Abstract

Missing data problem frequently occurs during data acquisition in ground-penetrating radar (GPR) and recovery of the missing entries prior to any processing is vital in GPR imaging. Existing missing data recovery methods are based on low-rank matrix completion or the recently proposed deep generative networks. However, the former approaches suffer from producing satisfying results under severe missing data cases and the latter require a large amount of data for training. This study proposes two methods based on deep networks for the missing data recovery. The first method uses pyramid-context encoder network (PEN-Net) architecture which consists of three parts: attention transfer network, guided Pyramid-context encoder, and a multi-scale decoder. Although the method needs training, it requires considerably less data compared to the existing U-Net based method. The second method, deep image prior (DIP), is a regularization based data recovery method which uses an untrained network as a prior. This method does not need any training, network weights are initialized randomly and updated during the iterations to minimize the cost function. Different experiments are reported for both pixel and column-wise missing cases in simulated and real data. The simulated data results show that the proposed methods have a noticeably better performance than conventional methods for the challenging pixel-wise case around 17–27% and moderate level column-wise missing case around 15%. Besides, they can also deal with extreme column-wise missing data cases where the conventional methods fail completely. Real data results further verify the superiority of the proposed methods.

Highlights

  • Published: 6 February 2022Ground-penetrating radar (GPR) is a non-destructive measurement method that allows the detection of objects lying beneath the surface

  • By incorporating deep image prior (DIP) [30], we extend the deep learning based approach of [42] to the single image data recovery case which is common for field studies

  • There are several ground-penetrating radar (GPR) images, together with the conventional methods as Go Decomposition (GoDec), NMC, low rank matrix fitting (LmaFit), and norm minimization (NNM) both pyramid-context encoder network (PEN-Net) and DIP models can be performed since PEN-Net needs training

Read more

Summary

Introduction

Ground-penetrating radar (GPR) is a non-destructive measurement method that allows the detection of objects lying beneath the surface. Various data recovery methods are introduced in the literature for natural images and seismic data These methods are based on interpolation [4], matrix completion [5] or more recently on deep networks [6]. There are other popular matrix completion methods in image processing domain and they have been recently applied to the GPR missing data problem in the work of [16]. Patch-based methods can be considered successful for inpainting with similar contexts This approach can not capture image semantics, it performs poorly on images with complex patterns and may not be applicable to GPR images. DIP method formulates the data recovery as an optimization problem with an already trained deep network structure used as prior through the optimization steps and it provides a solution for a highly corrupted single image.

Missing Data Recovery in GPR
Proposed Data Recovery Methods
Data Recovery by Generative Models
Experimental Results
Implementation
Simulation Dataset Results
Pixel-Wise Missing Data Case
Column-Wise Missing Data Case
Real Data Results
Real Data-I
Real Data-II
Conclusions
Full Text
Published version (Free)

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 call