Adaptive Thresholding using the Integral Image

Derek Bradley
Carleton University

Gerhard Roth
National Research Council of Canada

This paper appears in issue Volume 12, Number 2.
An electronic version of this article is available.

Abstract

Image thresholding is a common task in many computer vision and graphics applications. The goal of thresholding an image is to classify pixels as either “dark” or “light.” Adaptive thresholding is a form of thresholding that takes into account spatial variations in illumination. We present a technique for real-time adaptive thresholding using the integral image of the input. Our technique is an extension of a previous method. However, our solution is more robust to illumination changes in the image. Additionally, our method is simple and easy to implement. Our technique is suitable for processing live video streams at a real-time frame-rate, making it a valuable tool for interactive applications such as augmented reality. Source code is available online.

Author Information

Derek Bradley, 201-2336 Main Mall, Vancouver, BC V6T 1Z4, Canada bradleyd@cs.ubc.ca

Gerhard Roth, Institute for Information Technology, Building M50, 1200 Montreal Rd., Ottawa, ON K1A 0R6, Canada Gerhard.Roth@nrc-cnrc.gc.ca

Source Code

C/C++ implementation: source.zip

Examples

You can see examples at the author's website.

You can also download the images.

Input
image1.png
image2.png
image3.png
image4.png
image5.png

Wellner's technique
wellner1.png
wellner2.png
wellner3.png
wellner4.png
wellner5.png

Our technique
bradley1.png
bradley2.png
bradley3.png
bradley4.png
bradley5.png

BibTeX Entry

@article{BradleyRoth07,
  author = "Derek Bradley and Gerhard Roth",
  title = "Adaptive Thresholding using the Integral Image",
  journal = "journal of graphics, gpu, and game tools",
  volume = "12",
  number = "2",
  pages = "13-21",
  year = "2007",
}