Analysis And Application Of Seismic Texture Attributes Based On Gray Level Co-Occurrence Matrix

The content of this thesis is closely related to the current hot image processing and artificial intelligence.We hope to enrich the research content of geophysical exploration and find new methods for exploration through the research of gray symbiotic matrix texture analysis and self-organizing neural network algorithm in artificial neural network.In this thesis,the characteristics and analysis methods of texture are discussed from the definition of texture,then seismic texture attributes are introduced,and the characteristics and extraction methods of seismic texture attributes are discussed.This research focuses on the method of extracting the eigenvalues of seismic texture attributes based on GLCM,and discusses the formation of grayscale symbiotic matrix and its influencing factors in detail.Finally,combining the seismic data of the actual work area,the eigenvalues of seismic texture are used to predict and interpret the reservoir,and combined with the self-organizing neural network algorithm(SOM)to classify the seismic facies.By comparing with other seismic attributes,such as coherence and curvate,the seismic texture attributes show a more precise and accurate characterization in reservoir interpretation and seismic phase division.