Dissertation > Excellent graduate degree dissertation topics show

A Research for Geometry Shape Representation for Profiled Fibers

Author: GuoXiaoZuo
Tutor: TangLiPing
School: Donghua University
Course: Communication and Information System
Keywords: Image description TAR mean filtering TAR median filtering The mean integrated filter the TAR Histogram
CLC: TP391.41
Type: Master's thesis
Year: 2010
Downloads: 53
Quote: 0
Read: Download Dissertation

Abstract


The fiber blend ingredients detection plays a vital role in the import and export of textiles as well as production and sales, but the technical difficulty in the inspection, the inspection process is quite time-consuming. Research topics from the \The object of the component. Profiled fiber geometry research have different cross-sectional shape of the profiled fiber geometry recognition, to provide help for of profiled fiber in the textile fiber product performance. Restrictions exist due to the fiber samples in the production process, due to the acquisition of equipment deficiencies in performance at the same time, the most important part of the study is the fiber image the effective image description. Profiled fiber image description process in two steps: First, extract useful information on the picture, remove useless information on the picture; another step in the input images in a large number of fiber data saved independently. The processing of the above steps to ensure the correctness of the input data in the subsequent identification process. Triangular area of ??the multi-scale representation of (Multiscale Triangle AreaRepresentation, TAR) is a novel method of graphic description, the characteristics of the boundary of the area of ??a triangle sampling to judge graphics pits and bumps where. As local classified area of ??a triangle method has its unique characteristics of the curvature at different scales, so this paper the theoretical basis of in-depth research. The introduction of multi-scale description operator TAR mean filtering (TAR Average Filtering, for short the TAR-AVG), and TAR median filtering (TAR the Median Filtering TAR-MED), two methods for closed graphic boundaries bump characteristics. TAR-AVG method accuracy has improved significantly, but the algorithm does not have good versatility. TAR-MED with wide versatility, the calculation results, but there are more redundant point. In this paper, a TAR in the mean filter (TAR Median AverageMixed Filtering, referred to as TAR-MAMF) method. In order to extend the application field of the TAR-MAMF method, improve the TAR formula, modified to enable the results to deviation sinusoidal get modified in the TAR Mean Integrated Filter (represented with TAR-MAMFW). Longitudinally through the data, the method accumulation histogram output result, the effective characteristic parameters extracted from the histogram results purpose of geometric shape recognition. In the experimental section, first profiled fiber database fibers and adhesion fiber graphics were used TAR-AVG the TAR-MED, the TAR-MAMF well the TAR-MAMFW method obtained the geometry of the identification results. Secondly, the different types of graphics in standard graphics library MPEG7 above four methods were used for processing. The two types of results proved the versatility and accuracy of the TAR-MAMFW method.

Related Dissertations

  1. The socks color vision grading system based on computer vision,TP391.41
  2. Research of High Speed Image Pre-processing System Based on FPGA,TP391.41
  3. The Study of Alarming Surveillance Systems and Intelligentization,TP277
  4. Based on chi-square test and macroblock type statistics of the video shot boundary detection algorithm,TP391.41
  5. Study of Pedestrian Detection Methods Based on Images,TP391.41
  6. Fast gradient orientation histogram based human detection algorithm,TP391.41
  7. Video-based information technology research moving target tracking,TP391.41
  8. Research on the Techniques for Color-Based Image Retrieval,TP391.41
  9. FPGA-based high-speed image preprocessing algorithm research and implementation,TP391.41
  10. Content-based image copy detection algorithm,TP391.41
  11. Tongue image color correction and results of evaluation of technical research,TP391.41
  12. Particle Filter Algorithm in Intelligent Transport System Research,TP29-AC
  13. Feature Extraction and Identification of Subsurface Objects in GPR,TN959.71
  14. Based on the Edge-Characteristics of Image Content Retrieval Research,TP391.41
  15. Research of Image Retrieval Based on Improved Partition Histogram and Wavelet Texture,TP391.41
  16. Near-duplicate Image Detection and Application,TP391.41
  17. Connected Digit Speech Recognition Based on the Acoustic Universal Structure,TN912.34
  18. Radar signal sorting Tracker Design and Implementation,TN957.51
  19. Research and Realization on Visibility Enhancement for Fog Image and Video Based on PC Platform,TP391.41
  20. Low Quality Finger Vein Image Enhancement,TP391.41
  21. Application of Color and Texture Features in Image Retrieval,TP391.3

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