Efficient Denoising of Multidimensional GPR Data Based on Fast Dictionary Learning
Denoising plays a fundamental role in ground penetrating radar (GPR) data processing and determines the effect of anomaly extraction, inversion imaging, and other subsequent processing. In recent years, the sparse dictionary representation method k-singular value decomposition (K-SVD) based on K-mea...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10438015/ |
| _version_ | 1797289883879866368 |
|---|---|
| author | Deshan Feng Li He Xun Wang Yougan Xiao Guoxing Huang Liqiong Cai Xiaoyong Tai |
| author_facet | Deshan Feng Li He Xun Wang Yougan Xiao Guoxing Huang Liqiong Cai Xiaoyong Tai |
| author_sort | Deshan Feng |
| collection | DOAJ |
| description | Denoising plays a fundamental role in ground penetrating radar (GPR) data processing and determines the effect of anomaly extraction, inversion imaging, and other subsequent processing. In recent years, the sparse dictionary representation method k-singular value decomposition (K-SVD) based on K-means, which can adaptively change the basis function according to the data, has become a hotspot in the field of image denoising and data reconstruction. Nevertheless, the SVD is a time-consuming calculation, especially unacceptable in multidimensional problems; we introduce a dictionary learning method based on the sequential generalized K-means (SGK), where the dictionary atoms are updated by the arithmetic average of several training signals instead of a great deal of SVD calculation in K-SVD. We establish a 3-D road simulation model and conduct finite-difference time-domain forward numerical simulation to acquire 3-D GPR data. Through three sets of experiments on 3-D numerical examples and 3-D field data, the results show that both dictionary learning algorithms can successfully remove random noise from GPR data even at a lower input signal-to-noise ratio. The clutter interference in the random medium forward data can be effectively eliminated simultaneously, and both denoising methods exhibit promising applications in 3-D field data. However, the SGK method solves the serious problem of computational efficiency to a certain extent. The computational acceleration ratio of SGK remains consistently above 7.5× that of the K-SVD algorithm in multigroup experiments, with only a marginal decline in denoising performance. |
| first_indexed | 2024-03-07T19:12:28Z |
| format | Article |
| id | doaj.art-9774a635ffee47a7be10b564f28f41b3 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-03-07T19:12:28Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-9774a635ffee47a7be10b564f28f41b32024-03-01T00:00:13ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352024-01-01175221523310.1109/JSTARS.2024.336639710438015Efficient Denoising of Multidimensional GPR Data Based on Fast Dictionary LearningDeshan Feng0https://orcid.org/0000-0002-6290-7797Li He1https://orcid.org/0009-0000-8303-8407Xun Wang2https://orcid.org/0000-0002-3039-4683Yougan Xiao3https://orcid.org/0009-0001-6558-5535Guoxing Huang4https://orcid.org/0009-0004-3498-1465Liqiong Cai5https://orcid.org/0009-0000-1805-6828Xiaoyong Tai6https://orcid.org/0009-0004-5991-5122School of Geosciences and Info-Physics, Central South University, Changsha, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha, ChinaSchool of Geosciences and Info-Physics, Central South University, Changsha, ChinaFuzhou City Construction Design and Research Institute, Fuzhou, ChinaFuzhou City Construction Design and Research Institute, Fuzhou, ChinaFuzhou City Construction Design and Research Institute, Fuzhou, ChinaFuzhou City Construction Design and Research Institute, Fuzhou, ChinaDenoising plays a fundamental role in ground penetrating radar (GPR) data processing and determines the effect of anomaly extraction, inversion imaging, and other subsequent processing. In recent years, the sparse dictionary representation method k-singular value decomposition (K-SVD) based on K-means, which can adaptively change the basis function according to the data, has become a hotspot in the field of image denoising and data reconstruction. Nevertheless, the SVD is a time-consuming calculation, especially unacceptable in multidimensional problems; we introduce a dictionary learning method based on the sequential generalized K-means (SGK), where the dictionary atoms are updated by the arithmetic average of several training signals instead of a great deal of SVD calculation in K-SVD. We establish a 3-D road simulation model and conduct finite-difference time-domain forward numerical simulation to acquire 3-D GPR data. Through three sets of experiments on 3-D numerical examples and 3-D field data, the results show that both dictionary learning algorithms can successfully remove random noise from GPR data even at a lower input signal-to-noise ratio. The clutter interference in the random medium forward data can be effectively eliminated simultaneously, and both denoising methods exhibit promising applications in 3-D field data. However, the SGK method solves the serious problem of computational efficiency to a certain extent. The computational acceleration ratio of SGK remains consistently above 7.5× that of the K-SVD algorithm in multigroup experiments, with only a marginal decline in denoising performance.https://ieeexplore.ieee.org/document/10438015/Dictionary learningground penetrating radar (GPR)K-singular value decomposition (K-SVD)noise attenuationsparse representationsequential generalized K-means (SGK) |
| spellingShingle | Deshan Feng Li He Xun Wang Yougan Xiao Guoxing Huang Liqiong Cai Xiaoyong Tai Efficient Denoising of Multidimensional GPR Data Based on Fast Dictionary Learning IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Dictionary learning ground penetrating radar (GPR) K-singular value decomposition (K-SVD) noise attenuation sparse representation sequential generalized K-means (SGK) |
| title | Efficient Denoising of Multidimensional GPR Data Based on Fast Dictionary Learning |
| title_full | Efficient Denoising of Multidimensional GPR Data Based on Fast Dictionary Learning |
| title_fullStr | Efficient Denoising of Multidimensional GPR Data Based on Fast Dictionary Learning |
| title_full_unstemmed | Efficient Denoising of Multidimensional GPR Data Based on Fast Dictionary Learning |
| title_short | Efficient Denoising of Multidimensional GPR Data Based on Fast Dictionary Learning |
| title_sort | efficient denoising of multidimensional gpr data based on fast dictionary learning |
| topic | Dictionary learning ground penetrating radar (GPR) K-singular value decomposition (K-SVD) noise attenuation sparse representation sequential generalized K-means (SGK) |
| url | https://ieeexplore.ieee.org/document/10438015/ |
| work_keys_str_mv | AT deshanfeng efficientdenoisingofmultidimensionalgprdatabasedonfastdictionarylearning AT lihe efficientdenoisingofmultidimensionalgprdatabasedonfastdictionarylearning AT xunwang efficientdenoisingofmultidimensionalgprdatabasedonfastdictionarylearning AT youganxiao efficientdenoisingofmultidimensionalgprdatabasedonfastdictionarylearning AT guoxinghuang efficientdenoisingofmultidimensionalgprdatabasedonfastdictionarylearning AT liqiongcai efficientdenoisingofmultidimensionalgprdatabasedonfastdictionarylearning AT xiaoyongtai efficientdenoisingofmultidimensionalgprdatabasedonfastdictionarylearning |