GB2402574A

GB2402574A - Recovery of a watermark from a combined image file of the watermark and a compressed image based on the combined image file size

Info

Publication number: GB2402574A
Application number: GB0419833A
Authority: GB
Inventors: Tinku Acharya; Kannan Raj
Original assignee: Intel Corp
Current assignee: Intel Corp
Priority date: 2000-03-06
Filing date: 2001-01-18
Publication date: 2004-12-08
Anticipated expiration: 2021-01-18
Also published as: GB2402574B; GB0419833D0

Abstract

Recovery of a watermark from a combined image file of the watermark and a compressed image are disclosed. The watermark is recovered by determining locations in the combined image file that include groupings of bits of the watermark, the groupings of bits being inserted at locations in the compressed image determined at least in part on a size of the combined image file, extracting the bit groupings for the watermark from the locations and forming the watermark from the extracted bit groupings for the watermark. An independent claim is also included for recovering a watermark from a combined image file wherein the locations of groupings of bits including the watermark are selected in accordance with a pseudo-random number generation technique.

Description

METHOD OF RECOVERING A WATERMARK

FROM A COMPRESSED IMAGE

RELATED APPLICATION

This patent application is related to concurrently filed U.S. patent application serial no. 09/519,874, titled "Method of Integrating a Watermark into an Image" by Acharya et al., now U.S. 6654501B, assigned to the assignee of the current invention and herein incorporated by reference.

BACKGROUND

This disclosure is related to watermarks and, more particularly, to recovering an embedded watermark from an image, such as a compressed digital image.

The management of information is becoming a more challenging problem with the progress of information technology, especially in the area of digital imaging.

Identification of ownership and/or other rights with respect to an image when transmitted through a network, such as a public network, is receiving a great deal of attention. Furthermore, information management and the secure movement of image data are becoming complex tasks, particularly in light of the rapid evolution of a variety of technologies. "Digital Watermarking" may provide a potential solution to this problem.

SUMMARY OF THE INVENTION

According to a first aspect of this invention there is provided a method as claimed in claim 1 herein.

According to a second aspect of this invention there is provided an article as claimed in claim 8 herein.

According to a third aspect of this invention there is provided a computer program as claimed in claim 12 herein.

BRIEF DESCRIPTION OF THE DRAWINGS

The subject matter regarded as the invention is particularly pointed out and distinctly claimed in the concluding portion of the specification. The invention, however, both as to organisation and method of operation, together with objects, features, and advantages thereof, may best be understood by reference to the following detailed description when read with the accompanying drawings in which: Fig. 1 is a schematic diagram illustrating one example application of a method of integrating a watermark into a compressed image in accordance with the present invention; Fig. 2 is a flowchart illustrating a method of integrating a watermark into a compressed image; and Fig. 3 is a flowchart illustrating an embodiment of recovering the integrated watermark from the compressed image in accordance with the invention, specifically 3 for the embodiment illustrated in Fig. 2. 2'

DETAILED DESCRIPTION

In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be understood by those skilled in the art that the present invention may be practiced without these specific details. In other instances, welI-own methods, procedures, components, and circuits have not been described in detail so as not to obscure the present invention.

Some portions of the detailed description whuch follow are presented in. terms of algorithms and symbolic representations of operations on data bits or binary digital signals within a computer memory. These algorithmic descriptions and representations are the techniques used by those sl illed in the data processing arts to convey the substance of their work to others skilled in the art. An algorithm is here, and generally, considered to be a self- consistent sequence of acts or operations leading to a desired result. These include physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers or the like. It should be understood, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely j convenient labels applied to these quantities. Unless specifically stated otherwise, as apparent from the following discussions, it is appreciated that throughout the specification, discussions utilizing terms such as "processing" or "computing" or calculating" or "determining" or the lilts, refer to the action and processes of a computer or computing system, or similar electronic computing device, that manipulate and transform data represented as physical, such as electronic quantities, within the computing system's registers indoor memories into other data similarly represented as physical quantities within the computing system's memories, registers or other such information storage, transmission, retrieval, or display devices.

Operations, such as image acquisition, encoding, manipulation, transmission, retrieval, or storage particularly involve careful handling of noise that may be present in one or more images and which may appear as perceptually disturbing artifacts. The iropact of this may be reduced by reducing the impact at the source. However, this may not always be possible, and, furthermore, images that may be digitally captured may, at times, be left with rennanents of noise, even after the application of noise reduction techniques. Techniques and processes are available to reduce the Impact of spatial noise on sell images and digital cameras and both spatial and high frequency noise in video image capture systems.

