- Cervical Cancer and HPV Research
- Digital Imaging for Blood Diseases
- Radiomics and Machine Learning in Medical Imaging
- Image Retrieval and Classification Techniques
- Video Surveillance and Tracking Methods
- Human Pose and Action Recognition
- Brain Tumor Detection and Classification
- Image Processing Techniques and Applications
- Multimodal Machine Learning Applications
- Retinal and Optic Conditions
- Robotic Path Planning Algorithms
- Medical Imaging and Analysis
- COVID-19 diagnosis using AI
- Advanced Image Fusion Techniques
- Medical Image Segmentation Techniques
- Biometric Identification and Security
- Glaucoma and retinal disorders
- Robotics and Sensor-Based Localization
Central South University
2016-2025
China Mobile (China)
2016-2017
Northwest University of Politics and Law
2008
Chinese Academy of Sciences
2007
Chongqing University
2005
Ministry of Education of the People's Republic of China
2004
Autonomous exploration in unknown environments is a fundamental task of Unmanned Aerial Vehicles (UAVs). To choose goals wisely, we propose an information-driven strategy by applying the fast marching method to UAVs. A frontier point detection algorithm designed obtain Candidate Goals (CGs) utilizing structural characteristics octree-based map. With sum information gain during journey as evaluation indicator, present novel utility function evaluate CGs considering trade-off between and...
Cervical abnormal cell detection is a challenging task as the morphological discrepancies between and normal cells are usually subtle. To determine whether cervical or abnormal, cytopathologists always take surrounding references to identify its abnormality. mimic these behaviors, we propose explore contextual relationships boost performance of detection. Specifically, both cell-to-global images exploited enhance features each region interest (RoI) proposal. Accordingly, two modules, dubbed...
Medical image segmentation is indispensable for diagnosis and prognosis of many diseases. To improve the performance, this study proposes a new 2D body edge aware network with multi-scale short-term concatenation medical segmentation. Multi-scale modules which concatenate successive convolution layers different receptive fields, are proposed capturing representations fewer parameters. Body generation feature adjustment based on weight map computing via enlarging convolutions using Sobel...
Automated segmentation of hard exudates in colour fundus images is a challenge task due to issues extreme class imbalance and enormous size variation. This paper aims tackle these proposes dual-branch network with dual-sampling modulated Dice loss. It consists two branches: large exudate biased branch small branch. Both them are responsible for their own duties separately. Furthermore, we propose loss the training such that our proposed able segment different sizes. In detail, first branch,...
The autonomous exploration task we consider requires Unmanned Aerial Vehicles (UAVs) to actively navigate through unknown environments with the goal of fully perceiving and mapping environments. Some existing strategies suffer from rough cost budgets, ambiguous Information Gain (IG), unnecessary backtracking caused by Fragmented Regions (FRs). In our work, a hierarchical spatio-temporal-aware framework is proposed alleviate these problems. At local level, Asymmetrical Traveling Salesman...
In this paper, we propose a novel age estimation method based on gradient location and orientation histogram (GLOH) descriptor multi-task learning (MTL). The GLOH, one of the state-of-the-art local descriptor, is used to capture age- related spatial information face image. As extracted GLOH features are often redundant, MTL designed select most informative bins for problem, while corresponding weights determined by ridge regression. This approach largely reduces dimensions feature, which can...
Diabetic retinopathy (DR) is one of the leading cause blindness, but classification DR requires experienced ophthalmologist to distinguish presence various small features, which time-consuming and difficult. Convolution neural network (CNN), enables learning hierarchical discriminative features without experiences clinicians, an alternative method address above issue. In this paper, we investigate four factors employing deep CNN problem, including architectures, preprocessing, class...