Lots of sources can be used to derive unconfined rock compressive strength (UCS) along the wellbore as a key input of failure criteria for wellbore stability analysis. Laboratory tests are the most accurate methods to obtain rock strength which mainly are destructive and result in a set of discrete data. On the other hand, non-destructive methods (log based approaches), deliver a continuous rock strength log, however, the accuracy of these techniques is still dubious and they should be calibrated with lab data. If the evaluated rock strength contains some errors, final mud weight window will be affected by these errors which lead to inaccurate mud weight selection for safe drilling operation and consequently increase in total well cost. Whole data in UCS log shall be multiplied or added by a number to match laboratory results. This calibration approach may superficially show good match between results from log and laboratory, though all data may not represent real UCS and show same uncertainty. In this paper, a modified calibration approach based on lithology, compaction, fluid content and porosity is proposed for calibration of UCS logs. Performed cross validation results show a good match between laboratory data and modified UCS profile.