Abstract: In this paper, we present an algorithm, which is incorporated into geometrical calibration, to accurately and efficiently calibrate and correct the nonlinearity. The proposed method utilizes phase to obtain the intensity correspondences between projected patterns and recorded images. Based on the intensity correspondences, the nonlinear response of the system is further retrieved. Along with geometrical calibration, without adjusting setup or using any additional specially-designed structured light patterns, a look-up table for correcting the nonlinear response is directly and accurately derived, and is applied to the intensities of the patterns before projection. Experiments show that the root mean square of the phase error is significantly reduced by a factor of 21.8 times.