Status: closed
Supervisors: Markus Diem, Florian Kleber, and Fabian Hollaus

Problem Statement

Historical documents are often scanned or photographed to publish facsimile editions. Thus, it is important to know the exact scale of the image, and the correct color information. With this information, the document can be reproduced in such a manner that the dimension and color of the reproduction matches the original document.

For that reason documents are imaged together with color patches. The following image shows a historical document together with a color patch.

Goal

The goal of the Praktikum/Bachelor thesis is the detection of the color patch within the image, based on a template of the color patch. After the detection, the ruler of the patch should be analyzed to determine the scale of the image. Furthermore, the color patches must be extracted and a color correction should be performed automatically (e.g. white balance).

Workflow

• Literature research
• Implementation in Matlab or C++ with OpenCV
• Development and evaluation of the system
• Written report or bachelor thesis (in English) and final presentation

Requirements

• Matlab, or C++ and OpenCV