html5-img
1 / 10

Recognizing Hebrew alphabet of specific handwriting

Recognizing Hebrew alphabet of specific handwriting. Introduction to Computational and Biological Vision 2010 By Rotem Dvir and Osnat Stein. Motivation. Usage of existing function on the text. Reprocessing and rearranging the text. Redesign the text.

suzy
Download Presentation

Recognizing Hebrew alphabet of specific handwriting

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. Recognizing Hebrew alphabet of specific handwriting Introduction to Computational and Biological Vision 2010 By RotemDvir and Osnat Stein

  2. Motivation • Usage of existing function on the text. • Reprocessing and rearranging the text. • Redesign the text. • Combining the text with another document. • Using other applications that require text files.

  3. Goal • Recognize handwritten Hebrew alphabet of specific handwriting, known in advance. • When given a scanned image of an alphabetic letter, we would like to recognize this letter.

  4. Chain code - reminder • Absolute • Relative

  5. Generic Algorithm • Input: scanned letter image. • Output: Numerical chain code. • Find black pixel to start the encoding from. While there is black pixel who wasn't visited: • Find a direction to continue. • Update chain code and mark the pixel as visited. • Move forward to the chosen pixel. Return chain code.

  6. Question marks • How to choose the beginning pixel? • How to choose the next direction if there is more than one option? • Which encoder to use? • What to do if we reached dead-end, but there are still black pixels left?

  7. In our project • Three different approaches • Same beginning pixel • Encoder • Method for progressing • Dead end • State machine

  8. An easier problem • Easier problem then recognize is to determine if an input letter is a specific target letter or not. • cannot produce good results, as long as using chain codes as described before

  9. Conclusions • The chain code must describe the shape of the letter. • For thin letters we had good results, but for regular letters it is much harder. • Improvement • Reduction

  10. Questions…? Thank you! 

More Related