Caracterização de fraturas em imagens de amplitude acústica utilizando morfologia matemática

Abstract

Fractures analysis is of particular interest in the characterization of carbonate reservoir since the fractures are the classic geological setting for stock and produce hydrocarbon in this kinds of reservoirs. Particularly in Brazil is growing the interest in the characterization of carbonate reservoirs, with the recent discoveries in pre-salt. The acoustic imaging tools provide valuable information about the amplitude of the reflected waves in the borehole wall, which can be interpreted to allow the characterization of fractures. However, some problems arise due to the qualitative interpretation of these images that are basically performed with the use of vision and experience of the interpreter. This work presents a methodology that performing the fractures analysis of acoustic images and can be divided into three steps. The first one presents the image modeling, which is used to infer the aspect of the fractures in different geological settings. In the second step, the mathematical morphology is used as an edge detector and performs the fractures identification in the acoustic image. The last step deals with the extraction of geometric attributes of the fractures with the adoption of a four degree polynomial according to the least square criterion. The evaluation of this methodology is performed with synthetic images generated by the presented modeling, which supports the characterization of fractures performed in real images.