Buried object detection from B-scan ground penetrating radar data using Faster-RCNN

TL;DR

This paper adapts the Faster-RCNN framework for detecting buried objects in ground penetrating radar images, leveraging simulated data due to limited real data, and demonstrates improved detection performance over classical methods.

Contribution

It introduces a novel approach combining simulated radargrams and transfer learning with Faster-RCNN for underground object detection in GPR images.

Findings

Significant improvement over classical methods

Effective use of simulated data for training

Promising results with limited real data

Abstract

In this paper, we adapt the Faster-RCNN framework for the detection of underground buried objects (i.e. hyperbola reflections) in B-scan ground penetrating radar (GPR) images. Due to the lack of real data for training, we propose to incorporate more simulated radargrams generated from different configurations using the gprMax toolbox. Our designed CNN is first pre-trained on the grayscale Cifar-10 database. Then, the Faster-RCNN framework based on the pre-trained CNN is trained and fine-tuned on both real and simulated GPR data. Preliminary detection results show that the proposed technique can provide significant improvements compared to classical computer vision methods and hence becomes quite promising to deal with this kind of specific GPR data even with few training samples.