Three-dimensional (3-D) surface reconstruction of reflective objects and colored objects is a challenge. To this end, we proposed a novel approach. First, a new method of pixel matching between the projected image and the captured image is proposed. Then, we analyzed the impact of color texture and reflective surface on the captured image quality, and proposed a black-and-white (B/W) camera-based color information extraction method, an object surface reflectance and ambient light interference estimation method. Finally, the pixel color, pixel intensity and exposure time can be adaptively adjusted according to the color information and reflectivity of the measured objects. Experiments verified that the proposed method can not only obtain high-quality captured images, but also has a smaller number of additional images and a wider range of applications than existing methods.