Dissertation > Excellent graduate degree dissertation topics show
Research on Camera Calibration Method with Neural Network
Author: TianZhen
Tutor: Zhang
School: National University of Defense Science and Technology
Course: Instrument Science and Technology
Keywords: Machine Vision Camera Calibration Parallel calibration Neural Networks
CLC: TP391.41
Type: Master's thesis
Year: 2010
Downloads: 118
Quote: 2
Read: Download Dissertation
Abstract
|
Various nonlinear factors present in the imaging process , such as radial distortion and cut to the distortion , the ultimate camera imaging model for non-linear model . In order to better achieve the object point like the nonlinear relationship between the points , the neural network technology is the introduction of the camera calibration work . This article focuses on the camera calibration method based on neural network technology, in-depth research . Current use of neural network research for camera calibration , can only achieve a single location or a small range of calibration accuracy requirements of the space . When the calibration space has expanded to a certain range , calibration accuracy and calibration of the speed of the conflict will not be adjusted . This paper describes the respective characteristics of these two disciplines camera calibration and neural networks . On this basis , this paper re- positioning the point of integration between the two , presents a new calibration method , the neural network to focus on the calculation of non-linear factors . Starting from the camera pinhole model , derived camera imaging principle ; the proof object points , respectively, along the three axes of the world coordinate system (XW axis , YW axis and ZW axis) direction very different laws imaging , presents a new parallel calibration method, the calibration work of the three-dimensional information of the object point (XW, YW and ZW) for parallel processing ; proposed a new normalization algorithm , to further improve the calibration accuracy of the XW and YW . Calibration method proposed camera calibration experiments and experimental analysis . By error analysis , the paper pointed out the possible ways to improve the calibration accuracy . Experimental results show that : the calibration method can be expanded calibration accuracy and calibration speed camera calibration space ; proposed normalization algorithm is able to further improve the calibration accuracy XW and YW ; ZW calibration is a whole wide range of calibration XW and YW its reconstruction of the standard deviation is greater than reconstructed standard deviation . Improve the the ZW axis calibration accuracy , will be the focus of the next step .
|
Related Dissertations
- High Speed Frequency Measurment and Non-Linearity Correction of Frequency Modulated Capacitive Displacement Sensor,TH822
- FPGA/DSP Image Co-processor Technology and Ethernet Data Transmission,TP391.41
- The Research of 3D Human Motion Capture Based on Reference Points,TP391.41
- Camera Calibration and Position and Pose Detecting on Vision Measurement System of PCB,TP391.41
- Research on Anti-periodic Solutions of Delayed Cellular Neural Networks without Assuming Global Lipschitz Conditions,TP183
- Designs and Applications of Fuzzy Synthetic Evaluation Models Based on Parallel Algorithms,TP18
- Research on the Theory of Visual Object Tracking Based on Particle Filter for Autonomous Robot,TP242
- Research on Detecting Optical Fiber Geometric Parameter Based on Machine Vision,TN253
- Spectrum Variation of Vegetation in Yanzhou Coal Mine Area and Heavy Metal Stress Characteristic,X173
- Research into Visual Orientation in the Weeding Robot Based on COCM,TP391.41
- Multi- license plate location method based on CNN's Intelligent Transportation Systems Research,TP391.41
- Research on the Models of GPS Height Fitting Based on BP Neural Network and Their Applications,P228.4
- Research of QoS Optimization Based on Neural Network Prediction for Ethernet Passive Optical Network,TN929.1
- Research on Vision Detection Algorithm for Tracing Printing System,TP391.41
- The Research and Implement on Camera Calibration Technology Based on Trifocal Tensor,TP391.41
- Research on Image Denoising Algorithm in Machine Vision System,TP391.41
- Single Camera Calibration Method Based on Dual-mirror Imaging,TP391.41
- Sensitivity Study and Application of Complex Variabal Weight Function Neural Network,TP183
- Researches on the Application of the Neuron-MOS in the Neural Network,TP183
- A Reduction Method for Artificial Neural Network Inputs Based on An Improved Genetic Algorithm,TP18
- On-line Monitoring of Tool Wear for Mass Production,TG71
CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net Mobile
|