A5086685723- Robotics and Sensor-Based Localization
- Robotic Path Planning Algorithms
- Fault Detection and Control Systems
- Adaptive Control of Nonlinear Systems
- Fire Detection and Safety Systems
- Video Surveillance and Tracking Methods
- Machine Fault Diagnosis Techniques
- Target Tracking and Data Fusion in Sensor Networks
- Advanced Control Systems Optimization
- Fire effects on ecosystems
- Advanced Algorithms and Applications
- Distributed Control Multi-Agent Systems
- Advanced Image and Video Retrieval Techniques
- Anomaly Detection Techniques and Applications
- Industrial Vision Systems and Defect Detection
- Stability and Control of Uncertain Systems
- Blind Source Separation Techniques
- Image and Signal Denoising Methods
- Underwater Acoustics Research
- Human Pose and Action Recognition
- UAV Applications and Optimization
- Fractal and DNA sequence analysis
- Geophysics and Sensor Technology
- Indoor and Outdoor Localization Technologies
- Aerospace Engineering and Control Systems
Xi'an University of Technology
2016-2025
Xichang University
2008
Unmanned aerial vehicles (UAVs) are increasingly being used in forest fire monitoring and detection thanks to their high mobility ability cover areas at different altitudes locations with relatively lower cost. Traditional algorithms mostly based on the RGB color model, but speed accuracy need further improvements. This paper proposes a algorithm by exploiting YOLOv3 UAV-based images. Firstly, UAV platform for purpose of is developed. Then according available computation power onboard...
To promote the effect of variational mode decomposition (VMD) and further enhance recognition performances bearing fault signals, genetic algorithm (GA) is applied to optimize combination VMD parameters in this paper, GA-VMD put forward improve accuracy VMD. In addition, combined with center frequency, a feature extraction method based on frequency proposed ameliorate difficulty extraction. Firstly, signal decomposed into series intrinsic components (IMFs) by GA-VMD. Then, Center Frequency...
In order to accurately identify various types of ships and develop coastal defenses, a single feature extraction method based on slope entropy (SlEn) double SlEn combined with permutation (SlEn&PE) are proposed. Firstly, is used for the ship-radiated noise signal (SNS) compared (PE), dispersion (DE), fluctuation (FDE), reverse (RDE), so that effectiveness verified, has highest recognition rate calculated by k-Nearest Neighbor (KNN) algorithm. Secondly, PE, DE, FDE, RDE, respectively,...
Forest fires are very dangerous. Once they become disasters, it is difficult to extinguish. In this paper, an unmanned aerial vehicle (UAV) image-based forest fire detection approach proposed. Firstly, the local binary pattern (LBP) feature extraction and support vector machine (SVM) classifier used for smoke detection, so as make a preliminary discrimination of fire. order accurately identify in early stage fire, according convolutional neural network (CNN), has characteristics reducing...
With the new development in Unmanned Aerial Vehicle (UAV) recent years, UAV equipped with color and infrared cameras becomes a tool for carrying out forest fire detection fighting missions due to its advantages of low price, high maneuverability, easy use. In order detect potential early stage, UAV-based method using convolutional neural network is proposed this paper. The effectiveness algorithm verified by simulated flames an indoor experimental testbed.
The effectiveness and efficiency of using unmanned aerial vehicle (UAV) for automated power transmission line inspection is tightly related to the paths designed UAV. Different types UAV are suitable different tasks have requirements path planning. Based on contents inspection, it can be divided into tower monitoring corridor monitoring. First, according characteristics monitoring, multi-rotor used inspection. By considering safe distance bewteen features camera, genetic algorithm (GA)...
The analysis of ship radiation signals to identify ships is an important research content underwater acoustic signal processing. traditional fast Fourier transform (FFT) not suitable for analyzing non-stationary, non-Gaussian, and nonlinear In order realize the feature extraction accurate classification with higher accuracy, a method based on wavelet packet decomposition energy entropy proposed in this paper. According decomposition, decomposed into different frequency bands, its extracted....
This paper studies the distributed formation control problem for multiple unmanned aerial vehicles (UAVs), focusing on preserving connectivity and avoiding obstacles within constraints of a limited communication distance in presence dynamic obstacles. The UAV network is modeled as proximity graph, where edges are defined by distances between UAVs. A hierarchical strategy employed to manage position attitude subsystems independently. controller developed subsystems, utilizing bounded...
Traffic volume estimation is a fundamental task in Intelligent Transportation Systems (ITS). The highly unbalanced and asymmetric spatiotemporal distribution of traffic flow combined with the sparse uneven deployment sensors pose significant challenges for accurate estimation. To address these issues, this paper proposes novel framework. It combines dynamic adjacency matrix Graph Convolutional Network (GCN) multi-scale transformer structure to capture correlation. First, an adaptive...
As a kind of the forest "fault", fire is highly destructive and difficult to rescue. Fire segmentation helpful for firefighters understand scale formulate reasonable fire-fighting plan. Therefore, this paper proposes real-time method based on deep learning. This an improved version deeplbav3+, which encoder-decoder structure network. Encoder network composed convolutional neural atrous spatial pyramid pooling. Different from deeplabv3+, in order improve speed, uses lightweight mobilenetv3...
.This paper considers the learning-ability for discrete-time iterative learning control (ILC) systems with feedforward. More specifically, relation between output realizability and feedforward matrix is first established. Then, of four ILC considered. It shown that proportional type (P-type) update law can only ensure fully asymptotic learning-ability. By using matrix, a more efficient point-wise P-type developed, which \((T+2)\)-step learning-ability, where \(T\) trial length. In case state...
Permutation Lempel-Ziv complexity (PLZC) is a recently proposed method for analyzing signal complexity. However, PLZC only characterizes the from single scale and has certain limitations. In order to overcome these shortcomings improve performance of feature extraction underwater acoustic signal, this paper introduced coarse graining operation, multi-scale permutation (MPLZC), an automatic hybrid multi-feature ship-radiated noise (S-S) based on (MLZC), entropy (MPE) MPLZC. The results...