DDIS-based GPR Image Matching Method

Ground Penetrating Radar (GPR) has enabled efficient and high-resolution imaging of subsurface information using electromagnetic radiation. Despite of being cost-effective and non-destructive, GPR imaging process is sensitive to illumination variation, noise and scanning motion, making GPR image matching a challenging problem. In this paper, inspired by the success of Deformable Diversity Similarity (DDIS) in optical image matching, we propose to leverage DDIS to tackle the GPR image matching task. Extensive experiments with GPR images under different conditions are taken against other baseline methods. In addition, we explore different design choices in pre-processing and matching stages. Both quantitative and qualitative results demonstrate the promise of DDIS in the GPR image matching task.

• Publication year

  2022

• Venue

  ACM Cloud and Autonomic Computing Conference

• Publication date

  2022-11-25

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1109/CAC57257.2022.10055347
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Application of Combining YOLO Models and 3D GPR Images in Road Detection and Maintenance

  2021cited by this paper
• Scale-Adaptive NN-Based Similarity for Robust Template Matching

  2021cited by this paper
• Weighted Smallest Deformation Similarity for NN-Based Template Matching

  2020cited by this paper
• Template-matching based detection of hyperbolas in ground-penetrating radargrams for buried utilities

  2016cited by this paper
• Digital Image Processing

  2016cited by this paper
• Template Matching with Deformable Diversity Similarity

  2016influential reference
• Best-Buddies Similarity for robust template matching

  2015influential reference
• Automatic target detection in GPR images using Histogram of Oriented Gradients (HOG)

  2014cited by this paper
• Very Deep Convolutional Networks for Large-Scale Image Recognition

  2014cited by this paper
• Interpretation of Ground Penetrating Radar (GPR) image for detecting and estimating buried pipes and cables

  2013cited by this paper
• Fast-Match: Fast Affine Template Matching

  2013cited by this paper
• Using pattern recognition to automatically localize reflection hyperbolas in data from ground penetrating radar

  2013cited by this paper
• An improved Normalized Cross Correlation algorithm for SAR image registration

  2012cited by this paper
• Fast and high-performance template matching method

  2011cited by this paper
• Globally Optimal Estimation of Nonrigid Image Distortion

  2011cited by this paper
• Template Matching Techniques in Computer Vision

  2009cited by this paper
• Template Matching Techniques in Computer Vision: Theory and Practice

  2009cited by this paper
• Image Registration by Template Matching Using Normalized Cross-Correlation

  2009cited by this paper
• Robust Real-Time Pattern Matching Using Bayesian Sequential Hypothesis Testing

  2008cited by this paper
• Fast and robust template matching algorithm in noisy image

  2007cited by this paper
• Fast algorithm for robust template matching with M-estimators

  2003cited by this paper
• Template matching based object recognition with unknown geometric parameters

  2002cited by this paper
• Color-Based Probabilistic Tracking

  2002cited by this paper
• Real-time tracking of non-rigid objects using mean shift

  2000cited by this paper
• Neural detection of pipe signatures in ground penetrating radar images

  2000cited by this paper
• Automatic construction of 3D human face models based on 2D images

  1996cited by this paper

• No citing papers are available for this paper.