Recently, automatic face recognition has become a realistic target of biometrics research. Face fake attacks are truly a threat to face recognition systems. Exploration on non-invasive software based face spoofing detection schemes have been mainly concentrated on the analysis of the luminance information of the face images, accordingly discarding the chroma component, which can be very useful for discriminating fake faces from genuine ones. In this paper, we present a novel approach based on analyzing joint color-texture information of the facial image from the luminance and the chrominance channels using color local binary pattern (LBP) descriptor. Particularly the feature histograms are extracted from each image band separately. The resulting feature histograms are concatenated into an enhanced feature histogram in order to obtain an overall reproduction of the facial color texture. The final feature vector is fed to an ensemble subspace discriminant classifier and it describes whether there is a live person in front of the camera or a fake one. Also, we determine the performance measures of an ensemble classifier and compare with SVM.