However, another field of image representation is referred to as "Digital Waterrnarking," in which specific signal information may Epically be encoded as "noise." A Digital Waterrnarlc may be a specially encoded symbol or signature placed in the image for the purpose of image authentication or verification or simply for encoding secret information in one or more images.

It would be desirable to have a process or technique for encoding Digital Watermarks that was unaffected, or at least substantially unaffected, by the process of image compressionldecompression. One embodunent of a method of integrating a watermark into a compressed image in accordance with the present invention provides that capability.

Digital Watermarks, such as logos or signatures, may be introduced into the image in a variety of ways. As one example, although the invention is not limited in scope in this respect, the watermark itself may be, for example, a 32 x 32 image block of a logo or signature that may be encrypted or, alternatively, employed in the form provided. This watermark signal information, of a portion of the watermark may be combined into the image, or a portion thereof, after the image or image portion has been compressed. This may be implemented by inserting the watermark pixel signal values between selected pixel values i or between selected pixel locations using a suitable technique or process. Therefore, in this example, the signal information may be intermittently interspersed in the compressed image, or portion thereof. In particular, in one such embodiment, the watermark, or a portion thereof, may be encoded, the image, or a portion thereof, may be compressed, and the encoded watermark and the compressed image, or portions thereof, may then be combined in this manner, for example.

Typically, after an image is captured by an image capture device, such as an imager, for example, it undergoes image processing, followed usually by compression, before storage and/or transmission. To retrieve and display the image, a corresponding decompression process or technique is typically applied. Usually, feasible image compressiorUdecompression processes or techniques are "lossy" in nature, that is, the pixel signal values of the decompressed image do not correspond precisely to the original image.

If the Digital Watermark, in the form of pseudo-random noise, is encoded in the unage, the compression and decompression typically will corrupt the encoded watermark signal information and, hence, the watennak will not be recoverable for identification and/or authentication to occur.

In one arrangement an encoded watermark signal infonnation is integrated into a compressed image. Of course, as previously indicated, the watermark signal information may alternatively not be encoded. Furthermore, in one particular embodiment, some relevant signal information may also be provided in the header, although the invention is not limited in scope in this respect. The encoded watermark signal information, therefore, is combined in the compressed data of the compressed image, or a selected portion thereof, such as by intermittently interspersin,, the watermark signal information with the compressed ima,,e data. The watermark signal information may then be decoded from the compressed ima,,e either prior to decompression of the compressed image or, alternatively, during the decompression process, depending upon the particular integration methodology that may have been applied during integration of the watermark signal information with the compressed image. The decoded watermark signal information may then be blended with the decompressed image in'a perceptually non-detectible or barely detectible manner, if desired, as described, for example, in the aforementioned concurrently filed patent application, although the invention is not limited in scope in this respect For example, this may not be performed.

Encryption of watermark signal information may be done by encoding the infonnation in the spatial domain or the transform domain. Image or spatial domain encoding may involve random bits of scrambling with a nonscrambling key similar to what has been applied in 128-bit encryption for Internet applications and image transmission, for example. See, for example, "A method for obtaining digital signatures and public}fey cryptosystem," ACM Cornmun. 1978, Vol. 21, ppl20-126; 'valuation of selective encryption techniques for secure transmission of MPEGcompressed bit-streams," Alattar et al., Proc. ISCAS, pp 340-343. vol. 14, 1999; or "Combining, data compression and encryption," C. Finnila, HERE WESCON/94, pp 40409, although, these are merely provides as examples. Alternately, transform domain encodin,, may be done using any one of a number of ways, such as with Fourier transform encoding, wavelet transform encodin,,, Walsh-Hadamard transform encoding, etc. It will, of course, be appreciated by one of ordinary skill in the art that these are just examples of image domain encoding and frequency domain encoding, and the invention is not limited in scope to any particular approach. Any non of number of currently nown or to be later developed approaches may be employed.

In one particular arrangement of integrating a watermark into a compressed image, a process may include the following. For an image stored as a file of pLxel signal values, for example, the compressed image file is represented as C(F). Let B denote a 39 x 32 block representing the watermark signal information in this particular embodiment. Of course, the invention is not limited in scope to a watermark or watermark signal information of any particular size or shape. In this particular embodiment, watermarI; signal information B is encrypted. Let E(B) represent the bit stream of the encrypted block B. In this particular embodiment E(B) is divided into a series or sequence of groupings of bits, here bit "bytes," which represent the watermark encoded signal values.

For example,

E(B) = NoNll!TzN3.. No tl3 where each Ni represents a byte in the encrypted blocl: E(B). Therefore, E(13) is the binary digital signal representation or bits for the encrypted block. Of course, the invention is not restricted to dividing the watermark into bytes. For example, in an alternative embodiment, the watermark signal information may be divided into groupings of bits that comprise 4 bit nibbles instead. In this particular arrangement the watermark signal information E(B) may then be included or combined into the compressed file C(F) using the approach described! hereinafter.

