Chapter 1: Invited Addresses and Tutorials on Signals, Coding,
  Systems and Intelligent Techniques

Title: A system to enhance details on partially shadowed images

Author(s): Rafael Heitor Correia de Melo, Evelyn de Almeida Vieira, Aura Conci

Address: Computer Institute, UFF - Federal Fluminense University, Passo da Pátria 156, 24210-240 Niterói, RJ, Brazil | Computer Institute, UFF - Federal Fluminense University, Passo da Pátria 156, 24210-240 Niterói, RJ, Brazil | Computer Institute, UFF - Federal Fluminense University, Passo da Pátria 156, 24210-240 Niterói, RJ, Brazil

Reference: 12th International Workshop on Systems, Signals and Image Processing pp. 309 - 312

Abstract/Summary: The goal of this paper is to define a system to enhance details on partially shadowed images. These images are currently used on the processing of digital images when a natural scene on a sunny day is recorded or great difference on the intensity of the illumination occurs in the same image, for instance. They present significantly change in the reflectance producing reduced contrast in dark regions. To enhance these images for later use, a system based on homomorphic filter was developed. Examples illustrate the system performance and its possibility on security, restoration as well as any other kind of application, where the identification of objects and its details (not depending of it's illumination) are useful. Tests with images in grey-scale and colour were performed. RGB, YIQ, XYZ, YUV and HSL colour spaces were experimented to define the best combination for the approach. Low pass and high pass filters were implemented on the spatial domain improving the system speed.

Order a copy of this article