Colour Texture Analysis of Face Spoof Detection using CNN Classifier
Suresh Bojja1, K Naga Prakash2

1Suresh Bojja*, Department of ECE, JNTUK, Kakinada, India.
2K Naga Prakash, Department of ECE, Gudlavalleru Engineering College, Gudlavalleru, India
Manuscript received on September 23, 2019. | Revised Manuscript received on October 15, 2019. | Manuscript published on October 30, 2019. | PP: 5037-5046 | Volume-9 Issue-1, October 2019 | Retrieval Number: A3006109119/2019©BEIESP | DOI: 10.35940/ijeat.A3006.109119
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract: The emphasis on analysis of various research schemes of non – intrusive software based face spoofing detection is now a days gaining reputation in image and video processing tools. The analysis on luminance(Y)data of the various face images which provides the discrimination of forged faces from genuine faces by removing the chroma component. Here the work provides an innovative approach that perceives spoofed face using texture analysis (colour)by exploiting combined colour texture information from various channels such as luminance and chrominance. This helps to exploit joint information by removing degraded feature metaphors from dissimilar colour models.Precisely the featured histograms are figured over all images that obtained from the YCrCb colour model band distinctly.The concatenation of testing and training by using Neural Network for classification of spoofed images by the concept of blending of images gives the best possible outcomes. Wide-ranging researches on face data bases is most interesting target datasets paves the way for best processing face spoofing results than state of art. The proposed method gives stable performance when compared with the most unlike methods that conferred in the literature survey. The promising outcomes of evaluation suggests that facial colour texture depiction is added steady strange conditions associated to gray-scale complements.The favourableoutcomeswereattained using these CNN(Convolution Neural Network)designs for face antispoofing in diversesituations.
Keywords: Recognition of Face, spoofedrecognition, examination of colour texture