Face recognition applications are increasingly becoming the norm, especially in public places such as airports, sports stadiums, and shopping malls. They can recognize people and match them with their passport photos much faster than humans. But the technology also has controversial uses. Some people worry that it might be used for racial profiling, as in the case of racist software. This article explores some of the potential issues with these applications. Read on for more information.
Face recognition systems are more reliable than fingerprinting and iris recognition. They only take pictures of faces when a person enters a defined area. There are no complicated procedures that require the subjects to position their hands on the reader or to position their eyes in front of the scanner. In addition, the subjects do not feel that their privacy is being invaded. But this problem is hardly ever resolved. Some researchers are trying to make these systems more accurate under unconstrained conditions.
To address these issues, face recognition techniques worddocx can use non-linear extensions. In higher-dimensional space, face patterns are mapped onto a simplified manifold. Non-linear techniques such as kernel ICA can be applied to face images. The result is a much more accurate face recognition system. This method is especially useful in situations where the images of two individuals are similar. Face recognition systems can even recognise identical twins. Moreover, they can also use multi-spectral images.
