Publication Type : Conference Paper
Publisher : IEEE
Source : 2025 3rd International Conference on Intelligent Systems, Advanced Computing and Communication (ISACC)
Url : https://doi.org/10.1109/isacc65211.2025.10969241
Campus : Bengaluru
School : School of Computing
Year : 2025
Abstract : Recognizing architectural styles is essential for preserving & understanding cultural heritage, as it helps categorize & document diverse structures, highlighting their historical, cultural & artistic value. The study addresses the gap between technology & heritage by developing a framework for classifying 8 architectural styles: Buddhist, Indo Islamic, Rajput, Dravidian, Hindu, Sikh, British & Modern. Using the HOG method for feature extraction, the models capture intricate architectural details. Classification is done using machine learning models like Logistic Regression, Random Forest & XG-Boost, as well as advanced deep learning models such as DenseNet & InceptionV3. DenseNet achieves the highest accuracy of 86%, followed by InceptionV3 at 79%, showing the capability of deep learning in handling complex visual data. The study not only compares model performance but also provides a scalable method for architectural style classification that contributes to the heritage preservation.
Cite this Research Publication : Ananya Ganapathi, Deivanai Saravanan, Sneha T Raghavan, Jyotsna C, Aiswariya Milan K, Classification of Structures and Monuments based on Indian Architectural Styles, 2025 3rd International Conference on Intelligent Systems, Advanced Computing and Communication (ISACC), IEEE, 2025, https://doi.org/10.1109/isacc65211.2025.10969241