Image analysis – Histogram processing – For segmenting an image
Patent
1998-04-15
2000-09-26
Tadayon, Bijan
Image analysis
Histogram processing
For segmenting an image
382109, G06K 900
Patent
active
061252034
DESCRIPTION:
BRIEF SUMMARY
FIELD OF THE INVENTION
The invention relates to a method for automatically determining the bedding in a formation, in particular a geological formation, from images of the walls of boreholes or developments of core samples of this formation. It more particularly concerns the generation of images which very precisely represent the bedding in this formation and the boundaries of this bedding.
BACKGROUND OF THE INVENTION
The electrical images of walls of boreholes in a geological formation, as obtained for example using the tools which are referred to by the references FMI (Full-bore Formation Micro Imager) and/or FMS (Formation Micro Scanner) and have been developed by the company SCHLUMBERGER, are of great benefit to the oil industry because of the wealth of information which they contain. These images are used almost exclusively by structure analysts for fine determination of the geometrical characteristics of the bedding and fracture planes in boreholes.
The FMS and FMI tools make it possible to acquire electrical images from measurements of the local electrical conductivity of the wall of the borehole and, in order to do this, have four articulated arms, each equipped with a pad in the case of the FMS tool, or a pad and a flap (accompanied pad) in the case of the FMI tool. The pads of the FMS tool each have 16 electrodes, and the pads and flaps of the FMI tool each have 24 electrodes. The pads and flaps of the said tools are held against the wall of the borehole using a mechanical system, throughout the acquisition of the electrical images.
An electrical image of the wall of a borehole is a view of the wall of the borehole and, when the borehole is opened to develop the wall on to a plane, has a horizontal axis x representing the azimuthal distribution of the electrodes of the pads and a vertical axis y along which the depth (position) of the tool in the borehole is defined. An FMS or FMI electrical image of the wall in a borehole is reconstructed by 4 or 8 pad images. Since there are 16 electrodes on each pad of the FMS tool and 24 electrodes on each pad and each flap of the FMI tool, a pad (pad or flap) image is therefore formed by 16 or 24 columns (one column per electrode) and several thousand rows, each pixel of the image having a size of about 2.5 mm.sup.2. The vertical sampling interval for the FMS or FMI tool is 2.5 mm, and the lateral sampling shift is 3.8 mm for the FMS tool and 2.5 mm for the FMI tool. More generally, each pad image may be considered, in the aforementioned (x,y) axis system, as consisting of p sample columns (p represents the number of electrodes of the pad), which are each arranged along the y axis and are spaced apart along the x axis by a distance corresponding to the lateral sampling shift, the samples of each column being separated from one another by a distance corresponding to the sampling interval of the tool.
The electrical image is analysed for planar heterogeneities and point heterogeneities. Planar heterogeneity includes the bedding planes as well as the fracture plane of the geological formation which intersect the bedding; the rest of the electrical image represents the point heterogeneities, that is to say the variations which are associated with variations in petrophysical parameters (porosity) or variations in sedimentological parameters (for example bioturbations).
In terms of image analysis, the planar heterogeneities present on the electrical images can be categorized by their conductivity relative to the background of the image, their sharpness (grey scale contrast), their organization (isolated or grouped by family), their frequency (high or low frequency according to direction and depth) and their visibility (visible on all the pad images or only on some of them).
Distinction can be made between four important plane types, which have particular characteristics on the electrical image, namely: background of the image and which are visible on all the pads, the said planes being identified as the bedding boundaries; organized by family and visible on al
REFERENCES:
patent: 3069654 (1962-12-01), Hough
patent: 4251773 (1981-02-01), Cailliau et al.
patent: 4303975 (1981-12-01), Hepp
patent: 5299128 (1994-03-01), Antoine Jean-Noel et al.
Deriche R: "Fast Algorithms for Low-Level Vision" IEEE Transactions on Pattern Analysis and Machine Intelligence vol. 12, No. 1, Jan. 1, 1990, pp. 78-87, XP000087783 cited in the application.
Keskes Naamen
Rabiller Philippe
Ye Shinju
Choobin M.
Elf Exploration Production
Tadayon Bijan
LandOfFree
Method for automatically determining stratification beds in a si does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for automatically determining stratification beds in a si, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for automatically determining stratification beds in a si will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2107052