Here, the bytes of the encrypted watermark block E(B) are distributed over the compressed Image file between selected byte locations in the compressed image. Likewise, in this particular embodiment, the byte locations are selected in accordance with a pseudo- random number,v,eneration technique. Any one of a number of pseudo-random number generation techniques may be employed. In this particular arrangement' an initial "seed" of the pseudo-random process is selected substantially in accordance with the following relationship [2]: () L[size[(B)+ size(C(F)]/ 1024_| where "sized" denotes the size of the file in bytes in this embodiment, To generate the next pseudo-random number, apply the pseudo-random number =,eneration technique usin==, the seed equal to the existing value of Q. Let Q vow represent the newly generated pseudo random number.

For this particular arrangement, the encoded watermark signal information is combined with the compressed file signal information m the following manner. In the compressed file C(17, insert byte Nit of E(B) between bytes BQand BE of the compressed image file C(;), and i varies from O to m. A pixel P(m,n) may be mapped into Bp where P= Macon, c is the number of pixels in each row, r is the number of rows, m goes from O to r-1 by I, and n goes from O to c-1 by 1. By this technique, Q may be mapped tO a pixel location BQ, although, of course, any one of a number of other techniques for malting an assi=,nent between numbers and pixel locations may be employed. Therefore, after inserting the first byte No of encrypted watermark, E(B), in this particular arrangement, for example, to generate the next byte location tO insert the next byte N', apply the same pseudo-random number generation technique using the existing value of Q as the seed This is illustrated schematically in Fig. 1.

Fig. 2 is a flowchart illustrating the previously described embodiment, although, again, this is just one example. For example, as illustrated in Fig. 2, in block 1 10 the image is compressed, and in block 120, the watermark is encoded, in this arrangement, encrypted.

Furthermore, as previously described, the watermark may not even be encoded or encrypted.

In block 130, as illustrated, the initial seed, Q. is selected, and I, a counter, is set to zero. For this arrangement the initial seed is selected in accordance with the previously provided relationship [2], although, again, the invention is not limited in scope to this initial seed, or to using a counter, or one that begins with zero. In block 140, a pseudo-random number generation technique is applied to the initial seed, and Q is replaced with the value generated in this manner. As previously described. the byte Ni is then inserted in compressed image file C(F) between bytes Bo and Bq+ of C(F), in block 150. For this arrangement, this process is continued until the entire encoded watermark is combined into the compressed imge, although that may not be the case in alternative arrangements. An aspect of this particular arrangement is the secure transfer of information. For example, for this arrangement, if both the sender and the receiver are familiar with the approach employed to integrate the watermark into the compressed image, then someone else intercepting the image will generally be unable to extract or recover the watermark, and, therefore, will not be able to access the information being transmitted in this fashion. Likewise, it may be possible for the receiver to tell if image has been intercepted and whether some tampering has occurred.

For an arrangement of a method of integrating a watermark into a compressed image, there is a corresponding embodiment of a method of recovering a watermark or encoded watermark fi om a compressed image in accordance with the present invention. For example, for the previous described arrangement, a corresponding embodiment for recovering a watermark or an encoded watermark from a compressed image includes: determining the locations of the groupings of bits, such as, in this embodiment bytes, in the combined compressed image-encrypted watermark file, CE tF), that correspond to the groupings of bits, here bytes, Ni, of the encoded watermark 33(B), in this particular embodiment; extracting the bytes for the encoded watermarl; from those respective byte locations in file CE (F); and forming the encoded watermark from the extracted pixel signal values. It is noted that, of course, the waterrnarl; extracted and formed may be a watermark subportion, and the compressed image may be a compressed image subportion. Likewise, the encoded watermark, whether or not a subportion, may be decoded and blended with the decompressed image, whether or not a subportion, if desired, such as, for example, by using a technique or approach, such as one of those described in the concurrently filed aforementioned patent application, although, as previously indicated, the invention is not limited in scope in this respect. For example, the decoded watermark may not be blended with the decompressed image at all, or they may altematvely be blended using another technique other than one of those described in the aforementioned patent application.

