Secure Palmprint Recognition based on Multispectral Sequential Capture

The security of personal identities is a serious challenge in today's digital world, with so many daily transactions requiring secure solutions. use biometric characteristics the person presented as reliable solution to solve this problem. Indeed, effective, but it hides weak point which lies ability reproduce certain for fraud. To overcome point, we propose approach palmprints that relies on concept merging multiple features. these features will be extracted from multi-spectral images different spectra, allow extraction information under skin palm two spectrums sequentially times (T1, T2) instantly. instant fusion impossible replicate. used are grayscale. satisfy construction and system, kind patterns (palmprints), Compound Local Binary Pattern method, since method adds an additional bit each P bits coded by LBP corresponding neighbor local neighborhood, order build robust system. This feature descriptor, uses both sign tilt differences between central neighboring gray values. reliability proposed has been demonstrated Casia Multi-Spectral database. final experimental results show recognition rates vary 99% 100% left right palms.