A5100755966- Biometric Identification and Security
- Face and Expression Recognition
- Advanced SAR Imaging Techniques
- Geophysical Methods and Applications
- Image Processing Techniques and Applications
- Underwater Acoustics Research
- Image Retrieval and Classification Techniques
- Robotic Path Planning Algorithms
- Face recognition and analysis
- Industrial Vision Systems and Defect Detection
- Robotics and Sensor-Based Localization
- Tunneling and Rock Mechanics
- Domain Adaptation and Few-Shot Learning
- Metaheuristic Optimization Algorithms Research
- Military Defense Systems Analysis
- Advanced Neural Network Applications
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Antenna Design and Analysis
- Flow Experience in Various Fields
- Grouting, Rheology, and Soil Mechanics
- Non-Destructive Testing Techniques
- Fatigue and fracture mechanics
- Electromagnetic Scattering and Analysis
- Anomaly Detection Techniques and Applications
- UAV Applications and Optimization
Anhui University of Science and Technology
2025
Wuyi University
2009-2024
Yanshan University
2024
Wuyi University
2010-2024
Xiamen University
2023
Ocean University of China
2021
Xidian University
2021
State Key Laboratory of Remote Sensing Science
2014
South China University of Technology
2012-2013
High resolution range profile (HRRP) has attracted increasing attention in radar automatic target recognition (RATR). However, the target-aspect missing problem non-cooperative targets recognition, which is one of most challenging tasks RATR, received very few contributions recently. The proposed work motivated by a simple observation, i.e., as compared with HRRP signals interesting targets, sufficient unlabeled are much easier to acquire. these often neglected since there no label...
Though Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR) via Convolutional Neural Networks (CNNs) has made huge progress toward deep learning, some key issues still remain unsolved due to the lack of sufficient samples and robust model. In this paper, we proposed an efficient transferred Max-Slice CNN (MS-CNN) with L2-Regularization for SAR ATR, which could enrich features recognize targets superior performance. Firstly, data amplification method is presented reduce...
Semantic segmentation of the sensed point cloud data plays a significant role in scene understanding and reconstruction, robot navigation, etc. This work presents Graph Convolutional Network integrating K-Nearest Neighbor searching (KNN) Vector Locally Aggregated Descriptors (VLAD). KNN is utilized to construct topological graph each its neighbors. Then, we perform convolution on edges constructed extract representative local features by multiple Multilayer Perceptions (MLPs). Afterwards,...
Abstract As UAVs are more and widely used in military civilian fields, their intelligent applications have also been developed rapidly. However, high‐precision autonomous landing is still an industry challenge. GPS‐based methods will not work places where GPS signals available; multi‐sensor combination navigation difficult to be because of the high equipment requirements; traditional vision‐based sensitive scale transformation, background complexity occlusion, which affect detection...
Iris recognition has been a hot research topic in these years. In this paper, an iris method based on the Contourlet Transform (CT) and Biomimetic Pattern Recognition (BPR) proposed. proposed method, was used to extract significant features of preprocessed image, algorithm construct high-dimensional covering hypersurface order recognize images. Experiments CASIA image database show that CT BPR can achieve promising results compared with other methods.
Fusion biometric recognition modal contributes in two aspects. It can not only improve the accuracy, but also gives a comparatively safe strategy, since it is difficult for intruders to achieve multi-biometric information simultaneously, especially iris information. In this paper, novel fusion with and facial images based on biomimetic pattern proposed. The Contourlet transform (CT) directional dimensional principal component analysis (2D) <sup xmlns:mml="http://www.w3.org/1998/Math/MathML"...
Introduction Flow, defined as a heightened state of consciousness characterized by intense concentration during an activity, is influenced primarily the perceived challenge and dynamic equilibrium skills. This investigation focuses on patterns flow attainment its elicitation mechanisms within context piano performance among Chinese music college students. Methods Our study establishes framework for accessing flow, utilizing quantitative data from ontology to gauge level acquisition assess...
In order to enhance the stability and security of biometric features recognition, finger-knuckle-print (FKP) is used in this paper study high performance recognition problem based on image set. After extracting feature by method local phase quantization, an set can transform a closely related points affine space. Then models convex hulls are constructed these point sets. Finally, FKP was processed optimized model. Experiments publish database show that proposed algorithm achieves reliable...
Due to the needs of modern battlefield awareness and military reconnaissance, Synthetic Aperture Radar (SAR) automatic target recognition (ATR) has been receiving more attention. In this paper, a SAR ATR algorithm based on Local Phase Quantization (LPQ) plus Biomimetic Pattern Recognition (BPR) proposed. There are three main steps in proposed algorithm: firstly, simple preprocessing procedure centroid location is applied original image extract target's region interest. Secondly, short-term...
This paper presents a new method for effective iris recognition using Biomimetic Pattern Recognition (BPR), which is theory proposed by academician ShoujueWang. A model that based on BPR was introduced and thoroughly discussed here. Experimental results the Chinese Academy of Sciences, Institute Automation (CASIA) image database clearly demonstrates use makes it possible to achieve highly accurate efficient with model.
Synthetic aperture radar (SAR) automatic target recognition (ATR) has been receiving more and attention in the past two decades. A lot of methods have proposed studied for recognition. Among some these methods, they use supervised algorithms to extracts features. In this paper, we first a unsupervised algorithm, K-means clustering, which can learn features without known class training samples, As algorithm high demand on scale data, so method data augmentation get by clustering Experimental...
As a novel biometric, Finger-Knuckle-Print (FKP) has received great interest in recent years, and become hot research spot of biometric recognition. Due to its characteristic uniqueness, easy accessibility, none abrasion abundant texture, it been widely applied personal identification. But the spare representation based FKP method not reported yet. In this paper, smooth l <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> norm model...
Aiming at the problem of scarce defect data set in industrial product detection, a recognition method based on fusion learning multi-layer image feature is proposed this paper. The introduces three Gaussian pyramids with same size and different resolution to obtain more information, establishes Defect Recognition Convolutional Network extract information. experimental results show that our model capable extracting features, can superior accuracy, recall, precision, as well deal fuzzy sampling.
Midcourse recognition of ballistic missile is one the core problems in anti-missile early warning operations. Considering variety penetration means missile, aiming at attitude characteristics midcourse flight, this paper focuses on analysis five technologies suitable for warhead, such as RCS sequence, establishes a Bayesian network model and standardizes conditional probability representation results techniques according to actual usage. Finally, effectiveness verified by simulation...
Aimed at the high incident angle request in frequency selected surfaces (FSS) actual application problem, FSS design method irradiation state is proposed this paper. By studying state, phenomenon that band-pass feature deteriorates along with incidence angle's increase ameliorated, which offers evidence for engineering apply.
A novel finger-knuckle-print (FKP) recognition based on convex optimization is proposed in this paper. Convex hulls and image sets are adopted to construct the method. Experiments publish FKP database show that algorithm can achieve high performance compared with state-of-the-art