Duhok, a city in Kurdistan Region of Iraq, has important historical and cultural evident in its archaeological sites. The city center contains many neglected heritage buildings that necessitate documenting as a crucial preliminary step for their preservation, facilitating a comprehensive understanding of their architectural, historical, and social significance. Consequently, the study aims to document buildings in the study region employing several documentation methods and looking at various aspects, such as architectural, historical, social, and cultural dimensions. The second goal of the study is to build an automated model utilizing advanced machine learning techniques, such as convolutional neural networks, transfer learning, and neural architectural search, to create a robust model for identifying and classifying architectural styles across various regions of the Kurdistan Region. The initials result of the study shows the unique attributes of Islamic, vernacular, modern, and postmodern architectural styles within Duhok’s legacy. The machine learning categorization of the model is very accurate, highlighting its potential as a reliable analytical tool for identifying and classifying architectural styles in Duhok city and across Kurdistan Region.
Cebar J. Sadiq (Wed,) studied this question.