Zhen Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zhen Wang has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Ghassan AlRegib
Georgia Institute of Technology
7 shared publications - Mohamed Deriche
Artificial Intelligence Research Centre, Ajman University, Ajman, United Arab Emirates
4 shared publications - Muhammad Amir Shafiq
Georgia Institute of Technology, School of Electrical and Computer Engineering, Center for Energy and Geo Processing (CeGP), Atlanta, Georgia 30308, USA.
3 shared publications - Asjad Amin
Department of Information and Communication Engineering, The Islamia University of Bahawalpur (IUB), Bahawalpur 63100, Southern Punjab, Pakistan
2 shared publications - Haibin Di
SLB, Houston, Texas, USA.;;;
2 shared publications - Yazeed Alaudah
Center for Energy and Geo Processing (CeGP) Minerals (KFUPM) Georgia Institute of Technology King Fahd University of Petroleum
2 shared publications
Latest Publications
- 1 August 2019
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 7 (3) , SE251-SE267
Improving seismic fault detection by super-attribute-based classification
Fault interpretation is one of the routine processes used for subsurface structure mapping and reservoir characterization from 3D seismic data. Various techniques have been developed for computer-aided fault imaging in the past few decades; for example, the conventional methods of edge detection, curvature analysis, red-green-blue rendering, and the popular machine-learning methods such as the support vector machine (SVM), the multilayer perceptron (MLP), and the convolutional neural network (CNN). However, most of the conventional methods are performed at the sample level with the local reflection pattern ignored and are correspondingly sensitive to the coherent noises/processing artifacts present in seismic signals. The CNN has proven its efficiency in utilizing such local seismic patterns to assist seismic fault interpretation, but it is quite computationally intensive and often demands higher hardware configuration (e.g., graphics processing unit). We have developed an innovative scheme for improving seismic fault detection by integrating the computationally efficient SVM/MLP classification algorithms with local seismic attribute patterns, here denoted as the super-attribute-based classification. Its added values are verified through applications to the 3D seismic data set over the Great South Basin (GSB) in New Zealand, where the subsurface structure is dominated by polygonal faults. A good match is observed between the original seismic images and the detected lineaments, and the generated fault volume is tested usable to the existing advanced fault interpretation tools/modules, such as seeded picking and automatic extraction. It is concluded that the improved performance of our scheme results from its two components. First, the SVM/MLP classifier is computationally efficient in parsing as many seismic attributes as specified by interpreters and maximizing the contributions from each attribute, which helps minimize the negative effects from using a less useful or “wrong” attribute. Second, the use of super attributes incorporates local seismic patterns into training a fault classifier, which helps exclude the random noises and/or artifacts of distinct reflection patterns.
- 1 August 2019
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 7 (3) , SE251-SE267
Improving seismic fault detection by super-attribute-based classification
Fault interpretation is one of the routine processes used for subsurface structure mapping and reservoir characterization from 3D seismic data. Various techniques have been developed for computer-aided fault imaging in the past few decades; for example, the conventional methods of edge detection, curvature analysis, red-green-blue rendering, and the popular machine-learning methods such as the support vector machine (SVM), the multilayer perceptron (MLP), and the convolutional neural network (CNN). However, most of the conventional methods are performed at the sample level with the local reflection pattern ignored and are correspondingly sensitive to the coherent noises/processing artifacts present in seismic signals. The CNN has proven its efficiency in utilizing such local seismic patterns to assist seismic fault interpretation, but it is quite computationally intensive and often demands higher hardware configuration (e.g., graphics processing unit). We have developed an innovative scheme for improving seismic fault detection by integrating the computationally efficient SVM/MLP classification algorithms with local seismic attribute patterns, here denoted as the super-attribute-based classification. Its added values are verified through applications to the 3D seismic data set over the Great South Basin (GSB) in New Zealand, where the subsurface structure is dominated by polygonal faults. A good match is observed between the original seismic images and the detected lineaments, and the generated fault volume is tested usable to the existing advanced fault interpretation tools/modules, such as seeded picking and automatic extraction. It is concluded that the improved performance of our scheme results from its two components. First, the SVM/MLP classifier is computationally efficient in parsing as many seismic attributes as specified by interpreters and maximizing the contributions from each attribute, which helps minimize the negative effects from using a less useful or “wrong” attribute. Second, the use of super attributes incorporates local seismic patterns into training a fault classifier, which helps exclude the random noises and/or artifacts of distinct reflection patterns.
