Abstract: Plane porosity is a key indicator for assessing the quality and resource potential of tight reservoirs. The current reservoir pore intelligent extraction and plane porosity calculation methods based on single image analysis technology have problems such as cumbersome pre-configuration, weak learning ability of sparse samples, and low accuracy of complex pore morphology recognition. For this reason, this paper proposes an intelligent calculation method of plane porosity in tight sandstone thin section by integrating SOLO (segmenting objects by locations) v2 algorithm and OpenCV (open source computer vision library) based on the idea of hybrid intelligence. Using the instance segmentation algorithm SOLOv2 to segment and label the pore regions in the image, and the distribution and percentage of pores are extracted in combination with OpenCV, then the plane porosity is calculated. Comparative experimental results show that this method is superior to the comparative algorithms such as YOLACT (you only look at coefficients), Mask R-CNN (mask region-based convolutional neural network) and SOLO in terms of Dice coefficient (0.88), pixel accuracy (0.91), and plane porosity calculation error (<0.1), with a faster execution speed.

Key words: tight sandstone, plane porosity, hybrid intelligence, SOLOv2, OpenCV, reservoir