The phase unwrapping procedure is the most important step to achieve high accuracy, high resolution three-dimensional (3-D) shape data of objects. Among numerous techniques of phase unwrapping, the region-based algorithms are outstanding, which provide a trade-off between robustness and the computational complexity. However, there is indeed a weakness in the region-based algorithms that the unwrapped data may be incorrect due to the wrapped process generating the same wrapped data of two different areas. To overcome these problems, we propose a novel phase-unwrapping method which depended on region-based methods and quality guided techniques. First, in principle the whole procedure of generating, capturing images and three-step phase-shifting algorithm are introduced and analyzed in detail. Then, the size of the region can be calculated automatically to divide the wrapped image into homogenous areas. Afterwards, based on the method described here, a non-continuous region joining method is applied to reconstruct the continuous unwrapped phase map. In addition, Itoh and Iso phase-unwrapping methods are implemented to unwrap the phase data for the comparison and the results indicate that the proposed method can unwrap phase data accurately. At last, some reconstruction images are presented and the results are analyzed. Experimental results have demonstrated that the proposed algorithm can resolve the problems the Iso phase-unwrapping method encounter in complex wrapped phase cases.