Publication:
Image analysis on remote sensing images

Loading...
Thumbnail Image
Date
2006-05-01
Authors
Azuddin, Aratus Husna
Journal Title
Journal ISSN
Volume Title
Publisher
Research Projects
Organizational Units
Journal Issue
Abstract
Humans easily identify roads in remote sensing images, but this task is difficult to automate using computers. In order to identify road segments from satellite images, human beings (experts) seem to first search for a set of linear and curvilinear features and then apply knowledge or use experience to decide (or guess) whether these linear and curvilinear features are roads are not. From the image, the roads can be detected in a single line to perform as a map or digital map nowadays. The selected image that I used here is a satellite image of an area which had a network of roads. The satellite image is analyzed from one step to another step in different process using MATLAB software. The important processes must be done to the satellite image before the single road lines can be detected. A detection theory is used which can be done by computer to overcome drawbacks of current theories and detect major roads in an image with high speed and high precision. The technique is combining image segmentation and image enhancement process. In this project, the enhancement of the satellite image is the important process in road detection process. Firstly, an image in any file formats is load/save into the workspace folder in MATLAB. The input image is converted into grayscale form. Before continue the next step, first the grayscale image is enhanced for better intensity. In this process, many function of enhancement are used to get the better image for the output. The function that used for the enhancement processes are imadjust, histeq, and adapthisteq. Comparing all the result using those functions, the best-enhanced image is selected to the next step. The process of image segmentation that involve in this project is edge detection. The edge detection is done using Laplacian of Gaussian detector. The segmentation process required an intensity image for all the process involved.
Description
Keywords
Citation