A New Approach to Camera Image Indexing

This paper presents a color filter array (CFA) image indexing approach. To enhance the functionality of single-sensor consumer electronics such as digital cameras, imaging-enabled mobile phones and wireless personal digital assistants (PDAs), the proposed solution embeds the metadata information to a CFA image using a common discrete cosine transform (DCT) based watermarking scheme. Depending on a consumer electronic device employed, the metadata information can be used to indicate ownership, capturing device identification numbers, time and location information. The metadata information can be extracted from the gray-scale, mosaic-like CFA image or the full-color, demosaicked image using PC software commonly available by camera manufacturers or with conventional public image database tools. Simulation studies reported in the paper indicate that the proposed CFA indexing approach does not affect the performance of the demosaicking methods which produce full-color images that are visually identical to those obtained by demosaicking of the non-indexed CFA data.