OPUS


Detection and Classification of Video sequence during Night-Time

  • Object detection and classification is active field of research inmachine learning and computervision. Depending on the application there are different limitations to adjust to, but also possibilities to take advantage of. In my thesis, We focus on classification and detection of video sequence during night-time and the proposed method is robust since it does use image thresholding [8] which is commonly use in other methods and the thesis uses histograms of oriented gradients (HOG) [37] as features and support vector machine (SVM) [74] as classifier. It is of great importance that the extracted features from the images should be robust and distinct enough to help the classifier distinguish between high-beam and a low-beam. The classifier is part of the object detection which predicts whether or not a testing image matches one group or the other. In our case that is predicting whether or not an image belongs to high or low-beam sequence.

Download full text files

  • application/pdf HSMW-Thesis-Vorlage.pdf (6424 KB) eng

Export metadata

  • Export Bibtex
  • Export RIS

Additional Services

Share in Twitter Search Google Scholar
Metadaten
Author:Rudresh Naidu Satharasi
Advisor:Thomas Villmann, Tina Geweniger
Document Type:Bachelor Thesis
Language:English
Year of Completion:2018
Granting Institution:Hochschule Mittweida
Release Date:2019/03/04
GND Keyword:Objekterkennung
Institutes:Angewandte Computer‐ und Bio­wissen­schaften
Dewey Decimal Classification:006.4 Mustererkennung
Access Rights:Innerhalb der Hochschule
Licence (German):License LogoEs gilt das UrhG

$Rev: 13581 $