Seismic Data and Fault Detection Labels for Machine Learning Analysis
Description
This dataset contains seismic data and corresponding fault detection labels used for the research article titled "Fault Detection in Seismic Data Using Machine Learning Models." The seismic data includes preprocessed seismic traces in MATLAB format, while the fault labels are provided as binary matrices representing fault locations. The dataset was generated and analyzed using various seismic attributes, including similarity, coherency, and instantaneous phase, to enhance the accuracy of fault detection. The data can be used for benchmarking fault detection algorithms, developing machine learning models, and validating geophysical interpretation methods. Content Overview: Seismic Data: Processed seismic sections in MATLAB format. Fault Labels: Binary images indicating fault locations. Attribute Maps: Extracted seismic attributes (e.g., Coherency, Similarity) in .mat format. Metadata: Information about data preprocessing and attribute extraction. Data Usage: Researchers are encouraged to use this dataset for further studies in fault detection, seismic interpretation, and machine learning applications. Proper citation of the related article is appreciated. Contact: For inquiries regarding the dataset, please contact the corresponding author of the article.