Method and apparatus for image encryption and embedding and related applications

BACKGROUND OF THE INVENTION

00011. Field of the Invention

0002This invention relates to image encryption, and in particular, it relates to a method and apparatus for encrypting an image using an encryption key and embedding the encrypted image in a cover image. It also relates to methods and apparatuses of applying image encryption in scanning, printing, and electronic transmission of images.

00032. Description of the Related Art

0004Scrambling of image signals has been used in video transmission and distribution as well as other areas to prevent unauthorized use. Scrambling generally refers to shifting the locations of lines and/or blocks on an image (e.g. video screen) so that the image is not easily recognizable. For example, in U.S. Pat. No. 6,535,687, “Method and system for deterring electronic video piracy through image rearrangement”, blocks of 3×3 pixels are rearranged using a pseudo random number sequence of numbers 1 to 9. The encryption method is implemented in a projector. U.S. Pat. No. 5,815,572, “Video scrambling”, teaches scrambling video data using a combination of line reversal, line permutation, line inversion, and block permutation. The permutation tables and other information needed for unscrambling are stored in the receiver. In U.S. Pat. No. 5,583,941, “Imaging processor”, pixel blocks are divided into 2×2 sub-blocks which are transposed diagonally to achieve scrambling. Different levels of scrambling are performed using different block sizes. The scrambling process is implemented in a printer. In U.S. Pat. No. 5,535,275, “Apparatus and method for producing scrambled digital video signals”, the screen is divided into 5×10 “super-blocks” each made up of 27 macroblocks, each macroblock consisting of 6 DCT blocks Y<b>1</b> Y<b>2</b> CR Y<b>3</b> Y<b>4</b> CB. A DCT block is an 8×8 pixel area. The macroblocks are read out in a shuffled sequence, where super-blocks form five columns labeled “A B F D E” and are read out one row at a time in the column sequence of “F B D A ” (see <figref idref="DRAWINGS">FIG. 13</figref> and cols. <b>8</b>-<b>9</b>). This patent also teaches further scrambling by exchanging the data order of the Y, CR and CB components (see <figref idref="DRAWINGS">FIGS. 1-4</figref> and cols. <b>9</b>-<b>10</b>), as well as bit manipulation of the DCT coefficients (see <figref idref="DRAWINGS">FIGS. 5-7</figref> and cols. <b>10</b>-<b>11</b>). U.S. Pat. No. 5,321,748, “Method and apparatus for television signal scrambling using block shuffling”, describes video scrambling in which video lines are shuffled within a block of lines. One pseudo random number sequence (e.g. a sequence of numbers 1 to 8) is used for each block of lines and the seed numbers for the pseudo random number sequences are inserted into the video date stream. U.S. Pat. No. 5,208,85, “Method and device for scrambling-unscrambling digital image data”, describes scrambling of digital image where each block of N×N pixels is divided into 4 sub blocks and inter-block and intra-block scrambling are performed. Inter-block scrambling exchanges two sub blocks that are located diagonally. Intra-block scrambling exchange pairs of pixels that are located at mirror image locations with respect to a line.

0005Other scrambling techniques applies scrambling after the pixels have been transformed (e.g. by discrete cosine transform), such as the method described in U.S. Pat. No. 6,505,299, “Digital image scrambling for image coding systems”, and the method described in “Efficient Frequency Domain Selective Scrambling of Digital Video”, IEEE Trans. Multimedia, Vol. 5, No. 1, 118-128, March 2003.

0006Embedding data in a carrier data is known. For example, U.S. Pat. No. 6,023,511, “Cryptosystem for encrypting digital image or voice file”, describes embedding data representing encrypted image in the low bits of a camouflage image.

SUMMARY OF THE INVENTION

0007The present invention is directed to image encryption methods and apparatuses that substantially obviate one or more of the problems due to limitations and disadvantages of the related art.

0008An object of the present invention is to provide an image encryption method that encrypts images using encryption keys.

0009Another object of the present invention is to provide devices and methods for secure handling of digital images using an image encryption method.

0010Additional features and advantages of the invention will be set forth in the descriptions that follow and in part will be apparent from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.

0011To achieve these and other advantages and in accordance with the purpose of the present invention, as embodied and broadly described, the present invention provides a method for encrypting an input image, the input image comprising a plurality of pixels, each pixel having a pixel value of a plurality of bits, the method including the following steps: (a) dividing the input image into a plurality of pixel groups, each pixel group comprising one or more pixels of the input image; (b) shifting the location of the pixel groups according to an encryption key; and (c) reordering the bits of the pixel values of the input image according to an encryption key, whereby an encrypted image is generated. The method may further include embedding the encrypted image in a cover image, the cover image having a plurality of pixels each having a pixel value of a plurality of bits, the embedding step including replacing less significant bits of the pixel values of the cover image with bits of the pixel values of the encrypted image.

