1 / 9

Improving Image Contrast Using Locally Adaptive Binarization Techniques

Improving Image Contrast Using Locally Adaptive Binarization Techniques. Mike Anderson Oliver Nina Mike Wynn. Motivation. Looking at ways to improve the readability of text from historical documents Initial solution required a pre-generated binarized image for processing. Initial Solution.

milt
Download Presentation

Improving Image Contrast Using Locally Adaptive Binarization Techniques

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. Improving Image Contrast Using Locally Adaptive Binarization Techniques Mike Anderson Oliver Nina Mike Wynn

  2. Motivation • Looking at ways to improve the readability of text from historical documents • Initial solution required a pre-generated binarized image for processing

  3. Initial Solution

  4. Our Solution • Allows process to be used on any image even if a binarized version of the image isn’t available • We use a local window to identify and enhance the contrast in that part of the image • We can compare Otsu(1979) and Su(2010) algorithms. • We allow the user to control the amount of the information from the original image to blend with the binarized image

  5. Background - Otsu Algorithm • Published in 1979 • Global thresholding algorithm • Uses the image histogram to pick a threshold • Calculates Maximum Between-class Variance (MBV) of the modes • Chooses the threshold that allows MBV

  6. Background - Su Algorithm • Published at DAS 2010 (HDIBCO '10 winner) • Local binarization method • Performs edge detection using local min. and max. • Approximates the width of text stroke • Builds a small window based on the stroke width • Performs local thresholding based on neighboring pixels

  7. Demo

  8. Conclusions • Does provide a way to enhance contrast • Speed is still an issue • More research could provide way to tile image and generate full binarized image that could be reused instead of regenerated every time • Improve Su algorithm

  9. Thank You

More Related