Integrated Approach to Simultaneously Determine 3D Location and Size of Rebar in GPR Data

Wavelet Scattering Network-Based Machine Learning for Ground Penetrating Radar Imaging: Application in Pipeline Identification

“…Although the development of radargram interpretation approaches from manual processing to machine learning is taking place, most researches (see, e.g., in [41][42][43]) on object detection only determined two characteristics: whether the objects were detectable and where the objects were. Some researches have determined the sizes of rebars [44] and small-scale voids [45] inside concrete successfully, but investigations inside the complicated underground sections become difficult. For instance, Pasolli et al [46] attempted to estimate the buried object size, but they only utilized numerically produced data and the mean error was up to 18.6%; Luo and Lai [47] failed to determine the subsurface void sizes as the identified magnitudes significantly different from the actual sizes.…”

An Integrated Framework for BIM Development of Concrete Buildings Containing Both Surface Elements and Rebar

“…Furthermore, a three-dimensional rebar arrangement can be determined by scanning both directions of an element. Figure 1 presents the principle of generating rebar signals in GPR data [10]. Thus, identifying rebar from GPR data is synonymous with processing hyperbolic patterns.…”

“…For rebar recognition, since rebar signals in GPR data are identical and have distinguishable characteristics compared with other signals (i.e., strong noise, cross rebar signal, and direct wave [10]), researchers have proposed a number of methods to recognize rebar according to this property, and existing studies can be classified as (1) machine learningbased method and (2) pattern-based method. For the machine learning-based method, the features of rebar signals are learned automatically from training datasets by algorithms and then used to recognize rebar from testing datasets.…”

“…For rebar localization, Bungey [47] has stated that rebar depth and size determination is not a reliable application of GPR due to the inevitable noise, which blurs rebar signals dramatically. Nonetheless, researchers have already proposed some approaches, which can be classified as (1) software method that develops algorithms to directly derive rebar depth and size from GPR data [10,48], and (2) hardware method that modifies the GPR device itself or integrates GPR device with other tools [46,49]. However, both methods have significant limitations: the former is sensitive to noise, and the latter s high cost.…”

“…Ground penetrating radar (GPR) provides the potential to accomplish these objectives as it can obtain rebar's spatial information by scanning concrete elements and has the possibility to build the link between rebar and scanned elements. A number of solutions have been suggested to address the first two objectives [10][11][12] and research in this area is still ongoing. Thus, to fill the gap, this paper proposes an innovative procedure to combine the as-built BIM development and rebar identification.…”