Dissertation > Excellent graduate degree dissertation topics show
Multi-sensor information fusion based Target Recognition Algorithm
Author: ZhuNing
Tutor: WangQingChao
School: Harbin Institute of Technology
Course: General and Fundamental Mechanics
Keywords: Multi-sensor information fusion Air target identification Neural Network Evidence Theory Multi-class classification Support Vector Machine
CLC: TP202
Type: Master's thesis
Year: 2008
Downloads: 136
Quote: 2
Read: Download Dissertation
Abstract
|
Modern warfare, air defense battle command target recognition is fast, accurate, one of the key decisions . Aerial target recognition directly affect the success of the deployment of anti-aircraft fire , distribution, and effective strikes, then that would affect the entire battlefield . Therefore, the target recognition technology for improving the performance of the entire air defense system has important significance. Based on multi-sensor information fusion technique to study the air target identification method and its application. First introduced the theory of multi-sensor information fusion , analysis of the existing evidence combination method , and that its shortcomings , then the evidence for different focal element in the conflict between the degree of research , based on the mutual confidence based conflicting evidence combination method for aerial target identification. Simulation results show that : The synthesis of the new approach is good, even better synthesis result . For multi-sensor information fusion basic probability assignment problems are difficult to obtain , given a neural network technology to obtain basic probability assignment methods, and on this basis, the evidence presents a combination of theory and neural networks for target recognition fusion method . Focuses on the support vector machine (SVM) in a multi- sensor aerial target recognition . SVM is a statistical learning theory developed on the basis of a new generation learning algorithm based on structural risk minimization (SRM criterion ) , using the empirical risk and the confidence range two while minimizing the risk as functional . Several multi- class classification by support vector machines , will DAGSVM applied target identification. Simulation study of the neural network, neural network and evidence theory combining recognition algorithm , DAGSVM performance of the three algorithms , the results show that SVM DAG noise in any case , have the highest recognition rate, especially in large background noise under identify the more obvious advantages , indicating support vector machine in the air, the effectiveness of target recognition . On the other hand , in DAGSVM adding parallel algorithms, simulation results show that the algorithm speed has been effectively improved.
|
Related Dissertations
- Development of the Platform for Compressor Optimization Design and Aerodynamic Optimization Design in the Transonic Compressor,TH45
- Research on Automatic Detection Algorithm for Substructure Distress of Highway Pavement Based on SVM,U418.6
- Research on Autamatic Music Structrue Analysis,TN912.3
- Research on Joint Target Detection for Dual-Sensor Image and System Implementation,TP391.41
- Research on Transductive Support Vector Machine and Its Application in Image Retrieval,TP391.41
- Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
- Study on Virtual Detector of Infrared Hyper-Spectral Image,TP391.41
- The Application of Fuzzy Control and Neural Network in Planar Double Inverted Pendulum,TP273.4
- Research of Visual Servoing in 4-DOF Robotic Manipulator,TP242.6
- Research on Visual Servo System of Mechanical ARM,TP242.6
- Multi-Sensor Information Fusion and Its Applications on Wearable Computer,TP202
- Fault Diagnosis Method Based on Support Vector Machine,TP18
- Design of Weapon Detection Device Control System,TP183
- Process Support Vector Machine and Its Application to Satellite Thermal Equilibrium Temperature Prediction,TP183
- Municipal tourism land use planning environmental impact assessment,X820.3
- Research on the Inteligent System of High Performance Concrete Mix Design in Zhujiang Triangle Area,TU528
- Study on Taste Characteristic of Taste Peptide Enzymatic Production from Oyster Base on A Neural Network Method,TS254.4
- Smart Control Research in Paddy Drying Based on BP Neural Network,S226.6
- BP network optimization based on genetic algorithm optimization of the biodiesel process,TE667
- Fire Fighting System Research for the Offshore Platform,U698.4
- The Research on Evaluation of Living Status Systems of Expressway Relocated People,D523
CLC: > Industrial Technology > Automation technology,computer technology > Automation technology and equipment > General issues > Design, performance analysis and synthesis
© 2012 www.DissertationTopic.Net Mobile
|