Study on the Identification of Steel Slag Mixed in Fine Aggregate for Concrete Based on Color as Feature Elements

Abstract

The frequent quality problems caused by steel slag concrete construction have received extensive attention and attention from the construction industry. How to effectively prevent the mixing of steel slag in raw materials for concrete has become a difficult problem faced by the current industry. In this paper, the phase and chemical composition of steel slag were analyzed to clarify the characteristic identification elements of steel slags and common fine aggregate for concrete. With the content of high-iron steel slag particles mixed in common fine aggregate increasing, the color of the mixed samples deepens, the diffuse absorption intensity in the visible light and near-infrared light range gradually increases, and the main peak position of the grey value gradually decreases. Through deep learning analysis, the precision and recall of TransUNet model for the mixed sample containing common fine aggregate and steel slag particles can reach more than 90.00%. Therefore, based on the color as a characteristic identification element, it can effectively identify whether fine aggregates is mixed with steel slag, so as to ensure the safety and stability of concrete structures.