- 1 November 2018
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 6 (4) , T1055-T1066
A comparative study of texture attributes for characterizing subsurface structures in seismic volumes
We have explored how to computationally characterize subsurface geologic structures presented in seismic volumes using texture attributes. For this purpose, we conduct a comparative study of typical texture attributes presented in the image processing literature. We focus on spatial attributes in this study and examine them in a new application for seismic interpretation, i.e., seismic volume labeling. For this application, a data volume is automatically segmented into various structures, each assigned with its corresponding label. If the labels are assigned with reasonable accuracy, such volume labeling will help initiate an interpretation process in a more effective manner. Our investigation proves the feasibility of accomplishing this task using texture attributes. We also identify the advantages and disadvantages associated with each attribute.
- 1 November 2018
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 6 (4) , T1055-T1066
A comparative study of texture attributes for characterizing subsurface structures in seismic volumes
We have explored how to computationally characterize subsurface geologic structures presented in seismic volumes using texture attributes. For this purpose, we conduct a comparative study of typical texture attributes presented in the image processing literature. We focus on spatial attributes in this study and examine them in a new application for seismic interpretation, i.e., seismic volume labeling. For this application, a data volume is automatically segmented into various structures, each assigned with its corresponding label. If the labels are assigned with reasonable accuracy, such volume labeling will help initiate an interpretation process in a more effective manner. Our investigation proves the feasibility of accomplishing this task using texture attributes. We also identify the advantages and disadvantages associated with each attribute.
- 1 June 2018
- Published by Society of Exploration Geophysicists in The Leading Edge
- Vol. 37 (6) , 451-461
Successful leveraging of image processing and machine learning in seismic structural interpretation: A review
As a process that identifies geologic structures of interest such as faults, salt domes, or elements of petroleum systems in general, seismic structural interpretation depends heavily on the domain knowledge and experience of interpreters as well as visual cues of geologic structures, such as texture and geometry. With the dramatic increase in size of seismic data acquired for hydrocarbon exploration, structural interpretation has become more time consuming and labor intensive. By treating seismic data as images rather than signal traces, researchers have been able to utilize advanced image-processing and machine-learning techniques to assist interpretation directly. In this paper, we mainly focus on the interpretation of two important geologic structures, faults and salt domes, and summarize interpretation workflows based on typical or advanced image-processing and machine-learning algorithms. In recent years, increasing computational power and the massive amount of available data have led to the rise of deep learning. Deep-learning models that simulate the human brain's biological neural networks can achieve state-of-the-art accuracy and even exceed human-level performance on numerous applications. The convolutional neural network — a form of deep-learning model that is effective in analyzing visual imagery — has been applied in fault and salt dome interpretation. At the end of this review, we provide insight and discussion on the future of structural interpretation.
- 1 June 2018
- Published by Society of Exploration Geophysicists in The Leading Edge
- Vol. 37 (6) , 451-461
Successful leveraging of image processing and machine learning in seismic structural interpretation: A review
As a process that identifies geologic structures of interest such as faults, salt domes, or elements of petroleum systems in general, seismic structural interpretation depends heavily on the domain knowledge and experience of interpreters as well as visual cues of geologic structures, such as texture and geometry. With the dramatic increase in size of seismic data acquired for hydrocarbon exploration, structural interpretation has become more time consuming and labor intensive. By treating seismic data as images rather than signal traces, researchers have been able to utilize advanced image-processing and machine-learning techniques to assist interpretation directly. In this paper, we mainly focus on the interpretation of two important geologic structures, faults and salt domes, and summarize interpretation workflows based on typical or advanced image-processing and machine-learning algorithms. In recent years, increasing computational power and the massive amount of available data have led to the rise of deep learning. Deep-learning models that simulate the human brain's biological neural networks can achieve state-of-the-art accuracy and even exceed human-level performance on numerous applications. The convolutional neural network — a form of deep-learning model that is effective in analyzing visual imagery — has been applied in fault and salt dome interpretation. At the end of this review, we provide insight and discussion on the future of structural interpretation.
- 12 March 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Magazine
- Vol. 35 (2) , 82-98
Subsurface Structure Analysis Using Computational Interpretation and Learning: A Visual Signal Processing Perspective
Understanding Earth's subsurface structures has been and continues to be an essential component of various applications such as environmental monitoring, carbon sequestration, and oil and gas exploration. By viewing the seismic volumes that are generated through the processing of recorded seismic traces, researchers were able to learn from applying advanced image processing and computer vision algorithms to effectively analyze and understand Earth's subsurface structures. In this article, we first summarize the recent advances in this direction that relied heavily on the fields of image processing and computer vision. Second, we discuss the challenges in seismic interpretation and provide insights and some directions to address such challenges using emerging machine-learning algorithms.
