Intrinsic image extraction from a single image

Yun Chung Chung, Cherng Shen, Robert Bailey, Sei-Wang Chen

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

An image is often modeled as a product of two principal components: illumination and reflectance components. The former is related to the amount of light incident on the scene and the latter is associated with the scene characteristics. The images formed from the two components are referred to as the illumination and the reflectance images; both are called the intrinsic images of the original image. The illumination components of the images of a fixed scene vary from image to image, while the reflectance components of the images in principle remain constant. Both reflectance and illumination images have their own applications. Intrinsic image extraction has long been an important task for computer vision applications. However, this task is not at all simple because it is an ill- conditioned problem. The proposed approach convolves an input image with a prescribed set of derivative filters. The pixels of the derivative images are next classified as being reflectance or illumination according to three measures: chromatic, intensity contrast and edge sharpness, which are calculated in advance for each pixel from the input image. Finally, a de-convolution process is applied to the classified derivative images to obtain the intrinsic images. The results reveal the feasibility of the proposed technique in both rapidly and effectively decomposing intrinsic images from one single image.

Original languageEnglish
Pages (from-to)1939-1953
Number of pages15
JournalJournal of Information Science and Engineering
Volume25
Issue number6
Publication statusPublished - 2009 Nov 1

    Fingerprint

Keywords

  • Chromatic measurement
  • Intensity contrast and edge sharpness measures
  • Intrinsic images
  • Photometric reflectance model
  • Reflectance and illumination components

ASJC Scopus subject areas

  • Software
  • Human-Computer Interaction
  • Hardware and Architecture
  • Library and Information Sciences
  • Computational Theory and Mathematics

Cite this