The present study proposes a new method of particle identification in three-dimensional particle-tracking velocimetry (3D PTV) using binary image data. It is based on the necessary condition for particle identification and the 3D cross-correlation method for evaluating the similarity of particle distribution patterns in two spaces. Computer simulations are carried out to investigate the effects of the sizes of particles, search subregion and identification subregion on the particle identification performance. The method has two key advantages. One is quick and accurate particle identification performance and the other is that the computer memory required for the computation of particle identification is generally small compared to other methods.