Color correction involves converting device-dependent RGB values into a device-independent color space, like XYZ or sRGB. This paper introduces a novel method for deriving a matrix-based transformation using tensor product B-splines, extending previous polynomial approaches. The proposed spline-based model offers enhanced adaptability in color correction, consistently surpassing traditional polynomial methods across various cameras. Splines, being piecewise polynomials, offer increased flexibility in modeling, adapting to different spectral characteristics of cameras. Performance comparison of the spline-based model against previous methods was conducted using simulated data of natural scenes on two different cameras in both training and testing phases.