0012More specifically, step (b) in the above method may include: (b1) generating from the encryption key a bit sequence having a number of bits; (b2) indexing the pixel groups according to the number of bits in the bit sequence; (b3) defining a shifting pattern based on a shifting rule and bit values of the bit sequence; and (b4) shifting the pixel groups according to the shifting pattern. Step (c) in the above method may include: (c1) generating from the encryption key a bit sequence having a number of bits; (c2) indexing the bits of pixel values according to the number of bits in the bit sequence; (c3) defining a shifting pattern based on a shifting rule and bit values of the bit sequence; and (c4) shifting the bits of the pixel values according to the shifting pattern.

0013In another aspect, the present invention is directed to devices and computer program code capable of performing the above-described methods.

0014In yet another aspect, the present invention provides various applications of an image encryption method in secure handling of images. One such application is a secure scanning method, which includes: generating a digital image of a document; obtaining an encryption key; encrypting the digital image of the document using the encryption key to generate an encrypted digital image; obtaining a digital cover image; embedding the encrypted image in the cover image; and transmitting the cover image with the encrypted image embedded therein to a receiver. A secure printing method includes: receiving a cover image with an encrypted image embedded therein; obtaining an encryption key; extracting the encrypted image from cover image; decrypting the encrypted image using the encryption key to obtain the digital image of the document; and printing or storing the digital image of the document. The invention is also directed to scanners and printers capable of performing the above described secure scanning and secure printing methods.

0015It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory and are intended to provide further explanation of the invention as claimed.

BRIEF DESCRIPTION OF THE DRAWINGS

0016<figref idref="DRAWINGS">FIG. 1</figref> shows an image encryption and embedding method according to an embodiment of the present invention.

0017FIGS. <b>2</b>(<i>a</i>)-(<i>b</i>) illustrates an example of the block shifting step of the image encryption method.

0018<figref idref="DRAWINGS">FIG. 2</figref>(<i>c</i>) illustrates an example of the pixel set shifting step of the image encryption method.

0019<figref idref="DRAWINGS">FIG. 3</figref> illustrates an example of the bit shifting step of the image encryption method.

0020<figref idref="DRAWINGS">FIG. 4</figref> show an image after various stages of the image encryption process.

0021<figref idref="DRAWINGS">FIG. 5</figref> show a cover image before and after an encrypted image is embedded.

0022<figref idref="DRAWINGS">FIG. 6</figref> shows an image decoding and decryption method according to an embodiment of the present invention.

0023<figref idref="DRAWINGS">FIG. 7</figref> shows a secure scanning method according to an embodiment of the present invention.

0024<figref idref="DRAWINGS">FIG. 8</figref> illustrates the operation of a scanner that implements the secure scanning function.

0025<figref idref="DRAWINGS">FIG. 9</figref> illustrates the operation of a printer that implements a secure printing function according to an embodiment of the present invention.

0026<figref idref="DRAWINGS">FIG. 10</figref> shows a scanner, a printer, or a copier/printer/scanner “multifunction” device implementing a secure scanning or printing function according to an embodiment of the present invention.

0027<figref idref="DRAWINGS">FIG. 11</figref> illustrates a secure ID scanning device according to an embodiment of the present invention.

0028<figref idref="DRAWINGS">FIG. 12</figref> illustrates a secure check reader according to an embodiment of the present invention.

0029<figref idref="DRAWINGS">FIG. 13</figref> illustrates a remote picture ID verification system according to an embodiment of the present invention.

0030<figref idref="DRAWINGS">FIG. 14</figref> illustrates a remote document submitting and receiving system according to an embodiment of the present invention.

0031<figref idref="DRAWINGS">FIG. 15</figref> illustrates a multi-function digital service terminal according to an embodiment of the present invention.

0032<figref idref="DRAWINGS">FIG. 16</figref> illustrates a method of checking the fidelity of an encrypted image using a digital signature according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

0033An image encryption method according to embodiments of the present invention is described with reference to <figref idref="DRAWINGS">FIGS. 1-3</figref>. The encryption method is described using a four-digit encryption key “5511” for illustration purposes only, and those skilled in the relevant art will recognize that the method can be used with longer or shorter encryption keys.

