Kou-Yuan Huang

Image processing of seismograms: (A) Hough transformation for the detection of seismic patterns; (B) thinning processing in the seismogram

Abstract Hough transformation is used to detect the seismic patterns on the seismogram. The system of Hough technique includes envelope generation, thresholding, Hough transformation, and parameter determination. Visual inspection method, local peak detection method, and clustering algorithm are used to determine the parameters. Thinning algorithms are used for the skeletonization of the seismic reflectors in the seismogram. The thinning algorithms used are fast parallel thinning algorithm and 2D thinning algorithm preserving 8- and 4-point neighbor connectivities.

A neural network method for prediction of 2006 World Cup Football Game

A neural network method is adopted to predict the football games winning rate of two teams according to their previous stages official statistical data of 2006 World Cup Football Game. The adopted prediction model is based on multi-layer perceptron (MLP) with back propagation learning rule. The input data are transformed to the relative ratios between two teams of each game. New training samples are added to the training samples at the previous stages. By way of experimental results, the determined neural network architecture for MLP is 8 inputs, 11 hidden nodes, and 1 output (8-11-1). The learning rate and momentum coefficient are sequentially determined by experiments as well. Based on the adopted MLP prediction method, the prediction accuracy can achieve 76.9% if the draw games are excluded.

Neural network for robust recognition of seismic patterns

The multilayer perceptron is trained as a classifier and is applied to the recognition of seismic patterns. The principle of training the multilayer perceptron is described. Three classes of seismic patterns are analyzed in the experiment. Bright spot, pinch-out, and horizontal reflection patterns. Seven moments that are invariant to translation, rotation, and scale, are employed for feature generation of each seismic pattern. The training set includes noise-free, low-noise, and misclassified seismic patterns. The test set includes seismic patterns with various noise levels. The multilayer perceptron is initially trained with the training set of noise-free and low-noise seismic patterns. After convergence of the training, the network is applied to the classification of the test set of noisy seismic patterns. Some misclassified patterns with higher noise level are added to the training set for retraining. From experiments, the multilayer perceptron is shown to have the capability of robust recognition of seismic patterns.

Neural networks for seismic principal components analysis

The neural network, using an unsupervised generalized Hebbian algorithm (GHA), is adopted to find the principal eigenvectors of a covariance matrix in different kinds of seismograms. The authors have shown that the extensive computer results of the principal components analysis (PCA) using the neural net of GHA can extract the information of seismic reflection layers and uniform neighboring traces. The analyzed seismic data are the seismic traces with 20-, 25-, and 30-Hz Ricker wavelets, the fault, the reflection and diffraction patterns after normal moveout (NMO) correction, the bright spot pattern, and the real seismogram at Mississippi Canyon. The properties of high amplitude, low frequency, and polarity reversal can be shown from the projections on the principal eigenvectors. For PCA, a theorem is proposed, which states that adding an extra point along the direction of the existing eigenvector can enhance that eigenvector. The theorem is applied to the interpretation of a fault seismogram and the uniform property of other seismograms. The PCA also provides a significant seismic data compression.

Syntactic pattern recognition and Hough transformation for reconstruction of seismic patterns

Hierarchical syntactic pattern recognition and the Hough transformation are proposed for automatic recognition and reconstruction of seismic patterns in seismograms. In the first step, the patterns are hierarchically decomposed or recognized into single patterns, straight‐line patterns, or hyperbolic patterns, using syntactic pattern recognition. In the second step, the Hough transformation technique is used for reconstruction, pattern by pattern. The system of syntactic seismic pattern recognition includes envelope generation, a linking process in the seismogram, segmentation, primitive recognition, grammatical inference, and syntax analysis. The seismic patterns are automatically recognized and reconstructed.

Simulated Annealing for Pattern Detection and Seismic Applications

Simulated annealing (SA) is adopted to detect the parameters of circles, ellipses, hyperbolas, and to treat lines as asymptotes of hyperbola in image. Also, the algorithm is applied to seismic pattern detection. We use the general equation for ellipses and hyperbolas in detection and define the distance from a point to a pattern such that the detection becomes feasible. The system error between N points and K patterns is defined. The proposed simulated annealing parameter detection system has the capability of searching a set of parameter vectors with global minimal error with respect to the input data. Experiments on the detection of circles, ellipses, hyperbolas, and lines in images are quite successful. The detection system is also applied to detect the line pattern of direct wave and the hyperbolic pattern of reflection wave in the simulated and real one-shot seismogram. The results can improve seismic interpretations and further seismic data processing.

Very fast simulated annealing for pattern detection and seismic applications

We use three global optimization methods in the pattern parameter detection system, including simulated annealing (SA), fast simulated annealing (FSA) and very fast simulated annealing (VFSA). The sequential pattern parameter detection system can detect three types of patterns that include the lines, hyperbolas and ellipses in image. We use steps in the parameter detection for reducing the computation and getting fast convergence. This system has the capability of searching pattern parameter vectors with global minimal distance between the patterns and the image data. After the system is successful in image pattern detection, we apply it to detect the parameters of the hyperbolic patterns on real one-shot seismogram and seismic common depth point (CDP) gather data. This procedure provides an automatic velocity analysis method and improves the seismic interpretation and further seismic data processing.

Simulated Annealing for Pattern Detection and Seismic Application

Simulated annealing algorithm is adopted to detect the parameters of lines, circles, ellipses, and hyperbolic patterns. We define the distance from a point to a pattern such that the detection becomes feasible, especially in hyperbola. The proposed simulated annealing parameter detection system has the capability to find a set of parameter vectors with global minimal error to the input data. Using average minimum distance, we propose a method to determine the number of patterns automatically. Experiments on the detection of lines, circles, ellipses, and hyperbolas in images are quite successful. The detection system is also applied to detect the line pattern of direct wave and the hyperbolic pattern of reflection wave in the simulated one-shot seismogram. The results are good and can improve seismic interpretations and further seismic data processing.

Branch and bound search for automatic linking process of seismic horizons

Abstract Pattern growing technique is proposed in the automatic linking of seismic horizons. The branch and bound search algorithm and the distance calculation of Gram approximate functions are used to find the best way to correlate the seismic reflectors. The experimental results in simulated seismograms and in real seismic data are quite good.

Syntactic Pattern Recognition for Seismic Oil Exploration

Introduction to Syntactic Pattern Recognition Introduction to Formal Languages and Automata Error-Correcting Finite-State Automaton for Recognition of Ricker Wavelets Attributed Grammar and Error-Correcting Earleys Parsing Attributed Grammar and Match Primitive Measure (MPM) for Recognition of Seismic Wavelets String Distance and Likelihood Ratio Test for Detection of Candidate Bright Spot Tree Grammar and Automaton for Seismic Pattern Recognition A Hierarchical Recognition System of Seismic Patterns and Future Study.