Abstract
Traditional image recognition methods are difficult to accurately identify Ground Penetrating Radar (GPR) images. This paper presents a method to detect and locate underground objects automatically. First, we used the Faster R-CNN framework to detect buried objects in B-SCAN. Due to the lack of real data for training, we used the GPRMAX toolbox to synthesize more GPR images and pre-trained on the PASCAL voc2007 database. Then, the Faster R-CNN framework based on the pre-trained CNN is trained and fine-tuned on the real and simulated GPR data. The detected rectangular region is converted into a binary image and the edge is detected. Finally, the hyperbola can be fitted successfully by using Randomized Hough Transform (RHT). Compared with the traditional computer vision method, the method presented in this paper is significantly improved.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