0034As shown in <figref idref="DRAWINGS">FIG. 1</figref>, the image encryption method includes a sub-block shifting step <b>101</b>, a pixel set shifting step <b>102</b>, and a bit shifting step <b>103</b>. The original image to be encrypted (the input image) is a bitmap image, either color or gray scale, where each pixel has a pixel value of, for example, 0 to 255 (8 bit pixel value). The sub-block shifting and pixel set shifting steps <b>101</b> and <b>102</b> shift the pixel locations across the image, and the bit shifting step <b>103</b> shifts (reorders) the bits within each pixel value. In the sub-block shifting step <b>101</b> (see <figref idref="DRAWINGS">FIG. 2</figref>(<i>a</i>)), the image is divided into a plurality of blocks each having N×M pixels. Note that when the image is said to be divided into a plurality of blocks, it is understood that an integer number of blocks may not completely cover the image. For example, the row dimension of the image may not be divisible by N. Thus, it should be understood that an image may be divided into a plurality of blocks with a number of rows or columns at the edge that are not a part of any blocks. These extra rows or columns may be treated differently than the rest of the blocks, or they may be omitted, or blank rows or columns may be added to form complete blocks. N and M may be any appropriate values so long as they permit divisions of a block into a requisite number of sub-blocks (see below). Larger N and M numbers will result in pixels being shifted across a larger distance on the image, while smaller N and M numbers will make the shifted image more granular. <figref idref="DRAWINGS">FIG. 2</figref>(<i>a</i>) shows one such block <b>201</b> having 8×8 pixels.

0035In the sub-block shifting step <b>101</b>, each block is divided into four equal sized sub-blocks. In the example shown in <figref idref="DRAWINGS">FIG. 2</figref>(<i>a</i>), the block <b>201</b> is divided into four 4×4 quadrants <b>202</b><i>a</i>, <b>202</b><i>b</i>, <b>202</b><i>c </i>and <b>202</b><i>d</i>, but other divisions may be used, including four equal sized vertical strips or four equal sized horizontal-strips. The positions of the four sub-blocks are shifted in a manner according to one or more digits of the encryption key. The first digit (“5”) is used in the illustrated example. The digit(s) is expressed as a binary number (a bit sequence), here (0101)₂, and the bits of the bit sequence are shifted by sequentially moving all of the “1” bits to higher (more significant) positions and all of the “0” bits to lower (less significant) positions as shown in <figref idref="DRAWINGS">FIG. 2</figref>(<i>b</i>). In this example, the bit at position <b>2</b> (“1”) is shifted to position <b>1</b>, the bit at position <b>4</b> (“1”) is shifted to position <b>2</b>, the bit at position <b>1</b> (“0”) is shifted to position <b>3</b>, and the bit at position <b>3</b> (“0”) is shifted to position <b>4</b>. The four sub-blocks <b>202</b><i>a</i>-<b>202</b><i>d </i>are indexed as #1 to #4 corresponding to the four bit positions of the bit sequence in <figref idref="DRAWINGS">FIG. 2</figref>(<i>c</i>), and their locations are shifted according to how the bits of the bit sequence are shifted. Thus, sub-block #1 (<b>202</b><i>a</i>) is shifted to the location of sub-block #3 (<b>202</b><i>c</i>); sub-block #2 (<b>202</b><i>b</i>) is shifted to the location of sub-block #1 (<b>202</b><i>a</i>); sub-block #3 (<b>202</b><i>c</i>) is shifted to the location of sub-block #4 (<b>202</b><i>d</i>); and sub-block #4 (<b>202</b><i>d</i>) is shifted to the location of sub-block #2 (<b>202</b><i>b</i>). The resulting sub-block shifted block <b>201</b>' is shown in <figref idref="DRAWINGS">FIG. 2</figref>(<i>a</i>).

0036The number of sub-blocks in each block depends on the number of digits from the encryption key used for the sub-block shifting. For example, using two digits from the encryption key gives an 8-bit bit sequence, the bits of which are re-arranged according to a similar rule as that shown in <figref idref="DRAWINGS">FIG. 2</figref>(<i>b</i>). This allows the block <b>201</b> to be formed by eight sub-blocks which are then shifted among themselves.

