CNN Prediction Enhancement by Post-Processing for Hydrocarbon Detection in Seismic Images

Souza, José Fabrício Lima de; Santana, Gabriel L.; Batista, Leonardo Vidal; Oliveira, Gustavo P.; Roemers-Oliveira, Eduardo; Santos, M.D.

doi:10.1109/access.2020.3005916

Cited by 10 publications

(3 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Where C is the output characteristic graph of the convolution layer, X is the input data and the price curve established by us in the input layer; f(.) is the nonlinear activation function, ⊗ is the convolution operation, W is the weight vector of the convolution kernel, and b is the bias term [4].…”

Section: Cnn Prediction Modelmentioning

confidence: 99%

A Comprehensive Trading Strategy Model for Forecasting and Scheme-Planning

Zhang¹,

Wu²,

Li³

2022

Proceedings of the 2022 3rd International Conference on E-Commerce and Internet Technology (ECIT 2022)

View full text Add to dashboard Cite

Due to the complexity and variability of the financial market, the current model can not fully cover all aspects of investment trends, and there is room for improvement in forecasting accuracy and decision-making. Based on this, we propose a CNN-LSTM prediction method combining RMSE loss, and use DDPG algorithm based on Actor-Critic framework to make decision. This method innovatively combines RMSE loss functions of two models. In order to prove the rationality of this model, we choose several models to compare with it, and finally come to the conclusion that our model has a good feasibility and universality in prediction and decision-making.

show abstract

Section: Cnn Prediction Modelmentioning

confidence: 99%

A Comprehensive Trading Strategy Model for Forecasting and Scheme-Planning

Zhang¹,

Wu²,

Li³

2022

Proceedings of the 2022 3rd International Conference on E-Commerce and Internet Technology (ECIT 2022)

View full text Add to dashboard Cite

show abstract

“…Seismic image processing refers to the set of techniques and algorithms used to extract useful information from seismic data acquired during oil and gas exploration [5]. Seismic data is collected using specialized sensors and equipment that generate sound waves and record the reflections of these waves as they travel through earth's subsurface.…”

Section: Introductionmentioning

confidence: 99%

Semantic Segmentation and Augmentation of salt in seismic images using Deep Learning

2023

ijfans

View full text Add to dashboard Cite

Salt segmentation refers to the process of separating salt from an image or a video stream.This process is often used in computer vision which is a python library application to extract specific features or objects from an image or video and can be used for various purposes, such as object recognition, tracking, and classification. The salt segmentation process typically involves segmenting the image into regions based on colour, texture, or other visual features, and then applying a threshold to separate the salt regions from the rest of the image. This can be done with various deep learning techniques like Unet, ResNet-18 and ResNet-34, VGG16 and InceptionV3, and DeeplabV3Plus. Through the ensemble approach of these models, we can achieve an accurate prediction of the model.For locating the salt zone in the seismic pictures, two models are used. The main model is an amalgam of ResNet-18 and ResNet-34 with UNET. By combining Ensemble UNET with ResNet-34, VGG16, Inceptionv3, and DeeplabV3Plus, the secondary model produces segmentation results. Besides, Data augmentation is performed to generate new training data from existing data by applying a set of transactions or modifications to the original data. This technique is commonly used to increase the size and diversity of the training data which can improve the accuracy and robustness of the trained models.

show abstract

“…The application of CNN in the petroleum industry has been undertaken to detect hydrocarbon zones in seismic images where 2D-CNN is used in the image segmentation process, and the model achieved more than 80% accuracy [26]. The developed model used 2D-CNN because the input of seismic images is in two dimensions (2D).…”

Section: Introductionmentioning

confidence: 99%

One-Dimensional Convolutional Neural Network with Adaptive Moment Estimation for Modelling of the Sand Retention Test

et al. 2021

View full text Add to dashboard Cite

Stand-alone screens (SASs) are active sand control methods where compatible screens and slot sizes are selected through the sand retention test (SRT) to filter an unacceptable amount of sand produced from oil and gas wells. SRTs have been modelled in the laboratory using computer simulation to replicate experimental conditions and ensure that the selected screens are suitable for selected reservoirs. However, the SRT experimental setups and result analyses are not standardized. A few changes made to the experimental setup can cause a huge variation in results, leading to different plugging performance and sand retention analysis. Besides, conducting many laboratory experiments is expensive and time-consuming. Since the application of CNN in the petroleum industry attained promising results for both classification and regression problems, this method is proposed on SRT to reduce the time, cost, and effort to run the laboratory test by predicting the plugging performance and sand production. The application of deep learning has yet to be imposed in SRT. Therefore, in this study, a deep learning model using a one-dimensional convolutional neural network (1D-CNN) with adaptive moment estimation is developed to model the SRT with the aim of classifying plugging sign (screen plug, the screen does not plug) as well as to predict sand production and retained permeability using a varying sand distribution, SAS, screen slot size, and sand concentration as inputs. The performance of the proposed 1D-CNN model for the slurry test shows that the prediction of retained permeability and the classification of plugging sign achieved robust accuracy with more than a 90% value of R2, while the prediction of sand production achieved 77% accuracy. In addition, the model for the sand pack test achieved 84% accuracy in predicting sand production. For comparative model performance, gradient boosting (GB), K-nearest neighbor (KNN), random forest (RF), and support vector machine (SVM) were also modelled on the same datasets. The results showed that the proposed 1D-CNN model outperforms the other four machine learning models for both SRT tests in terms of prediction accuracy.

show abstract

CNN Prediction Enhancement by Post-Processing for Hydrocarbon Detection in Seismic Images

Cited by 10 publications

References 21 publications

A Comprehensive Trading Strategy Model for Forecasting and Scheme-Planning

A Comprehensive Trading Strategy Model for Forecasting and Scheme-Planning

Semantic Segmentation and Augmentation of salt in seismic images using Deep Learning

One-Dimensional Convolutional Neural Network with Adaptive Moment Estimation for Modelling of the Sand Retention Test

Contact Info

Product

Resources

About