Αναγνώριση και αποφυγή οδικών εμποδίων με χρήση εικόνας και μηχανικής μάθησης (Bachelor thesis)

Νικολάου, Πόπης


In this thesis we will talk about machine learning and how it has helped and continues to help in computer science. At first we will refer to autonomous driving, the benefits that it has and the levels of autonomous driving. Then there will be an introduction to machine learning, the following process, we will mention the categories of machine learning and we will deal with one of them, the Deep Learning one, which was used for the development of this thesis. We will go on with parallel processing, how it helps in speeding up calculations and we will mention the two parallel processing libraries that were used, CUDA and Tensorflow. The fourth chapter will describe the data set that was used, and the convolutional neural network SqueezeDet, which detects objects in image data. Finally, we will look at the results obtained from the completion of this thesis and the conclusions drawn during it’s preparation
Institution and School/Department of submitter: Σχολή Τεχνολογικών Εφαρμογών / Τμήμα Μηχανικών Πληροφορικής
Keywords: μηχανική μάθηση;αποφυγή οδικών εμποδίων;αναγνώριση οδικών εμποδίων;χρήση εικόνας;Deep Learning;CUDA;Tensorflow;SqueezeDe
Description: Πτυχιακή εργασία--ΣΤΕΦ-Τμήμα Μηχανικών Πληροφορικής, 2018—10038
URI: http://195.251.240.227/jspui/handle/123456789/11906
Appears in Collections:Πτυχιακές Εργασίες

Files in This Item:
File Description SizeFormat 
NIkolaou.P.pdf2.68 MBAdobe PDFView/Open



 Please use this identifier to cite or link to this item:
http://195.251.240.227/jspui/handle/123456789/11906
  This item is a favorite for 0 people.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.