Scholarlink

<p>Accurate classification of clay bricks is vital for ensuring structural quality and reliability in construction. Traditionally, brick sorting at kilns relies on manual visual inspection followed by laboratory testing, a process that is subjective, time-consuming, and prone to human error. In this study, deep learning–based convolutional neural networks (CNNs) were explored as an automated approach for visual brick classification. A comprehensive image dataset of kiln-fired bricks was developed and used to train and test several pre-trained CNN architectures representing different levels of complexity and computational efficiency. The comparative analysis demonstrated that deep learning models can effectively distinguish between multiple classes of bricks, offering a fast, consistent, and objective alternative to manual inspection. The findings highlight the potential of CNN-based models to support automation and quality control in the masonry industry.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Classification of clay bricks using vision-based machine learning algorithms

  • 期刊论文

  • 发表日期：

    2026年1月27日

查看全文
  • Muhammad Faizan,
  • Muhammad Salman,
  • Muhammad Noman,
  • Ali Akhtar,
  • Afaq Ahmad,
  • Matiullah Bakhtyar

摘要

Accurate classification of clay bricks is vital for ensuring structural quality and reliability in construction. Traditionally, brick sorting at kilns relies on manual visual inspection followed by laboratory testing, a process that is subjective, time-consuming, and prone to human error. In this study, deep learning–based convolutional neural networks (CNNs) were explored as an automated approach for visual brick classification. A comprehensive image dataset of kiln-fired bricks was developed and used to train and test several pre-trained CNN architectures representing different levels of complexity and computational efficiency. The comparative analysis demonstrated that deep learning models can effectively distinguish between multiple classes of bricks, offering a fast, consistent, and objective alternative to manual inspection. The findings highlight the potential of CNN-based models to support automation and quality control in the masonry industry.