In this particular embodiment, due to the approach in which the bytes of E(B) have been inserted into the combined file of the compressed image and the encoded watermark, represented by CE (F), will have a size, for example, in bytes, equal to the sum of the respective sizes of C(F) and E(B), a;,ain, for this embodiment, in bytes. Therefore, the location of the particular bytes in the compressed image file may be recovered. In particular, the seed for the random process previously described above is in accordance with the following relationship [3]: Q Lsize[CE tF)]tlO24] g It is noted that although any pseudo-random number generation technique may be employed, the same pseudo-random number generation techruque must be employed for codin,, and decoding for this particular embodiment. Therefore, the pseudo-random process is applied to the seed, and that number is designated Q A byte Ni is located in the compressed image right after BQin CE(F) and, therefore, it may be extracted. After all the bytes, Ni of the encrypted watermarI; E() have been extracted, they may be concatenated to provide E(). This particular embodiment is illustrated by the flowchart in Fig. 3, although tier. invention is not limited in scope ir1 this respect, of course.

This particular arrangement of a method of integrating a watermark into a compressed image and this embodiment of recovering an encoded watermari; from a compressed image provides a number of advantages. It allows secure image data representation and secure movement of signal information, as previously described. Likewise, as previously indicated, the watermarks may be encoded and encrypted in either the image or frequency domain which provides an added level of security for this particular arrangement. An additional level of security may be obtained by encrypting the data placed in the image header, although the invention is not limited in scope in this respect. Likewise, a double level of encryption may be provided, in this particular embodiment, because, initially, the waterrnarl; pixel signal values may be encrypted, and, then, the choice of pixel locations may depend on the seed and pseudo-random number generation employed. Likewise, another advantage of this particular embodiment is the relative ease, in terms of computational complexity, to extract the watermark from the image and verify authenticity. Likewise, due to the nature of this particular embodiment, it is relatively easy to remove the watermark signal information prior to compression and decompression for image transmission/recovery.

It will, of course, be understood that, although a particular embodiment has just been described, the invention is not limited in scope to a particular embodiment or implementation.

For example, one embodiment may be in hardware, whereas another embodiment may be in software. Likewise, an embodiment may be in ironware, or any combination of hardware, software, or firmware, for example. Likewise, although the invention is not limited in scope in this respect, one embodiment may comprise an article, such as a storage medium. Such a storage medium, such as, for example, a CD-ROM, or a disl;, may have stored thereon instructions, which when executed by a system, such as a computer system or platform, or an imaging system, may result in a method of upscaling a color image in accordance with the invention, such as, for example, the embodiment previously described.

While certain features of the invention have been illustrated and described herein, many modifications, substitutions, changes and equivalents will now occur to those skilled in the art. iO

Claims

CLAIMS: 1. A method of recovering a watermark from a combined image file

of the watermark and a compressed image, said method comprising: determining locations in the combined image file that include groupings of bits of the watermark, the groupings of bits being inserted at locations in the compressed image determined based, at least in part, on a size of the combined image file; extracting the bit groupings for the watermark from the locations; and forming the watermark from the extracted bit groupings for the watermark.
2. The method of claim 1, wherein the watermark as formed from the extracted bit groupings is a subportion of a watermark.
3. The method of claim 1, wherein the compressed image is a compressed subportion of an image.
4. The method of claim 1, wherein the watermark is an encoded watermark.
5. The method of claim 4, and further comprising: decoding the encoded watermark.
6. The method of claim 1, wherein the groupings of bits comprise bytes.
7. The method of claim 1, wherein the size of the combined image file is equal to a sum of respective sizes of the watermark and the compressed image.
8. An article comprising: a storage medium having stored thereon instructions to recovering a watermark, or portion thereof, from a combined image file of the watermark, or portion thereof, and a compressed image, or portion thereof, the instructions, when executed by a system, resulting in: 1 1 determining locations in the combined image file that include groupings of bits for the watermark, or portion thereof, the groupings of bits being inserted at locations in the compressed image, or portion thereof, selected in accordance with a pseudo-random number generation technique; extracting the bit groupings for the watermark from the locations; and forming the watermark, or portion thereof, from the extracted bit groupings for the watermark.
9. The article of claim 8, wherein the instructions, when executed, result in recovering an encoded watermark, or portion thereof, from the combined image file.
10. The article of claim 8, wherein the instructions, when executed, further result in: decoding the encoded watermark, or portion thereof.
11. The article of claim 8, wherein the combined image file has a size that is equal to a sum of respective sizes of the watermark, or portion thereof, and the compressed image, or portion thereof.
12. A computer program comprising computer program code means adapted to perform all the steps of any of claims 1-7 when that program is run on a computer.
13 A method as claimed in claim 1 and substantially as herein described with reference to Figure 3 of the accompanying drawings.
14. An article as claimed in claim 8 and substantially as herein described with reference to Figure 3 of the accompanying drawings.
15. A computer program as claimed in claim 12 and substantially as herein described with reference to Figure 3 of the accompanying drawings. in,