초록 ▼

This invention has as its object to provide a technique for embedding digital-watermark information in image data while minimizing deterioration of the original image quality. To this end, a data processing apparatus of this invention has input means for inputting image data consisting of a pluralit

This invention has as its object to provide a technique for embedding digital-watermark information in image data while minimizing deterioration of the original image quality. To this end, a data processing apparatus of this invention has input means for inputting image data consisting of a plurality of coefficients, and embedding means for embedding digital-watermark information in coefficients having values falling within a predetermined range of the input image data.

대표청구항 ▼

1. A data processing apparatus comprising: input means for inputting image data composed of a plurality of coefficients, from which a plurality of subsets are formed, coefficients of each subset representing a same image area; specifying means for specifying a group of coefficients whose values

1. A data processing apparatus comprising: input means for inputting image data composed of a plurality of coefficients, from which a plurality of subsets are formed, coefficients of each subset representing a same image area; specifying means for specifying a group of coefficients whose values are within a predetermined range from the plurality of coefficients; and embedding means for embedding digital-watermark information to at least one coefficient selected at random from the coefficient group specified by said specifying means, wherein the random selection is executed in subset units, and wherein said embedding means error correction encodes the digital-watermark information, and embeds the obtained digital-watermark information thereafter. 2. An apparatus according to claim 1, wherein the plurality of coefficients correspond to a plurality of frequency components obtained by wavelet-transforming predetermined image data. 3. An apparatus according to claim 2, wherein said embedding means embeds the digital-watermark information in a random coefficient other than a lowest frequency component included in the coefficient group. 4. An apparatus according to claim 1, wherein said embedding means embeds the digital-watermark information in a random coefficient having a large absolute value. 5. An apparatus according to claim 1, further comprising: output means for outputting information for specifying the coefficient to which the digital-watermark information is embedded, together with the image data which has undergone embedding by said embedding means. 6. An apparatus according to claim 1, further comprising: output means for outputting interpolation data for reconstructing image data before embedding, together with the image data to which embedding is executed by said embedding means. 7. A data processing method comprising the steps of: inputting image data composed of a plurality of coefficients, from which a plurality of subsets are formed, coefficients of each subset representing a same image area; specifying a group of coefficients whose values are within a predetermined range from the plurality of coefficients; and embedding digital-watermark information to a coefficient selected at random from the coefficient group specified by said specifying step, wherein the random selection is executed in subset units, and wherein said embedding step error correction encodes the digital-watermark information, and embeds the obtained digital-watermark information thereafter. 8. A memory medium storing computer-executable code to perform a data processing method, said code comprising: code for inputting image data composed of a plurality of coefficients, from which a plurality of subsets are formed, coefficients of each subset representing a same image area; code for specifying a group of coefficients whose values are within a predetermined range from the plurality of coefficients; and code for embedding digital-watermark information to a coefficient selected at random from the coefficient group specified by said specifying code, wherein the random selection is executed in subset units, and wherein said code for embedding error correction encodes the digital-watermark information, and embeds the obtained digital-watermark information thereafter.