0037In the pixel set shifting step <b>102</b> (see <figref idref="DRAWINGS">FIG. 2</figref>(<i>c</i>)), the image (after processing by the sub-block shifting step <b>101</b>) is divided into blocks of N'×M' pixels, which may or may not be the same size as the blocks in the sub-block shifting step. <figref idref="DRAWINGS">FIG. 2</figref>(<i>c</i>) shows an 8×8 pixel block <b>203</b>. Each block is divided into pixel sets as follows. Starting from the upper left hand corner, counting individual pixels from left to right in each row (with wrap-around to the next row) and from the top row to the bottom row, every n pixels is grouped into a pixel set n may be an appropriate fixed value, or a value derived from the encryption key. In the example shown in <figref idref="DRAWINGS">FIG. 2</figref>(<i>c</i>), n is derived by taking the middle four bits of the 8-bit sequence representing the first two digits of the encryption key (“55”). The 8-bit bit sequence is 01010101 and the middle four bits are (0101)₂, which is 5. Thus, n=5 and the five-pixel pixel sets are <b>204</b><i>a</i>, <b>204</b><i>b</i>, <b>204</b><i>c</i>, <b>204</b><i>d</i>, . . . as shown in <figref idref="DRAWINGS">FIG. 2</figref>(<i>c</i>). Within every four consecutive pixel sets (e.g. <b>204</b><i>a</i>-<b>204</b><i>d </i>in <figref idref="DRAWINGS">FIG. 2</figref>(<i>c</i>)), the positions of the pixel sets are shifted with respect to each other in a manner according to a digit of the encryption key, which is the second digit (“5”) in the illustrated example. The shifting method is similar to that used in the sub-block shifting step: First, the digit from the encryption key is expressed as a 4-bit bit sequence, and the digits in the bit sequence are rearranged by moving all of the “1” bits to higher positions and all “0” bits to lower positions in a similar way as shown in <figref idref="DRAWINGS">FIG. 2</figref>(<i>b</i>). Then, the four pixel sets (indexed as #1 to #4) are shifted according to how the bits in the 4-bit bit sequence are shifted. Thus, pixel set <b>204</b><i>a </i>is shifted to the position of pixel set <b>204</b><i>c</i>; pixel set <b>204</b><i>b </i>is shifted to the position of pixel set <b>204</b><i>a</i>; pixel set <b>204</b><i>c </i>is shifted to the position of pixel set <b>204</b><i>d</i>; and pixel set <b>204</b><i>d </i>is shifted to the position of pixel set <b>204</b><i>b</i>. The next four pixel sets <b>204</b><i>e</i>-<i>g </i>are shifted among themselves in a similar manner. The resulting pixel set shifted block <b>203</b>' is shown in <figref idref="DRAWINGS">FIG. 2</figref>(<i>c</i>).

0038The number of pixel sets that are shifted among themselves is determined by the number of digits from the encryption key used for the pixel set shifting. For example, using two digits from the encryption key gives an 8-bit bit sequence, which allows eight pixel sets (n pixels per set) to be shifted among themselves.

0039The bit shifting step <b>103</b> shifts (reorders) the bits within each pixel value. In a preferred embodiment, the pixel value is 8 bit, and may represent one of the color components in a color image or the gray scale value in a gray scale image. The bits of the pixel value are shifted according to the last two digits of the encryption key in the illustrated example, as shown in <figref idref="DRAWINGS">FIG. 3</figref>. In this example, the pixel value is 3, or (00000011)₂. The last two digits of the encryption key, “11” in this example, are expressed as an 8-bit bit sequence, 4 bits from each digit. The key bit sequence is 00010001 in this example (Row (b) in <figref idref="DRAWINGS">FIG. 3</figref>). Row (a) in <figref idref="DRAWINGS">FIG. 3</figref> designates the bit positions <b>1</b>-<b>8</b> for ease of reference. The bits of the key bit sequence (Row (b)) are shifted by moving all of the “1” bits to higher (more significant) positions and all “0” bits to lower (less significant) positions, and shown in Rows (b)-(d) of <figref idref="DRAWINGS">FIG. 3</figref>. The binary pixel value is bit-reversed, i.e., the least significant bit becomes the most significant bit, the second least significant bit becomes the second most significant bit, etc. Then the bits of the bit-reversed pixel value are shifted in the same way the bits in the key bit sequence (Row (b)) are shifted, as illustrated in Rows (e)-(g) of <figref idref="DRAWINGS">FIG. 3</figref>. Row (g) is the output pixel value after bit shifting, here having a value of 0×30.

0040Computationally, the above bit shifting of the pixel value may be accomplished as follows. Let the bit sequence of the pixel value be (p₇p₆p₅p₄p₃p₂p₁p₀) and let the bit sequence derived from the key be (k₇k₆k₅k₄k₃k₂k₁k₀). Form two 16-bit bit sequences: <tables id="TABLE-US-00001" num="1"><table frame="none" colsep="0" rowsep="0" tabstyle="monospace"><tgroup align="left" colsep="0" rowsep="0" cols="3"><colspec colname="1" colwidth="49PT" align="left" /><colspec colname="2" colwidth="168PT" align="left" /><colspec colname="3" colwidth="0PT" align="left" /><tbody valign="top"><row><entry /></row><row><entry>p-sequence:</entry><entry>p₀p₁p₂p₃p₄p₅p₆p₇p₀p₁p₂p₃p₄</entry><entry /></row><row><entry></entry></row><row><entry /><entry>p₅p₆p₇</entry></row><row><entry>and</entry></row><row><entry></entry></row><row><entry>k-sequence:</entry><entry>k₇k₆k₅k₄k₃k₂k₁k_{0 (1-k}₇) (1-k₆)</entry></row><row><entry></entry></row><row><entry /><entry>(1-k₅) (1-k₄) (1-k₃) (1-k₂) (1-k₁)</entry></row><row><entry></entry></row><row><entry /><entry>(1-k₀).</entry></row></tbody></tgroup></table></tables><br /> Then, the bits in the p-sequence that correspond in position to all “1” bits in the k-sequence are taken in that order and form the 8-bit value which is the output pixel value. Note that similar computation method can be applied to the shifting illustrated in <figref idref="DRAWINGS">FIG. 2</figref>(<i>b</i>) as well.

