In this presented paper, we demonstrate the use of Deep Learning techniques that precisely and automatically recognizes whether a targeted subsurface is a salt or not. The traditional procedures for determining salt bodies are analyzing the processed data and using image evaluation to make informed decisions. These methods are not very accurate and involve excessive human involvement, which leads to a low success rate of drilling, causing harm to the environment. A recent development in the domain of Artificial Intelligence has led to outstanding pattern recognition procedures. The paper aims to explore the capabilities of semantic segmentation for automatic seismic interpretation. We took a huge amount of imagery data to address the problem of subjective interpretation of seismic data in Geophysics. U-Net (Convolutional neural network architecture) was used to locate the salt boundaries and after rigorous training, the result obtained shows that we can segment regions that contain salt with high precision. Evaluating results displayed that using the U-Net model instead of other segmentation methods gives much more accurate prediction as it behaves like an encoder and decoder. Keywords: Data Analysis, Computer Vision, Deep Learning, Semantic Segmentation, Geophysics.
• Hardware: Processor: i3 ,i5 RAM: 4GB Hard disk: 16 GB • Software: operating System : Windws2000/XP/7/8/10 Anaconda,jupyter,spyder,flask Frontend :-python Backend:- MYSQL
₹10000 (INR)
NON IEEE -2021