Determination of porosity value and type is one of the most important parameters in reservoir quality assessments. In this study, using sequential digital images from 30 samples of a carbonate reservoir, the 3D porosity model has been reconstructed. A very thin layer of rock (200 micron) is picked up in any stage and a digital photo was prepared. As the main purpose is porosity network modeling, rock surface was saturated with a green colored paste for porosity separation. The excessive green paste then removed from the rock surface. Prepared images were analyzed with MATLAB software and green channel was extracted. Each value of this channel with its x, y and z coordination stored in a new matrix. By modeling of the matrix in RockWorks software and filtering based on mean ± standard deviation porosity distribution model was reconstructed and quantified. In addition of porosity volume and 3D network, other rock parameters such as reducible and irreducible water saturation, permeability and pore throat can be also determined using the proposed model.