0041The key bit sequence in the above bit-shifting example has 8 bits, which correspond to the 8-bit pixel values. Alternatively, longer or shorter key bit sequences may be used. For example, if a 4-bit sequence is used, then the 8 bits in a pixel value may be divided into two 4-bit groups and shifted. If a 16-bit sequence is used, then two 8-bit pixel values may be combined into 16 bits and shifted.

0042Although FIGS. <b>2</b>(<i>b</i>) and <b>3</b> illustrate particular rules by which the bits in the bit sequence of the key are shifted, other shifting rules may be used. For example, all of the “0” bits may be shifted to higher order positions, or additional shifting may be done, etc. The shifting pattern (i.e. which bit is shifted to which position) depends on the shifting rule and the bit values of the original bit sequence. It should be understood that while FIGS. <b>2</b>(<i>b</i>) and <b>3</b> illustrate shifting of the bits of the key bit sequence, the key bit sequence may not actually be shifted in the computation; these figures merely illustrate how a shifting pattern may be defined by the shifting rule and the bit values of the key bit sequence. A variety of computational methods may be used to implement the shifting.

0043<figref idref="DRAWINGS">FIG. 4</figref> shows an original image <b>401</b>, an image <b>402</b> after sub-block shifting, an image <b>403</b> after pixel set shifting <b>403</b>, and an image <b>404</b> after bit shifting of pixel values (i.e. the encrypted image). It can be seen that sub-block shifting achieves a certain level of image scrambling, and pixel set shifting achieves a higher level of image scrambling, while the scrambled image may still have some structure. The final bit shifting step results in a bitmap image that looks like white noise. The images shown in <figref idref="DRAWINGS">FIG. 4</figref> were processed with an encryption key “123456”. The first two digits (“12”) were used to perform sub-block shifting; the next two digits (“34”) were used to perform pixel set shifting; and the last two digits (“56”) were used to perform bit shifting of the pixel values. In the sub-block shifting step, the image (1150×1450 pixels) was divided into sub-blocks of 115×50 pixels each (thus 10×29 sub-blocks in total), and every eight consecutive sub-blocks in the same row (with wrap-around to the next row) formed a block. Within each block, the eight sub-blocks were shifted using the key “12”. In the pixel-set shifting step, the entire image was treated as a block. Each pixel set consisted of 50×1 pixels, and every eight consecutive pixel sets were shifted among themselves using the key “34”.

0044It can be seen that the sub-block shifting step <b>101</b> and pixel set shifting step <b>102</b> are similar in that they both involve shifting the locations of pixel groups within the image according to an encryption key. The pixel groups are the sub-blocks in the sub-block shifting step and the pixel sets in the pixel set shifting step. Both steps may be described as a spatial shifting process in which the shifting of the pixel group is determined by an encryption key. Note that the term “shifting” does not imply that the pixel groups maintain their shape after shifting. In the pixel set shifting step <b>102</b>, for example, the shape of the pixel set <b>204</b><i>b </i>is changed after the shifting it is shifted to a different location. Also, the term “group” does not imply contiguous pixels within the pixel groups, as similarly demonstrated in the pixel set shifting step <b>102</b>.

0045Described broadly, an encryption method according to embodiments of the present invention takes a bitmap image and generates a bitmap image of the same size that looks like white noise. The method includes one or more spatial shifting steps and a bit shifting step. Spatial shifting involves dividing the image into a plurality of pixel groups, and shifting the locations of the pixel groups according to an encryption key. Spatial shifting may be performed a number of times, and different pixel group definition and shifting pattern may be used each time. Bit shifting shifts (reorders) the bits of each pixel value according to an encryption key. Bit shifting may be performed before, after, or concurrently with spatial shifting (e.g., performing spatial shifting and bit shifting for one area of the image, and then another area of the image). In this regard, the method claims should be understood to require no particular order of the method steps unless an order is expressly recited. For example, when a method step is said to be performed on the input image, it does not imply that that step must be performed as the first step in the claimed method.

