Scholarlink

For concrete constructions to be long-lasting, safe, and aesthetically pleasing, form-finished concrete quality is crucial. Defects including cracks, cavities, surface roughness, and discoloration can reduce structural integrity and raise maintenance expenses. A comprehensive assessment of computer vision-based techniques for analyzing the quality of form-finished concrete is presented in this work. It discusses fault kinds, their classification, and the significance of quality evaluation in construction. Defect detection and classification are studied using a variety of image processing and deep learning techniques, including edge detection, texture analysis, convolutional neural networks (CNNs), and machine learning algorithms. To improve the accuracy and efficiency of quality evaluation in the construction sector, the study also discusses present issues and potential directions for future research, such as incorporating automation, artificial intelligence (AI), and real-time defect detection systems.

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

Innovative Machine Learning Approaches for Real-Time Inspection of Form-Finished Concrete Using Image Processing

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Supriya Sawant,
  • R. B. Ghongade,
  • Aditi Sharma,
  • Milind Gayakwad,
  • Anand Shinde,
  • Shwetambari Chiwhane

摘要

For concrete constructions to be long-lasting, safe, and aesthetically pleasing, form-finished concrete quality is crucial. Defects including cracks, cavities, surface roughness, and discoloration can reduce structural integrity and raise maintenance expenses. A comprehensive assessment of computer vision-based techniques for analyzing the quality of form-finished concrete is presented in this work. It discusses fault kinds, their classification, and the significance of quality evaluation in construction. Defect detection and classification are studied using a variety of image processing and deep learning techniques, including edge detection, texture analysis, convolutional neural networks (CNNs), and machine learning algorithms. To improve the accuracy and efficiency of quality evaluation in the construction sector, the study also discusses present issues and potential directions for future research, such as incorporating automation, artificial intelligence (AI), and real-time defect detection systems.