Fingerprint-based recognition systems are widely used in the field of biometrics. Most of the fingerprint sensors developed so far acquire fingerprint images through a solid plate of the surface, which degrades the recognition performance due to deformations. These deformations are caused by the pressure of the physical contacts that are costly and hard to estimate. In this paper, a touchless fingerprint recognition system is devised using a camera sensor to resolve this problem. However, this system raises some new problems such as defocusing, low ridge-valley contrast, 3D-to-2D mapping and so forth. Herein, adequate solutions will be discussed by introducing modified steps of the algorithm. Our proposed system is compared with the primitive touchless sensor with no manipulation and the algorithm of the preceding touch-based system, and future works are suggested.