Thermal human face recognition based on GappyPCA

Over the last forty years several different techniques have been proposed for human face recognition. This has been used in different application especially in the field, where security is the main concern. Most of the face recognition algorithms work on visual face images. But the visual face images have several drawbacks because the quality of the visual images degrades gradually with the increase of darkness. So, face recognition system doesn't work properly when illumination changes. But thermal infrared camera can capture the same images even in dark situation which means thermal images are illumination independent. We have proposed a GappyPCA based thermal human face recognition technique for frontal face only. The results obtained on testing sets, about 98.61% correct recognition using this technique.