Fringe projection profilometry (FPP) is one of the most effective methods to obtain three-dimensional surface topography information about an object. However, for objects with large variations in surface reflectance, it is difficult to measure the saturated regions with conventional FPP. In this paper, a new adaptive fringe projection (AFP) method is proposed to solve this problem. Compared with existing methods, the proposed method can automatically generate low-intensity projected fringe patterns instead of requiring manual adjustment. First, we built a projection intensity model for the captured pixels and defined the surface reflection factor and environmental factor for saturated pixels. Then, a small number of uniform grayscale pattern sequences were projected to mark the saturated areas, while the surface reflection factor and the environment factor of individual pixels were calculated and a surface coefficient look-up table (SCLUT) was created. The low-intensity projected fringe patterns were automatically calculated and generated based on the captured uniform grayscale sequences of images to establish the mapping relationship between the camera image plane and the projector image plane. Finally, the optimal projection intensities of pixels in the saturated regions were calculated according to the SCLUT, and adaptive fringe patterns were generated. Experiments show that using the adaptive fringe patterns generated by the proposed method to measure objects can accurately adjust the projection intensity and substantially improve the measurement accuracy of high-reflective surface.