Image Log Processing and Interpretation

A powerful borehole image analysis tool.

Textural image analysis using histogram and threshold techniques and image porosity calculation
 
Geolog provides extensive borehole image processing workflows and advanced analysis capabilities for both interactive interpretation and automatic detection of bedding dip, fracture analysis, classification of rock texture and quantitative image analyses.  Together, these functionalities provide structural geologists, sedimentologists and petrophysicists with a powerful tool for interpreting borehole image data, leading to improved reservoir characterization.

Geolog can process all types of LWD images, as well as wireline images from Schlumberger, Halliburton, Baker Atlas and Weatherford. Image processing capabilities include data QC, speed correction, image equalization, bad button correction, static and dynamic normalization, image filtering, eccentricity correction and image calibration.
 

Integration with core photos

The Core Photo Importer allows for quick and easy import of core images into the Geolog database. Core photos can be fully integrated into analyses (e.g. sand count analysis). Full 360° core scans can be orientated and used for dip calculation.
 

Borehole image analysis

Geolog enables both interactive and fully automatic image analysis. Images can be visualized and analyzed in user-defined displays alongside other log data, core photos, rose plots, stereonets, etc.  A variety of customizable color maps can be applied to the images to emphasize certain features. Bedding, natural and drilling induced fractures, breakouts, and other features can be manually picked and classified into user-defined categories. In addition, image facies can be interactively picked.

Advanced automatic dip computation algorithms utilizing the information from the whole image enable the identification of bedding features, from bed boundaries to fine-scaled lamination (e.g. cross-bedding).

Stereonets, rose diagrams and walkout plots facilitate the interpretation of the dip data. Image statistics, histograms and threshold techniques can be utilized for quantitative and textural image analyses. Porosity can be estimated from the image data.

Autotexture routines that extract textural parameters from the images using either Kohonen Self-Organizing Maps (Neural Network) or Multi Resolution Graph-based Clustering (MRGC) do not require prior assumptions about texture types. Therefore, results are not biased by the interpreter, and textural classes can be identified consistently for multiple wells.

*Geolog image log processing and analysis incorporates technology developed by Total.