Smart Construction - The Application of Deep Learning and Computer Vision in Construction Automation

The construction industry heavily relies on photo and image data. Our internal investigations have shown that over 70% of the data generated during a new construction project consists of images captured on-site at various phases of construction. This immense volume of data provides a valuable opportunity to leverage deep learning techniques for automation and smart construction.

In this repository, I have summarized my key computer vision codes aimed at streamlining construction automation efforts. This includes image classification from field observation reports, failure detection in building envelope systems in constrcuction photos, and high-rise glazing system image classifications.