Jef Caers, Xiaojin Tan and Pejman Tahmasebi Stanford University, USA

Slides:



Advertisements
Similar presentations
Feature Selection as Relevant Information Encoding Naftali Tishby School of Computer Science and Engineering The Hebrew University, Jerusalem, Israel NIPS.
Advertisements

StatisticalDesign&ModelsValidation. Introduction.
Object Recognition with Features Inspired by Visual Cortex T. Serre, L. Wolf, T. Poggio Presented by Andrew C. Gallagher Jan. 25, 2007.
Salvatore giorgi Ece 8110 machine learning 5/12/2014
Multivariate analysis of community structure data Colin Bates UBC Bamfield Marine Sciences Centre.
Multipoint Statistics to Generate Geologically Realistic Networks 1 Hiroshi Okabe supervised by Prof. Martin J Blunt Petroleum Engineering and Rock Mechanics.
K Means Clustering , Nearest Cluster and Gaussian Mixture
Face Recognition Method of OpenCV
A multi-scale, pattern-based approach to sequential simulation annual scrf meeting, may 2003 stanford university burc arpat ( coaching provided by jef.
Robust Multi-Kernel Classification of Uncertain and Imbalanced Data
Uncertainty Representation. Gaussian Distribution variance Standard deviation.
Content Based Image Clustering and Image Retrieval Using Multiple Instance Learning Using Multiple Instance Learning Xin Chen Advisor: Chengcui Zhang Department.
HMM-BASED PATTERN DETECTION. Outline  Markov Process  Hidden Markov Models Elements Basic Problems Evaluation Optimization Training Implementation 2-D.
Applied Geostatistics Geostatistical techniques are designed to evaluate the spatial structure of a variable, or the relationship between a value measured.
M-HinTS: Mimicking Humans in Texture Sorting Egon L. van den Broek Eva M. van Rikxoort.
Visual Querying By Color Perceptive Regions Alberto del Bimbo, M. Mugnaini, P. Pala, and F. Turco University of Florence, Italy Pattern Recognition, 1998.
Distance Measures Tan et al. From Chapter 2.
Pattern Recognition. Introduction. Definitions.. Recognition process. Recognition process relates input signal to the stored concepts about the object.
Relevance Feedback Content-Based Image Retrieval Using Query Distribution Estimation Based on Maximum Entropy Principle Irwin King and Zhong Jin The Chinese.
1 Variable Selection for Tailoring Treatment S.A. Murphy, L. Gunter & J. Zhu May 29, 2008.
The Context of Forest Management & Economics, Modeling Fundamentals Lecture 1 (03/30/2015)
Geo479/579: Geostatistics Ch13. Block Kriging. Block Estimate  Requirements An estimate of the average value of a variable within a prescribed local.
Clustering methods Course code: Pasi Fränti Speech & Image Processing Unit School of Computing University of Eastern Finland Joensuu,
IDL GUI for Digital Halftoning Final Project for SIMG-726 Computing For Imaging Science Changmeng Liu
Statistics Definition Methods of organizing and analyzing quantitative data Types Descriptive statistics –Central tendency, variability, etc. Inferential.
Texture analysis Team 5 Alexandra Bulgaru Justyna Jastrzebska Ulrich Leischner Vjekoslav Levacic Güray Tonguç.
Uncertainty Maps for Seismic Images through Geostatistical Model Randomization Lewis Li, Paul Sava, & Jef Caers 27 th SCRF Affiliates’ Meeting May 8-9.
On board processing - OPB Hot Spots 18 channels Ring Buffer Map Manager Request Interpreter Ground Station Relational Database (SQLite) Preprocess ing.
What is Modeling?. Simplifying Complex Phenomena v We live in a complex world v Most of the scientific relationships we study are very complex v Understanding.
A Two-level Pose Estimation Framework Using Majority Voting of Gabor Wavelets and Bunch Graph Analysis J. Wu, J. M. Pedersen, D. Putthividhya, D. Norgaard,
Computer Graphics and Image Processing (CIS-601).
Stochastic inverse modeling under realistic prior model constraints with multiple-point geostatistics Jef Caers Petroleum Engineering Department Stanford.
Céline Scheidt and Jef Caers SCRF Affiliate Meeting– April 30, 2009.
A NOVEL METHOD FOR COLOR FACE RECOGNITION USING KNN CLASSIFIER
SCRF 2012 Erosion in Surface-based Modeling Using Tank Experiment Data Siyao Xu, Andre Jung, Tapan Mukerji, Jef Caers.
BEYOND SLIDING WINDOW: Object Localization by Efficient Subwindow Search Christoph H. Lampert, Matthew B. Blaschko, and Thomas Hofmann.
Geo479/579: Geostatistics Ch12. Ordinary Kriging (2)
1 Cluster Analysis – 2 Approaches K-Means (traditional) Latent Class Analysis (new) by Jay Magidson, Statistical Innovations based in part on a presentation.
A new approach for the gamma tracking clustering by the deterministic annealing method François Didierjean IPHC, Strasbourg.
Stanford Center for Reservoir Forecasting
Maria Volkova, Mikhail Perepechkin, Evgeniy Kovalevskiy*
Requirements Basis Requirements of an Image Visualization System (IVS), to support the verification of the correct functioning of some components under.
mps-tk : A C++ toolkit for multiple-point simulation
Cheolkyun Jeong, Céline Scheidt, Jef Caers, and Tapan Mukerji
A strategy for managing uncertainty
Unsupervised Riemannian Clustering of Probability Density Functions
Basic machine learning background with Python scikit-learn
Metric Learning for Clustering
Addy Satija and Jef Caers Department of Energy Resources Engineering
SCRF 26th Annual Meeting May
Pejman Tahmasebi and Jef Caers
Clustering.
Scott Tan Boonping Lau Chun Hui Weng
S-GEMS-UQ: An Uncertainty Quantification Toolkit for SGEMS
Lecture 7 Implementing Spatial Analysis (Cont.)
Jef Caers, Céline Scheidt and Pejman Tahmasebi
Pejman Tahmasebi, Thomas Hossler and Jef Caers
Céline Scheidt, Jef Caers and Philippe Renard
Fast Pattern Simulation Using Multi‐Scale Search
Physics-based simulation for visual computing applications
Brief Review of Recognition + Context
Céline Scheidt, Pejman Tahmasebi and Jef Caers
Brent Lowry & Jef Caers Stanford University, USA
Siyao Xu Earth, Energy and Environmental Sciences (EEES)
Clustering Wei Wang.
SCRF High-order stochastic simulations and some effects on flow through heterogeneous media Roussos Dimitrakopoulos COSMO – Stochastic Mine Planning.
Energy Resources Engineering Department Stanford University, CA, USA
Siyao Xu, Andre Jung Tapan Mukerji and Jef Caers
Multivariate analysis of community structure data
Random Neural Network Texture Model
Presentation transcript:

