The primary motivation of this paper is to present an accurate and universal method for the nonlinear system modeling. An Unconstrained Bezier Calibration Method (UBCM) is presented. By constraining the boundary freedom of the UBCM, it could become a constrained form called the Bezier Calibration Method (BCM), which is suitable for situations with an accurate boundary modeling requirement. A comparison study is performed to compare the presented method with the most widely utilized Polynomial Calibration Method (PCM) through several nonlinear behavior modeling examples, including sensor inherent nonlinearity calibration, sine wave, gauss nonlinearity and nonlinear broken line. A comparison study demonstrates that in the present verification examples the UBCM has better calibration performance than that of the BCM and PCM.