In this research, data from contactless air-coupled ultrasonic measurements are used to characterize internal defects within a full-scale reinforced concrete column. The measurements are carried out using an automated scanning frame, which maintains constant air gap between the air-coupled transducer and the concrete surface, and ensures accurate spatial position of the transducer. Because concrete surface preparation and physical coupling process are not needed, this system collects a large amount of test data across the cross-section of the concrete column in a reasonably short amount of time. The acquired large amount of data enables reconstruction of internal images within the concrete element. Images are reconstructed from the data in two ways: 1) modified filtered back projection and 2) algebraic reconstruction technique. The modified filtered back projections successfully applied using the collected data and sinogram interpolation to develop internal image of structure. As compared to the iterative method, the speed of computation significantly improves without a significant difference with tomogram quality. The results prove that the proposed ultrasonic method is applicable to full-scale concrete structures, and developed images can localize internal defects in an efficient way.