Download PDFOpen PDF in browser

Iranian Architectural Styles Recognition Using Image Processing and Deep Learning

EasyChair Preprint 10061

10 pagesDate: May 10, 2023

Abstract

Iranian architecture or Persian architecture is the design of buildings of Iran and parts of the rest of West Asia, the Caucasus, and Central Asia. Its history with its unique characteristics goes back to at least 5,000 BC. As regards, Iran is located in the Middle East and the Middle East is in danger of possible wars, such as Iraq, Afghanistan, and Syria. Always during war, some historical monuments are unintentionally damaged or bombed and easily destroyed or suffer a lot of damage, and since the historical monuments of each country belong to all the people of the world, so weed to try to preserve them. In this paper, we propose a system for the automatic detection and recognition of Monuments based on Deep Learning methods. This system can be activated on the attacker's war equipment and the attacker can find out the date of construction of the building according to its architectural style and receive the necessary warning not to target the building. We have also prepared a dataset with about 3000 photos according to six styles of Iranian architecture of Iranian historical monuments that can be used in other sciences and applications. Also, this way can be helpful for tourists to familiarize themselves with Iranian historical monuments without the need of a guide to present by using cellphone photos to get information about the period of the historical monument and the style of that architectural building.

Keyphrases: Iranian architecture, deep learning, image classification, image processing

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
@booklet{EasyChair:10061,
  author    = {Mohammad Tayarani Darbandy and Benyamin Zojaji and Fariba Alizadeh Sani},
  title     = {Iranian Architectural Styles Recognition Using Image Processing and Deep Learning},
  howpublished = {EasyChair Preprint 10061},
  year      = {EasyChair, 2023}}
Download PDFOpen PDF in browser