Deep Learning for Detecting Building Defects

EasyChair Preprint 15661

Abstract

This review article identifies the deep learning methods for building defects detection, adapting an updated version of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure a rigorous and transparent review process. The review reveals that convolutional neural networks (CNNs) and their variations are the most popular in this field. By systematically identifying and categorizing the most relevant articles, we present a detailed taxonomy of the methods and applications. Additionally, the article explores current trends and discusses future directions, including advancements in real-time defect detection and the utilization of more diverse and comprehensive datasets.

Keyphrases: Artificial Intelligence, building defects, deep learning

@booklet{EasyChair:15661,
  author    = {Mohammed Mudabbiruddin and Imre Felde and Amir Mosavi},
  title     = {Deep Learning for Detecting Building Defects},
  howpublished = {EasyChair Preprint 15661},
  year      = {EasyChair, 2025}}