About our Sessions

According to the Texas Commission on Environmental Quality (TCEQ), concrete is Class 3 insoluble nonhazardous industrial solid waste which requires a renewable permit for disposing construction debris. A renewable permit application requires not only an application fee but an annual generation fee, an annual facility fee, and a monthly management fee. In addition, fees in disposals are determined by the ton. The Environmental Protection Agency (EPA) published a report on Construction and Demolition Debris Management in the United States in March 2020 and demonstrated that concrete contributed greater than 50% of debris to landfill (approximately 66.5M tons). For the City of Dallas, assuming a concrete density of 150 lb/ft3, generating 2.15M tons of concrete solid waste annually is detrimental for ecosystems and contributes to climate change because carbon dioxide is emitted during compressive test specimen generation, transportation, and disposal activities. A reduction in the management of concrete solid waste could be significant using a digital QA/QC inspection technique. The concrete vision framework will have a direct and measurable impact on reduced energy consumption, fuel savings, reallocation of resources, and reduced concrete solid waste from infrastructure projects through digital QA/QC inspections, benefiting a broad range of communities, taxpayers, and stakeholders. This technology will contribute to equipping transportation and public works agencies with future AI-powered platforms that performs rapid, easy-to-use, environmentally-friendly, non-destructive prediction of concrete compressive strength simplifying concrete strength testing.

Digital image analysis of RGB images have demonstrated a promising relationship between concrete surface color and hardness during the early stages of curing which may indicate an increase in compressive strength as the concrete ages. This topic will demonstrate the usefulness of concrete image recognition and hardness estimation from a single RGB image.

1. Explain the characteristics of concrete color pixels from images when light reflects from concrete surfaces.
2. Learn how concrete color pixels can be used as a feature extractor in machine learning models for image recognition.
3. Learn how to leverage the power of digital image processing and analysis to estimate the hardness of concrete from a single RGB image.