Dissertation > Excellent graduate degree dissertation topics show

Feature-based image defect detection algorithm research and application

Author: YeXingZuo
Tutor: WuJianJie
School: Huazhong University of Science and Technology
Course: Software Engineering
Keywords: Image feature Defect Detection Printing images Machine Vision
CLC: TP391.41
Type: Master's thesis
Year: 2011
Downloads: 88
Quote: 0
Read: Download Dissertation

Abstract


With the production and technological progress, people increasingly demanding high quality products, based on machine vision inspection systems become an important quality control tool. With the rapid development of machine vision detection system accuracy and speed the development of machine vision to become a bottleneck. To overcome this difficulty, proposed feature-based image defect detection algorithms. In order to improve the stability of the feature detection algorithm, proposed two preprocessing algorithm. Conventional methods for adaptive image enhancement shortcoming, we propose a fast dynamic image enhancement algorithms, based on the distribution of gray areas, gray stretch the image automatically to the desired position, the image of the original grayscale pixel stretched to the maximum 255, and the image of the original minimum grayscale pixel shrink to zero, the other gray gray original distribution according to proportional change. Algorithm experiment used four typical experiment map and respectively for low brightness and high brightness image enhancement, the experimental results show that for different brightness and different types of image enhancements are ideal, and compared to the conventional method for enhancing adaptability strong. Conventional methods for image denoising easily cause image blur shortcomings proposed intelligent filtering algorithm, according to a pixel neighborhood of the pixel difference between whether a threshold range to determine whether the pixel filtering. This method can remove noise while preserving image edges. Algorithm experiment used a common median filtering, Gaussian filtering algorithm and intelligent edge information-rich image filtering processing, experimental results show that the filtering and conventional methods, intelligent filtering algorithm for edge retention capacity is ideal, the time complexity is not high, practical, strong image for conventional detection methods can not distinguish between the type of image defects disadvantages propose a feature-based detection methods. According to the test image defect classification, selection and extraction test chart detection feature set, individually analyze and compare the detection feature set for an exact match to treat mapping for defect detection. Since only a relatively small amount of the characteristic data, to achieve a fast and precise detection purposes. Experimental results show that the algorithm, feature detection algorithm matching information according to the characteristic of the image defect classification algorithm and high efficiency. Good image detection is to solve the contradiction between accuracy and speed of a train of thought, has a high application value.

Related Dissertations

  1. FPGA/DSP Image Co-processor Technology and Ethernet Data Transmission,TP391.41
  2. Multi-currency Notes Technology Research and Implementation,TP391.41
  3. Camera Calibration and Position and Pose Detecting on Vision Measurement System of PCB,TP391.41
  4. Research on the Theory of Visual Object Tracking Based on Particle Filter for Autonomous Robot,TP242
  5. Research on Detecting Optical Fiber Geometric Parameter Based on Machine Vision,TN253
  6. Design of Image Inspection Subsystem of the Horizontal Industrial CT-DR,TP391.41
  7. Research into Visual Orientation in the Weeding Robot Based on COCM,TP391.41
  8. Research on Vision Detection Algorithm for Tracing Printing System,TP391.41
  9. Spatial Uncertainty in Stereo Visual Odometry,TP391.41
  10. Research on Image Denoising Algorithm in Machine Vision System,TP391.41
  11. Woven Fabric Linear Research of Detection Based on AR Model,TP391.41
  12. Research on Machine Vision High-speed Positioning System Based on Linear Motor,TM33
  13. Study on Defect Inspection of Flip Chip Based on Ultrasonic Excitation,TN407
  14. Based on active infrared and ultrasound scan flip-chip Defects Detection,TN407
  15. Based on image processing printed circuit board defect detection of,TN41
  16. Helmet based on image processing spatial position measurement,TP391.41
  17. Fabric based on genetic programming adaptive orthogonal wavelet construction and optimization,TP274
  18. Orchidaceae based on embedded intelligent control system for plant tissue culture,TP273.5
  19. Embedded ampoule liquid foreign online testing system developed,TP368.1
  20. Rules based on support vector machine parts machine vision inspection technology,TP274
  21. Crystal wafer automatic sorting technology research,TP274

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