Publication Date: 2011-Jul-07
The IP.com Prior Art Database
There are many known ways to compress a digital image. The standard created by the joint photographic experts group for photographic or painted images, usually abbreviated jpeg or .jpg, is especially common with digital cameras and on the web. For line art, or barcode labels, .tiff, .gif, and other lossless graphics formats are common. But these widely used standards still maintain more information than is required in some common applications using common card printers, label printers or RFID equipment. It may be desirable to hold a digital photograph in forms of non-traditional memory, such as an RFID tag or a two-dimensional barcode. But even though these compression methods reduce the total amount of data that must be stored, the net result may still require a particularly high capacity RFID tag, resulting in higher costs, or a particularly large 2D barcode, resulting in the need for a larger label or plastic card. The image size could be reduced, but then it is harder to make out important details. The image could also be modified to black and white, but this may also be unsuitable for some applications, particularly when quick recognition is required.
One particular application where this is important is with photographs of a person’s face. As security standards evolve, it is particularly valuable to law enforcement, security personnel, nurses, border crossing officials, bouncers, guards, and others to access a photograph. Comparing a person’s face to the photograph of a person on a card or RFID tag is a fast and easy way to confirm identity for access control or for verifying identity. This is even in the stages of being standardized for drivers’ licenses and passports. A problem arises however in the speed with which this comparison can be done. A big, full color photograph is easiest to compare to a live person’s face, but a small black and white photograph would be optimal from a data storage and data communication perspective.
The JPEG standard allows for embedding a particular color space, but this is often ignored by applications. Today, most commonly used color spaces use three or four dimensions, such as CMYK, or Y’CBCR. However, for the types of applications described above, the two-dimensional colorspace Red-Cyan is completely acceptable, since nearly the entire tonal range of human skin falls can be captured in only a small portion of the colorspace which is visible by the human eye. Although two dimensional colorspaces have been used in the past, such as Disney cartoons in Technicolor, they have rarely been applied to digit...