NIR Face Databases
Nov 01, 2016Author:
PrintText Size A A

Near infrared (NIR) face imaging technique is proposed to overcome the illumination variation problem arising from uncontrolled environmental lights in visible light images and to produce face images of a good illumination condition for face recognition.

The NIR imaging is a more like normal visible light imaging, though NIR is invisible to the naked eyes. We use NIR light-emitting diodes (LEDs) as active radiation sources, which are strong enough for indoor use and are power-effective. Meanwhile, we use a long pass optical filter to cut off visible light while allowing NIR light to pass. Fig. 1 shows the capture devices. The device consists of 18 NIR LEDs, an NIR camera, a color camera, and the box. The NIR LEDs and camera are for NIR face image acquisition. The color camera capture color face images may be used for fusion with the NIR images or for other purposes. The hardware and the face are relatively positioned in such a way that the lighting is frontal and NIR rays provide nearly homogenous illumination on face. The imaging hardware works at a rate of 30 frames per second with the USB 2.0 protocol for 640*480 images.


Fig. 1 Active NIR imageing system (Upper) and its geometric relationship with the face (Lower).

A NIR Face Database captured by the device introduced above is collected. It is released now to promote research and development of NIR face detection, NIR eye detection and NIR face recognition and other related researches. The NIR face database consists of 3,940 images with the image size of 640x480 from 197 subjects in total. Fig.2 shows some NIR example images. More details about the NIR face image and the capture device are available in reference below.


Fig. 2 NIR face example images