Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/17013
Title: Tech-X DSTA challenge vision-360 indoor cognitive navigator
Authors: Antonius, Tio Favian
Keywords: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
Issue Date: 2009
Abstract: The purpose of this project is to design and implement a strategy to visually recognize corridors and plan paths for building a tree-based map. This project is a continuation of the previous final year project about robot localization. Through experiments, various image processing techniques such as polar transform, color image segmentation, image histogram equalization, image erosion, etc has been reviewed. The final purpose of reviewing those techniques is to provide best images as an input to enable the module to correctly identify corridors and navigate the robot towards it. Corridors are detected on a segmented color image, which looks like a matrix with colored cells. The main idea is to get the floor color, and then compare the similarity of the color profile in a column starting from the lowest cell to the highest cell. If the color is similar, then the column number is kept as a possible corridor. Experiments have been conducted to examine the reliability of the module. The results have shown that this module can work best if the corridor has sufficient brightness. However it still can work in low light intensity corridors with help of histogram equalization technique. Integration with the main robot has been done. Therefore now the module is able to give directions to the robot through a UDP communication protocol. Lastly, the proposed technique for map construction is to generate the map by using a sequence number as will be discussed later on the report.
URI: http://hdl.handle.net/10356/17013
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
TioFavianAntonius09.pdf
  Restricted Access
2.09 MBAdobe PDFView/Open

Page view(s) 50

316
checked on Sep 30, 2020

Download(s) 50

7
checked on Sep 30, 2020

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.