0046The encrypted image <b>404</b> can now be embedded in a cover image (carrier image) to be transmitted (see <figref idref="DRAWINGS">FIG. 1</figref>, step <b>104</b>). Embedding can be done in a variety of ways. For example, each 8-bit pixel value in the encrypted image may be broken into four 2-bit sequences, and each 2-bit sequence may be embedded in the two lowest (least significant) bits of the pixel values of the cover image. In other words, the two lowest bits of the pixel values of the cover image are replaced by two bits of the encrypted image. Note that such embedding requires the cover image to be at least four times the size (in terms of the number of pixels) of the encrypted image. Alternatively, each pixel in the encrypted image may be broken into two 4-bit sequences and embedded in the lower four bits of the pixel values of the cover image. This requires a cover image of at least two times the size of the encrypted image, but the resulting cover image with the encrypted image embedded therein may have poor quality. Alternatively, in a lossy embedding method, the six more significant bits of each pixel of the encrypted image may be broken into three 2-bit sequence and embedded in the lower two bits of three of every four pixels of the cover image. Other variations are also possible.

0047<figref idref="DRAWINGS">FIG. 5</figref> shows an original cover image <b>501</b> and a cover image <b>502</b> after the encrypted image <b>404</b> is embedded in the lower 2 bits of the cover image.

0048The embedded image can be extracted from the cover image and decrypted (decoded) to recover the original image in a process that is the reverse of the encrypting and embedding method, as shown in <figref idref="DRAWINGS">FIG. 6</figref>. In other words, the lower 2 bits of the cover image are extracted and the 8-bit pixel values of the encrypted image are reconstructed (step <b>601</b>). The bits of the pixel values are reverse-shifted in a manner that is the reverse of the bit shifting in the encryption process (reverse bit shifting step <b>602</b>). The pixel sets and sub-blocks are then reverse-shifted in a manner that is the reverse of the pixel set and sub-block shifting in the encryption process (reverse pixel set shifting step <b>603</b> and reverse sub-block shifting step <b>604</b>). This gives the original image without any loss of information (unless a lossy embedding method is used). Further detailed description of image recovery method is unnecessary.

0049The image encryption method described above can be applied to encrypt color and gray scale images.

0050The encryption keys in the examples given above contain only numbers. Alternatively, they may contain both numbers and letters, which may then be converted to 4-bit, 8-bit, or other suitable bit sequences for use in the encryption. The encryption key may be inputted by a user (such as the user of a scanner or printer in the secure scanning and printing application described later), or it may be supplied by the encryption program without involving the user, such as by using a random number generator. The encryption key may be transmitted to the receiver separately from the encrypted and embedded image. Alternatively, as shown in <figref idref="DRAWINGS">FIG. 1</figref> (step <b>105</b>), the encryption key may be embedded in the cover image in a pre-defined manner, such as by embedding it in one or more corners of the cover image. The encryption key may also be conveyed in the cover image using other methods. In one example, a 128-bit encryption key is converted to four sets of CMYK values representing four colors, and the four colors are added to the cover image as color patches. Alternatively, other decorative elements that encode the key may be added in the cover image.

0051Image encryption and embedding method has wide applications, some of which are described with reference to <figref idref="DRAWINGS">FIGS. 7-15</figref>.

0052Secure Scanning and Secure Printing

0053Scanning is a process that generates a digital copy of an input image in print form. In a secure scanning process according to embodiments of the present invention, an input image (the secret image) is scanned into a digital form, encrypted using an image encryption method (optional), and embedded in a cover image. The image encryption method may be the ones described earlier in this disclosure, or any other suitable image encryption method; or the secret image may be embedded in the cover image without having been encrypted. Preferably, the scanner only outputs the cover image with the encrypted image embedded therein for transmission or saving to a memory, and does not output an unencrypted digital form of the secret image or a cover image without the encrypted image embedded therein. As shown in <figref idref="DRAWINGS">FIG. 7</figref>, the secret image is typically scanned in (step <b>701</b>), while the cover image may be either scanned in (step <b>702</b>) or obtained as an existing digital image, which may be either pre-stored in the scanner (step <b>704</b>) or obtained from an external device connected to the scanner (step <b>703</b>). Typically, the cover image is larger in size (in terms of the number of pixels) than the secret image in order to embed the encrypted image. Thus, if both images are being scanned in and have similar physical sizes, the cover image will be scanned in at a higher resolution. Thus, steps of checking the input image size (steps <b>705</b> and <b>706</b>) may be performed.

0054<figref idref="DRAWINGS">FIG. 8</figref> illustrates the operation of a scanner that implements the secure scanning function. After selecting the Secure Scan function (step <b>801</b>), the user selects the embedding method (step <b>802</b>) (optional; a default embedding method may be used) and input the encryption key (step <b>803</b>). The cover image is scanned in or obtained as digital data from the scanner memory or an external source (step <b>804</b>). The secret image is canned in (step <b>805</b>). The scanner than performs an image encryption (optional) and embedding process (step <b>806</b>) to embed the secret image into the cover image. The encryption key may also be embedded in the cover image (step <b>807</b>). The cover image is then transmitted or saved (step <b>808</b>).

