In the non-contact optical measurement techniques, the depth of field limits the range of the measurement. If the step height sample is out of the depth of field, and it will cause the measurement to fail. In this study, the proposed integration framework integrates the movement of PZT (piezoelectric transducer) stage and the algorithm of the multiple-height image fusion. It can precisely define the height discontinuity of the step height sample which is out of the depth of field. In detail, the lead zirconate titanate actuator, namely PZT actuator, moves a fixed distance with 1 m. And then, the camera captures one sheet of the 2D image. The working distance of PZT is 800 m. After repeating the cycle of PZT and the camera with 800 times, the 800 sheets of the 2D image are obtained. The algorithm of the multiple-height image fusion enables the 800 sheets of the 2D image to become the one of 3D reconstruction image with the noise. Finally, the filtering algorithm is used to remove the noise in the 3D reconstruction image. In the experiment system with the depth of field 100 m, the accuracy of the step height sample with 200 m is about 1 m (i.e. 0.5%).