BioHashing is a popular biometric template protection scheme defined in the last decade. Most of previous studies on this algorithm focus on the performance optimization or the use of this privacy protection scheme on many types of biometric modalities (face, fingerprint, palmprint. . . ). The objective of this paper is to study the robustness of this algorithm on a texture based representation by testing and simulating operational attacks. We consider in this study some quantitative measures of the robustness of the BioHashing algorithm and we show some new results on its security on fingerprints represented by texture features.