Ebook: Image Processing with Scilab and Image Processing Design Toolbox

Учебник, 2011
Contents
Introduction
Characteristic Properties of Images
Basic Operations on Images
Introduction
Thresholding
Blob Analysis
Filtering
Watershed Transform and Distance Transform
How to Detect Objects in Images
Introduction
Thresholding and Blob Analysis
Edge Detection and Watershed Transform
References
There are various applications of image processing, e. g. detection of surface defects in in-
dustrial quality control, detection of anatomical landmarks in surgery, counting cells in bio-
technology and classification of regions in remote sensing. Images are generated by opti-
cal cameras, ultrasound, x-ray machines and other imaging devices.
When processing an image with a computer, it must be digitized or created in a digital for-
mat. There are some basic methods to distinguish between objects and background and to
describe regions in digital images. This document explains how digital images can be re-
presented mathematically and basic methods for finding objects and describing regions.
Scientists, engineers and medical doctors who intend to analyze images usually know how
their images were generated and how the objects of interest look like. However, these ob-
jects must be described mathematically when they are to be detected automatically with a
computer. The goal of this document is introducing these mathematical descriptions in a
way easily to understand.
The remaining part of the document is organized as follows: In chapter 2, a mathematical
representation of digital images and characteristic properties of images are introduced. In
chapter 3, basic image processing operations such as segmentation into connected regi-
ons and filtering are introduced. In chapter 4 methods for object detection are presented.
References are given in the last chapter.
Examples are implemented with Scilab 5.3.1 and Image Processing Design Toolbox (IPD)
8.0. The photograph used in this document is taken from IPD. Scilab is available for down-
load at www.scilab.org. IPD can be downloaded from atoms.scilab.org/toolboxes/IPD.