Research on non-intrusive software package based mostly face spoofing detection schemes has in the main been targeted on the analysis of the light info of the face pictures, thence discarding the intensity element which may be terribly helpful for discriminating faux faces from real ones. This work introduces a completely unique and appealing approach for sleuthing face spoofing mistreatment color texture analysis. we have a tendency to exploit the joint color-texture info from the light and therefore the chrominance channels by extracting complementary low-level feature descriptions from completely different color areas. the color native binary patterns (LBP) descriptor, explore the facial color texture content mistreatment four different descriptors: the native part division (LPQ), the co-occurrence of adjacent native binary patterns (CoALBP), the binarized applied math image options (BSIF) and therefore the scale-invariant descriptor (SID) that have shown to be effective in gray-scale texture based mostly face anti-spoofing. Here by mistreatment these options {the color the color} texture is analyzed and extracted by face descriptors from completely different color bands. To gain insight into that color areas square measure most fitted for discriminating real faces from faux ones, thought of 3 color areas, namely RGB, HSV and YCbCr. a brand new and appealing approach mistreatment color texture analysis and demonstrate that the intensity element is terribly helpful in discriminating faux faces from real ones. First, the face is detected, cropped associated normalized into an M×N component image. Then, holistic texture descriptions square measure extracted from every color channel and therefore the ensuing feature vectors square measure concatenated into associate increased feature vector so as to induce associate overall illustration of the facial color texture. the ultimate feature vector is fed to a binary classifier and therefore the output score worth describes whether or not it's a true or a faux image.