1 / 26

Single Character Recognition

Single Character Recognition. CS 5185-MULTIMEDIA TECHNOLOGIES AND APPLICATIONS 2007 1st Semester - Group Project Progress Presentation Presented on 2007-10-23. Project Group. Group 08 Members: Ku Heung Chin (Ku) Yu K a m Fung (Kam) Fung K a Hang (Harry) W ang Yang       .

Télécharger la présentation

Single Character Recognition

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Single Character Recognition CS5185-MULTIMEDIA TECHNOLOGIES AND APPLICATIONS 2007 1st Semester - Group Project Progress Presentation Presented on 2007-10-23

  2. Project Group • Group 08 • Members: • Ku Heung Chin (Ku) • Yu Kam Fung (Kam) • Fung Ka Hang (Harry) • Wang Yang       

  3. Usage of Content Recognition • Printed content • Business Card Recognition • Car License plate recognition • Handwritten content • Postal address recognition • Bank cheque signature verification

  4. A top-down approach Single Character Recognition is the basis for Content Recognition !!

  5. Implementation • Programming Language • Java (J2SE 1.6) • Platform independent • Rich graphics and image processing APIs • Vector data structure APIs • Development platform • Eclipse - Java SDK • User friendly interface • Java complier and debugger • Java Syntax and spelling checking

  6. Character DB Preprocessing A … G … G Recognition How does it work?

  7. Preprocessing • Image to *.bmp format • Color(RGB) image to Grayscale image • Noise Filter • Binarization • Slant Correction • Thinning • Size Normalization

  8. Image to *.bmp format • FileChooser Class to select input image. • Accepted different image format, e.g. jpg, gif, etc… • Call some API to convert to 24-bit RGB Bitmap format (will be implemented in later phase)

  9. RGB to 8-bits Grayscale 0.3*Red + 0.59*Green + 0.11*Blue

  10. Median:80 Noise Filter • 3x3 pixels Median Filter

  11. Binarization Purpose : • Clear cut the background and the character

  12. Binarization Implementation (Simplest Approach) : • Assumption that noises are filtered out and dark text and white background • Transform gray color to black (0x00) and white (0xFF)

  13. Binarization Implementation (Simplest Approach) : • Average the color as a cut off 0xFF 0x00

  14. Slant Difficulty : • Different slant style @ ppl Solution : • Estimate the slant by a slope • Skew the image back

  15. Slant Correction (Linear Regression)

  16. Thinning Purpose : • Transform a binary image into one pixel thickness

  17. Thinning Implementation (Simple Approach) : • http://fourier.eng.hmc.edu/e161/lectures/morphology/node2.html

  18. Thinning • Will add more on implementation

  19. Image Scaling • Image scaling is a process of resizing an image. Mapping the pixels from the original image to the destination image. • It’s non-trivial process that involves a trade-off between efficiency, smoothness and sharpness.

  20. Image Scaling • Scaling method used in this prototype 1st: Edge Detection: 2nd: Calculate scaler. 3rd: Do scaling: Convert scaler into integer, enlarge every pixel by scaler times. i.e. Scaler = 2.

  21. Image Scaling • Pros: Easy to implement. • Cons: Suitable for integer scaler, not so good for rational scaler.

  22. Recognition • Usually machine learning algorithms • Simple pixel-by-pixel difference would be used as not closely related to this course. • Will be implemented in later phase.

  23. Gantt Chart

  24. Program Live Demo

  25. Q &A

  26. The End Thank you for you attentions!!

More Related