- 12 March 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Magazine
- Vol. 35 (2) , 82-98
Subsurface Structure Analysis Using Computational Interpretation and Learning: A Visual Signal Processing Perspective
Understanding Earth's subsurface structures has been and continues to be an essential component of various applications such as environmental monitoring, carbon sequestration, and oil and gas exploration. By viewing the seismic volumes that are generated through the processing of recorded seismic traces, researchers were able to learn from applying advanced image processing and computer vision algorithms to effectively analyze and understand Earth's subsurface structures. In this article, we first summarize the recent advances in this direction that relied heavily on the fields of image processing and computer vision. Second, we discuss the challenges in seismic interpretation and provide insights and some directions to address such challenges using emerging machine-learning algorithms.
- 31 August 2017
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 5 (3) , SJ61-SJ79
Automated salt-dome detection using an attribute ranking framework with a dictionary-based classifier
We have developed a dictionary-based classification approach for salt-dome detection within migrated seismic volumes. The proposed workflow uses seismic attributes derived from the gray-level co-occurrence matrix, Gabor filter, and higher order singular-value decomposition to effectively learn and detect the salt bodies. We use an information theoretic framework to rank the seismic attributes as per their salt-dome classification performance. Based on this ranking, we select the top K attributes for dictionary training, testing, and classification. To improve the accuracy of the detected salt bodies and make the proposed workflow robust to different data sets, we introduce a refining step that uses edge strength and energy values to detect the shape of the salt-dome boundary within the classified patches. The optimal set of attributes and the refining step ensure that the proposed workflow yields good results for detecting salt-dome boundaries even in the presence of weak seismic reflections. We use the seismic data from the Netherlands offshore F3 block (North Sea) to demonstrate the effectiveness of the proposed workflow in detecting salt bodies. Using subjective and objective evaluation metrics, we compare the results of the proposed workflow with existing gradient-, texture-, and patch-based classification methods. The experimental results show that the proposed workflow outperforms existing salt-dome delineation techniques in terms of accuracy and precision.
- 31 August 2017
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 5 (3) , SJ61-SJ79
Automated salt-dome detection using an attribute ranking framework with a dictionary-based classifier
We have developed a dictionary-based classification approach for salt-dome detection within migrated seismic volumes. The proposed workflow uses seismic attributes derived from the gray-level co-occurrence matrix, Gabor filter, and higher order singular-value decomposition to effectively learn and detect the salt bodies. We use an information theoretic framework to rank the seismic attributes as per their salt-dome classification performance. Based on this ranking, we select the top K attributes for dictionary training, testing, and classification. To improve the accuracy of the detected salt bodies and make the proposed workflow robust to different data sets, we introduce a refining step that uses edge strength and energy values to detect the shape of the salt-dome boundary within the classified patches. The optimal set of attributes and the refining step ensure that the proposed workflow yields good results for detecting salt-dome boundaries even in the presence of weak seismic reflections. We use the seismic data from the Netherlands offshore F3 block (North Sea) to demonstrate the effectiveness of the proposed workflow in detecting salt bodies. Using subjective and objective evaluation metrics, we compare the results of the proposed workflow with existing gradient-, texture-, and patch-based classification methods. The experimental results show that the proposed workflow outperforms existing salt-dome delineation techniques in terms of accuracy and precision.
- 31 August 2017
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 5 (3) , SJ1-SJ19
A texture-based interpretation workflow with application to delineating salt domes
We propose a texture-based interpretation workflow and apply it to delineate salt domes in 3D migrated seismic volumes. First, we compute an attribute map using a novel seismic attribute, 3D gradient of textures (3D-GoT), which measures the dissimilarity between neighboring cubes around each voxel in a seismic volume across the time or depth, crossline, and inline directions. To evaluate the texture dissimilarity, we introduce five 3D perceptual and nonperceptual dissimilarity functions. Second, we apply a global threshold on the 3D-GoT volume to yield a binary volume and demonstrate its effects on salt-dome delineation using objective evaluation measures such as receiver operating characteristic curves and the areas under the curves. Third, with an initial seed point selected inside the binary volume, we use a 3D region growing method to capture a salt body. For an automated 3D region growing, we adopt a tensor-based automatic seed point selection method. Finally, we apply morphological postprocessing to delineate the salt dome within the seismic volume. Furthermore, we also develop an objective evaluation measure based on the curvature and shape to compute the similarity between detected salt-dome boundaries and the reference interpreted by the geophysicist. Experimental results on a real data set from the North Sea show that the proposed method outperforms the state-of-the-art methods for salt-dome delineation.