0055A secure printing function decodes and decrypts an input cover image that has an encrypted secret image embedded therein and prints the decrypted secret image. The input cover image may originate from a scanner that implements a secure scanning function, or a computer that implements an image encryption method, etc. Again, the image may have been encrypted with a method described earlier in this disclosure or any other suitable method, or it may be merely embedded but not encrypted, in which case decryption will be unnecessary. Optionally, before printing the decrypted image, the printer may add a visible watermark to the image which is intended to warn against or prevent further distributing. <figref idref="DRAWINGS">FIG. 9</figref> illustrates the operation of a printer that implements the secure printing function. After selecting the secure printing function (step <b>901</b>), the user inputs the encryption key (step <b>902</b>) that has been used to encrypt the image. Optionally, if the encryption key is also embedded in the cover image, the encryption key is verified by comparing the inputted key with the embedded key (step <b>903</b>). The embedded image is extracted from the cover image and decrypted to obtain the secret image (step <b>904</b>). Watermark is optionally added (step <b>905</b>) and the secret image is printed (step <b>906</b>).

0056<figref idref="DRAWINGS">FIG. 10</figref> shows a scanner, a printer, or a copier/printer/scanner “multifunction” device <b>1000</b> that includes an image encryption section <b>1001</b> that implements image encryption and embedding and/or image recovery. The image encryption section <b>1001</b> may be implemented as dedicated hardware logic, dedicated processor, or firmware or software executed by the CPU of the scanner. The other components of the device—scanning section <b>1002</b>, printing section <b>1003</b>, control section <b>1004</b>, memory <b>1005</b>, network interface <b>1006</b>, and input section <b>1007</b>—are conventional. A scanner only or printer only device may only have a subset of these components.

0057Secure ID Scanning Device

0058<figref idref="DRAWINGS">FIG. 11</figref> illustrates a dedicated scanning device for securely scanning ID cards or similar documents and securely transmitting the scanned images through a network such as the Internet. The secure ID scanning device <b>1100</b> has a scanning section <b>1101</b> for scanning an ID card or similar document <b>1110</b> and generating a digital image, an encryption section <b>1102</b> for encrypting the scanned image, a selection section <b>1103</b> for obtaining a cover image (which may be selected by a user or by the scanning device from its database, or a default cover image may be used), and an embedding section <b>1104</b> for embedding the encrypted image in the cover image. An encryption key may be inputted by the user or supplied by the scanning device. The ID scanning device <b>1100</b> may be connected to a computer <b>1120</b> via a wired or wireless connection <b>1130</b>, such as a USB connection. Alternatively (not shown), the ID scanning device may be directly connected to the Internet and transmit the encrypted and embedded images to a destination via the Internet. The ID scanning device <b>1100</b> may optionally provide normal (non-secured) scanning of documents. Such ID scanning devices may be installed at businesses or other public places to facilitate secure business transactions.

0059Secure Check Reader

0060Although electronic payment methods are pervasive today, personal checks are still being widely used at point-of-sale locations. Shoppers need to carry their checkbooks and shops need to process many checks. A dedicated check reader may be used to facilitate and simplify transactions involving personal checks. As shown in <figref idref="DRAWINGS">FIG. 12</figref>, the check reader <b>1200</b> is connected to a cash register or computer <b>1220</b> to provide check reading and processing functions. First, a scanning section <b>1201</b> scans a user's blank personal check <b>1210</b>, and a signature and data input section <b>1202</b> allows the user to input relevant data and sign his or her name. The input data may include the amount of the payment, the check number, the payee's name, a short message to be included on the check (the “For” line), etc. A cash register data input section <b>1206</b> may import certain data automatically from the cash register, such as the amount of the payment, the payee's name, etc. Then a check image generation section <b>1204</b> combines the image of the blank check, the relevant input data and the user's signature to generate an image of the check. An encryption and embedding section <b>1205</b> encrypts the image of the check and embeds it in a cover image provided by a cover image section <b>1203</b>. The cover image having the embedded image is then transmitted by the computer or cash register <b>1220</b> to the intended bank via the Internet. With such a check reader, a shopper can use the familiar check-writing method for payments, but only needs to carry one blank personal check. Processing of the checks at banks becomes easier because it reduces manual handling of checks and because part of the information associated with the check is already in the digital form.

