This book deals with image-processing problems that arise in the pro- cess of automating some aspects of well log analysis. Each problem is first described in log analysis terms - that is, what task is performed by a log analyst and how it is accomplished in manual processing. Then algorithms for automating each function are presented and their meanings from the point of view of log analysis and image processing are explained. The term image processing is understood here, in its broadest sense, as any processing of any images. I developed many of the algorithms presented in this book for par- ticular independent applications. Later, when I realized that they used some common techniques for analysis of logging curves, I applied these techniques in designing new algorithms. To present the algorithms here, I first formulate a minimization principle that has proved useful in a number of applications. Then I describe image-processing problems and their solutions based on this principle and some other common techniques. Finally, I describe al- ternative approaches. At first reading, readers may choose to skip the chapter describing the minimization principle and come back to it later when they have seen how the principle can be applied. This order of reading is further justified by the fact that the formulas that apply the general principle are different for each application, so their derivation is repeated each time independently.

Table of Contents:
1. An Overview.- Depth Matching.- Dipmeter Computation.- Dipmeter Postprocessing.- Detection of Dipmeter Patterns.- Log Blocking.- Computation of Primary and Shear Wave Velocity Logs.- Problems Not Yet Solved.- References.- For Further Reading.- 2. The Minimization Principle.- Initial Formulation.- Modified Formulation.- Discussion of the Principle.- Conclusion.- 3. Computing the Shear Wave Velocity Log: Curve Detection in Noisy Pictures.- The Sonic Tool and the Full Waveform Log.- Detection of Shear Wave Arrival.- Selection of Correct Log Values.- An Image-Processing Interpretation.- Log Despiking.- Conclusion.- References.- 4. Dipmeter Correlation: A Pattern-Matching Problem.- The Dipmeter Tool.- Manual Processing of Dipmeter Logs.- Computer Processing of Dipmeter Logs.- Activity.- Segmentation of Dipmeter Curves.- Correlation Using the Minimization Principle.- Correlation Using Syntactic Pattern Recognition.- Fixed-Interval Correlation.- Conclusion.- References.- 5. Depth Matching: A Pattern-Matching Problem.- Depth Errors.- Manual Depth Correction of Logs.- An Interactive Algorithm.- An Algorithm Based on the Minimization Principle.- A Physical Model.- Depth Matching of Logs with Varying Deflections - and Other Options.- Examples.- Conclusion.- References.- 6. Dipmeter Postprocessing: An Image-Sharpening Problem.- Testing Dipmeter Logs for Plausibility.- A Mathematical Model.- Applications of the Model.- Taking the Dip Angle into Account.- Conclusion.- References.- 7. Coloring a Computed Dipmeter Log: Feature Identification.- Definition of the Problem.- An Algorithm Based on the Minimization Principle.- An Algorithm Based on Statistical Pattern Recognition.- Conclusion.- References.- 8. Blocking: Curve Shape Analysis.- Definition and Applications of Blocking.- Zonation.- Determination of Representative Log Values.- Examples.- Extension of Zoning to Other Logs.- Conclusion.- References.- 9. Conclusion: Directions for Future Research.- Appendix A: An Algorithm to Implement the Minimization Principle.- Appendix B: An Algorithm to Implement the Modified Minimization Principle.- Appendix C: Forms of Activity Function.