- 31 August 2017
- Published by Society of Exploration Geophysicists in Interpretation
- Vol. 5 (3) , SJ1-SJ19
A texture-based interpretation workflow with application to delineating salt domes
We propose a texture-based interpretation workflow and apply it to delineate salt domes in 3D migrated seismic volumes. First, we compute an attribute map using a novel seismic attribute, 3D gradient of textures (3D-GoT), which measures the dissimilarity between neighboring cubes around each voxel in a seismic volume across the time or depth, crossline, and inline directions. To evaluate the texture dissimilarity, we introduce five 3D perceptual and nonperceptual dissimilarity functions. Second, we apply a global threshold on the 3D-GoT volume to yield a binary volume and demonstrate its effects on salt-dome delineation using objective evaluation measures such as receiver operating characteristic curves and the areas under the curves. Third, with an initial seed point selected inside the binary volume, we use a 3D region growing method to capture a salt body. For an automated 3D region growing, we adopt a tensor-based automatic seed point selection method. Finally, we apply morphological postprocessing to delineate the salt dome within the seismic volume. Furthermore, we also develop an objective evaluation measure based on the curvature and shape to compute the similarity between detected salt-dome boundaries and the reference interpreted by the geophysicist. Experimental results on a real data set from the North Sea show that the proposed method outperforms the state-of-the-art methods for salt-dome delineation.
- 23 September 2015
- Published by Society of Exploration Geophysicists in Geophysics
- Vol. 80 (6) , WD101-WD116
Noise-robust detection and tracking of salt domes in postmigrated volumes using texture, tensors, and subspace learning
The identification of salt-dome boundaries in migrated seismic data volumes is important for locating petroleum reservoirs. The presence of noise in the data makes computer-aided salt-dome interpretation even more challenging. We have developed noise-robust algorithms that could label boundaries of salt domes effectively and efficiently. Our research is twofold. First, we used a texture-based gradient to accomplish salt-dome detection. We found that by using a dissimilarity measure based on the 2D discrete Fourier transform, the algorithm was capable of efficiently detecting salt-dome boundaries with accuracy. At the same time, our analysis determined that the proposed algorithm was robust to noise. Once the detection is performed for an initial 2D seismic section, we track the initial boundaries through the data volume to accomplish an efficient labeling process by avoiding the parameter tuning that would have been necessary if detection had been performed for every seismic section. The tracking process involves a tensor-based subspace learning process, in which we built texture tensors using patches from different seismic sections. To accommodate noise components with various levels in a texture tensor, we used noise-adjusted principal component analysis, so that principal components corresponding to greater signal-to-noise-ratio values might be selected for tracking. We validated our detection and tracking algorithms through experiments using seismic data sets acquired from the Netherlands offshore F3 block in the North Sea with very encouraging results.
- 23 September 2015
- Published by Society of Exploration Geophysicists in Geophysics
- Vol. 80 (6) , WD101-WD116
Noise-robust detection and tracking of salt domes in postmigrated volumes using texture, tensors, and subspace learning
The identification of salt-dome boundaries in migrated seismic data volumes is important for locating petroleum reservoirs. The presence of noise in the data makes computer-aided salt-dome interpretation even more challenging. We have developed noise-robust algorithms that could label boundaries of salt domes effectively and efficiently. Our research is twofold. First, we used a texture-based gradient to accomplish salt-dome detection. We found that by using a dissimilarity measure based on the 2D discrete Fourier transform, the algorithm was capable of efficiently detecting salt-dome boundaries with accuracy. At the same time, our analysis determined that the proposed algorithm was robust to noise. Once the detection is performed for an initial 2D seismic section, we track the initial boundaries through the data volume to accomplish an efficient labeling process by avoiding the parameter tuning that would have been necessary if detection had been performed for every seismic section. The tracking process involves a tensor-based subspace learning process, in which we built texture tensors using patches from different seismic sections. To accommodate noise components with various levels in a texture tensor, we used noise-adjusted principal component analysis, so that principal components corresponding to greater signal-to-noise-ratio values might be selected for tracking. We validated our detection and tracking algorithms through experiments using seismic data sets acquired from the Netherlands offshore F3 block in the North Sea with very encouraging results.