0061Remote Picture ID Verification System

0062<figref idref="DRAWINGS">FIG. 13</figref> shows a remote picture ID verification system useful for verifying the identity of an ID card holder. At a sending terminal <b>1300</b>, a scanner <b>1301</b> scans a picture ID card presented by the ID card holder. At the same time, a camera <b>1302</b> takes a picture of the face of the ID card holder. A coding section <b>1303</b> encrypts the scanned ID card image and embeds it into the picture of the ID card holder (the cover picture). The cover picture with the embedded information is transmitted through a network such as the Internet to a remote receiving terminal <b>1330</b>. A decoding section <b>1331</b> of the receiving terminal extracts the embedded image of the ID card <b>1340</b> from the cover picture and decrypts it. A human operator may then view the decrypted image of the ID card and compare the picture on the ID card to the picture of the card holder (the cover picture) to verify the identify of the card holder. Alternatively (as shown in <figref idref="DRAWINGS">FIG. 13</figref>), a photo area extraction section <b>1332</b> extracts the picture on the ID card and a face recognition section <b>1333</b> of the receiving terminal automatically compares the cover picture with the picture extracted from the ID. The face recognition section <b>1333</b> may be implemented by appropriate pattern recognition algorithms.

0063Remote Document Submitting and Receiving System.

0064The secure scanning and secure printing functions may be combined to provide a remote document submitting and receiving system shown in <figref idref="DRAWINGS">FIG. 14</figref>. The submitting terminal <b>1400</b> has a scanner <b>1401</b> that scans the document and a coding section <b>1403</b> that encrypts the scanned image an embeds it in a cover image. The submitting terminal <b>1400</b> optionally has a camera <b>1402</b> that captures a digital image that can be used as the cover image. The receiving terminal <b>1410</b> has a decoding section <b>1411</b> that extracts the embedded image and decrypts it to produce the original scanned image (document) that may be displayed by a display <b>1413</b> or printed by a printing section <b>1414</b>. An optional customized watermark embedding section <b>1412</b> adds a watermark to the decrypted image before display or printing.

0065The submitting terminals of the remote document submitting and receiving system may be installed at businesses or other public places. The submitting terminals may also be a part of a multi-function digital service terminal, which may provide services such as digital document submitting, digital photography, scanning, printing, copying, Internet access, etc. as shown in <figref idref="DRAWINGS">FIG. 15</figref>.

0066The implementation of the above secure scanning device, secure printing device, secure ID scanning device, secure check reader, remote picture ID verification system, and remote document submitting and receiving system, although not described in extensive detail, is within the capabilities of those skilled in the relevant art based on the description given in this disclosure. It should be noted that these devices and systems may use the image encryption and embedding method described in this disclosure, or any other suitable image encryption and embedding method.

0067A digital signature feature may optionally be included to enhance the security of the above-described applications or other applications of an image encryption method. Digital signature is a known technique used in data transmission. For example, digital signature technique can be used to indicate whether the document is authentic or has been altered. A digital signature is a bit stream generated based on the content of the input document. In an embodiment shown in <figref idref="DRAWINGS">FIG. 16</figref>, a digital signature is generated from the secret image (step <b>1603</b>) at the transmitting side <b>1600</b>. The digital signature may be added to the secrete image (step <b>1604</b>), such as by adding the digital signature to a fixed location or locations of the secret image. Alternatively, the digital signature may be transmitted to the receiving side separately (step <b>1607</b>), such as by electronic mail. The secret image (with or without the added digital signature) is encrypted (step <b>1605</b>) and embedded in the cover image (step <b>1606</b>). As another alternative, the digital signature may be embedded in the cover image directly instead of being added to the secret image and encrypted as a part of the secret image. The cover image with the encrypted image embedded therein is transmitted via a network such as the Internet. At the receiving side <b>1610</b>, the received image is extracted and decrypted (step <b>1611</b>) to obtain the decrypted secret image. The digital signature is obtained depending on how it has been transmitted: it may be extracted from the decrypted image (step <b>1612</b>); it may be directly extracted from the cover image; or it may be separately received from the transmitting side (step <b>1613</b>) such as by electronic mail. The digital signature is then used to check the fidelity of the secret image (step <b>1613</b>). This is done by comparing such digital signature with a digital signature re-generated from the decrypted image. This step will reveal whether the secret image has been altered.

0068Another variation of the encryption method described in this disclosure is to use the digital signature to generate the encryption key. A digital signature generated from the secret image may be encrypted using a suitable method such as ACE. A bit sequence (e.g., 64-bit or 128-bit) is then derived from the encrypted digital signature and used as the encryption key. An advantage of generating the encryption key from the digital signature is that the encryption key does not need to be separately inputted by the user.

0069It will be apparent to those skilled in the art that various modification and variations can be made in the image encryption an embedding method and apparatus of the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention cover modifications and variations that come within the scope of the appended claims and their equivalents.