Jef Caers, Xiaojin Tan and Pejman Tahmasebi Stanford University, USA Comparing multiple-point geostatistical algorithms using an analysis of distance Jef Caers, Xiaojin Tan and Pejman Tahmasebi Stanford University, USA

A simple question Which one is better ? Two algorithm aim to reproduce training image statistics Training image dispat ccsim Which one is better ?

Two fundamental variabilities Target statistics within realization variability “pattern reproduction” between realization variability “space of uncertainty” realization generated by a geostatistical algorithm

Comparing two geostatistical simulation algorithms Target statistics Algo 2 Algo 3 Algo 4 Definition of best: an algorithm that maximizes reproduction of statistics (within) while at the same time maximizes spatial uncertainty (between)

How to quantify this? Statistical science Computer Science x1 x2 x3 Form Matrix of realizations X Statistical science ANOVA Computer Science ANODI C: covariance D: Dot-product E: euclidean distance

Creating a distance multi-resolution view (34 x 34) Multi-resolution g=2 (51 x 51) Multi-resolution g=1 (101 x 101) Pyramid of one single realization

Creating a distance multiple-point histogram (MPH) Realization MPH But works only for binary, small 2D cases and small templates

Creating a distance cluster-based histogram of patterns (CHP) Pattern database Class-prototype Class 1 Class 2 Class 3 Class 4 Class 5 Class 6 Cluster patterns into classes based on a measure of similarity (distance)

Illustration case

Creating a distance cluster-based histogram of patterns (CHP)

Creating a distance Jensen-Shannon divergence Basic equation In this context multi-resolution algorithm

MDS visualizing distances

Multi-scale visualization

Ranking with ANODI Definition of best: an algorithm that maximizes reproduction of statistics (within) while at the same time maximizes spatial uncertainty (between) Use ratios

Back to illustration case

Ranking based on MPH algo m algo m algo m dispat ccsim sisim 1 1.15 0.38 * 0.33 dispat ccsim sisim 1 1.63 0.24 * 0.15 dispat ccsim sisim 1 0.70 1.58 * 2.20 algo k MPH approach: 1 : 0.70 : 0.46 (ccsim : dispat : sisim)

Ranking based on CHP dispat ccsim sisim 1 0.88 0.86 * 0.98 dispat ccsim sisim 1 1.31 0.43 * 0.33 dispat ccsim sisim 1 0.67 2.00 * 2.90 CHP approach: 1 : 0.67 : 0.35 (ccsim : dispat : sisim) MPH approach: 1 : 0.70 : 0.46 (ccsim : dispat : sisim)

Trade-off pattern reproduction for uncertainty in MPS

Trade-off Space of uncertainty (“between”) 1.65 : 1 : 1 (ns=10 : ns=50 : ns=200) Pattern reproduction (“within”) 2:75 : 1 : 1 (ns=10 : ns=50 : ns=200) Total (“between/within”): 0.60 : 1 : 1 (ns=10 : ns=50 : ns=200)

CHP works for 3D MPH does not Total (“between/within”): 0.60 : 1 (dispat : ccsim)

Conclusions Need: a repeatable quantitative comparison going beyond visual subjectivity Two fundamental variabilities Pattern reproduction (often the main focus) Space of uncertainty (often considered a by-product) What this presentation does not discuss which statistics to reproduce conditioning to data