Dissertation > Excellent graduate degree dissertation topics show

Research of the Identification Arithmetic of Fabric Defaults Using Computer Software

Author: QiLin
Tutor: WangSanWu
School: Wuhan University of Technology
Course: Mechanical and Electronic Engineering
Keywords: Fabric defect Automatic detection Texture recognition Perceptron neural network
CLC: TS107
Type: Master's thesis
Year: 2006
Downloads: 225
Quote: 8
Read: Download Dissertation

Abstract


As a textile giant, China so far has not developed a practical automatic fabric inspection system, introduced from abroad the price is too high, the machine configuration and mode of operation is not well adapted to our specific production situation. On the basis of the study and research on nearly 20 years of domestic and international fabric defect automatic detection technology development course and successfully market outcomes, we pointed our fabric defect automatic detection of the direction of development, which is based on computer software The detection method. And propose a set of effective defect detection software algorithms, to launch self-developed automatic fabric defect detection system provides a theoretical basis and technical support for our country. In this paper, the detection algorithm: parallel to the white woven fabrics fabric photographed with a digital camera in a uniform light irradiation fabric image entered into the computer; First fabric image is converted to grayscale, use the histogram equalization method to enhance the clarity of the image and contrast; then use the texture density similar to the the GLCM \gray threshold segmentation the fabric image binarization extracted the defect part of the expansion \hidden layer) perceptron neural network recognition and classification of defects. We successfully identified research fabric texture density used in the fabric defect detection, specifically for image data compression, greatly reduces the amount of data processing to accelerate the speed of defect detection, and removal of a large number of redundant data reduces noise interference, so that more accurate detection result. In the experiment we missing by the white plain woven fabrics, heavy by the lack of latitude, heavy weft, holes, oil and other kinds of the most common, most defect identification, correct identification rate of 96%. The method proposed in this paper in texture recognition algorithms, compression algorithms, the defect eigenvalues ??definition and extraction also further complement and optimize. Improve and perfect the algorithm can further improve the processing speed and degree of automation, can identify the type of defect can also be increased.

Related Dissertations

  1. Woven Fabric Linear Research of Detection Based on AR Model,TP391.41
  2. Based on DSP and FPGA fabric inspection machine image processing system design,TP391.41
  3. Space vehicle test system automation research and design,U471
  4. Based on Artificial Immune Algorithm Research and Application of On-line Product Testing,TP274
  5. Study of Image Feature Extraction and Texture Classification Algorithm,TP391.41
  6. Research on Remainder Detection Technique for Military Electronic Components Based on Random Vibration,TN607
  7. Development of the Electric Motor Shaft Manufacturing Quality Detection and Control System,TM303.5
  8. Highway traffic incident detection modeling and application research,U491.116
  9. Automatic detection system of the organic electroluminescence light emitting device spots,TN383.1
  10. Based on Virtual Instrument Technology projectile multi-parameter detection system,TP274
  11. Research of the Algae Image Texture Based on Morphology and None Entirely Tree Wavelet Decomposition,TP391.41
  12. ECG Automatic Analysis Based on Wavelet Transforms,TN911.6
  13. Research on Foreign Fiber Detecting and Clearing Online System,TS112.6
  14. Based on digital image processing technology rock fracture fast acquisition and processing of information,TP391.41
  15. Fabric Defects Detection Research Based on Image Processing Technology,TP391.41
  16. The Identification of Fabric Defects Based on Curvelet Transform and BP Neural Network,TP391.41
  17. Technology Research for Engine Oil Coolergas Tightness Detection Testing Process Optimization and Implementation,TK403
  18. Design of the Freight Car’s Fault Automatic Detection System Based on Image Processing,TP274
  19. Localized Generalization Error Model of Multilayer Perceptron Neural Networks,TP183
  20. Design of Railway Truck Off-Gauge Detection System,TP274
  21. The Design and Application of the Air-conditioning Production Line Detection System,TP311.52

CLC: > Industrial Technology > Light industry,handicrafts > Textile industry,dyeing and finishing industry > General issues > Standards and testing of textiles
© 2012 www.DissertationTopic.Net  Mobile