| description abstract | The information of exact locations of underground utilities is an essential piece of evidence for preventing utility strikes in excavation work. Ground penetrating radar (GPR), which has emerged as a promising, nondestructive solution for this purpose, is capable of capturing radar reflections that are then recorded as GPR scans. To determine the location, dimension, size, and spatial configuration of pipes, radargrams must be further interpreted to extract the shapes (e.g., hyperbolas and lines) and to identify the feature components (e.g., hyperbola apex, rising and trailing segment, and junctions of intersecting hyperbolas). This paper introduces a new drop–flow algorithm that automates the detection and decomposition of GPR signatures into feature components in two-dimensional scans. Commencing at a strip of pixels from the top of the edge of the scan image, the algorithm mimics the motion of a raindrop falling or flowing as it touches the edge pixels of the image. The movement of the raindrop completes the decomposition of the GPR signature when it touches the ground (i.e., the bottom of the edge image). This new algorithm was tested using both synthetic and field data. The promising results indicate the drop–flow algorithm’s ability to segment the intersecting hyperbolas and to identify the feature components of each hyperbola, forming the basis for estimating the spatial configuration, size, and location of underground pipes. | |
