JP4308715B2 - Image processing apparatus, image processing method, and image processing program - Google Patents

Description

The present invention relates to an image processing apparatus, an image processing method, and an image processing program. Specifically, information transmitted from an image data is embedded in the image data itself, and the standard format is input. the image processing apparatus in the image processing system having a reception-side image processing apparatus of the primary objects and secondary objects to restore the image data itself based on the information extracted from the image data, image processing method, and an image processing program.

  In general, in an image processing system configured by a transmission-side image processing device that reads a paper document or the like with a scanner and transmits the electronic image data, and a reception-side image processing device that receives and visualizes the transmitted image data, 2. Description of the Related Art A receiving-side image processing apparatus is known that includes a primary-purpose receiving-side image processing apparatus and a secondary-purpose receiving-side image processing apparatus. However, the receiving side image processing apparatus of this system is assumed to be a monitor or the like for the primary purpose (main purpose) and a printer or the like for the secondary purpose (secondary purpose). In the transmission-side image processing apparatus, conversion (mainly reducing the specification value) that matches the basic specifications (resolution, number of colors, sampling frequency) of the scanned image data with the primary target device (monitor, etc.) Conversion) is performed, and the converted data is transmitted.

  Here, the basic image specification value necessary and sufficient for displaying an image with appropriate quality in the image processing apparatus on the receiving side is actually slightly different for each device. For example, regarding the resolution, the printer is more than the normal monitor. Requires large specification values. As described above, when the secondary-purpose reception-side image processing device requires a larger specification value than the primary-purpose reception-side image processing device, as described above, the transmission-side image processing device uses the primary-purpose reception-side image processing device. If conversion is performed to reduce the specification value according to the image processing apparatus, the image quality of the secondary-purpose image processing apparatus cannot be guaranteed. This is because high-quality image data having a large specification value cannot be restored again from image data whose specification value has been once reduced. That is, the reduced information cannot be restored again.

  Even in the above-described case, two image data conversion methods for guaranteeing image quality with a secondary-purpose image processing apparatus are conventionally known. One is a conversion method (first conventional method) in which the basic specifications are adapted to the secondary-purpose image processing apparatus, and the other is image data obtained by conversion in which the basic specifications are adapted to the primary-purpose image processing apparatus, This is a method (second conventional method) in which high-quality image data is restored by conversion into two files of information data to be deleted by conversion and using both data.

  However, the first conventional method has a drawback in that a load for converting the basic specifications of the image data is generated on the primary target device side. In addition, the second conventional conversion method has the disadvantage that it is complicated due to two files.

  Patent Document 1 is known as a similar prior art. In Patent Document 1, a transmission-side image processing device embeds a histogram of electronic watermark information in image data, a reception-side image processing device extracts this, and attempts to obtain high image quality by color processing based on this information. This is intended to reduce the load on the receiving side by making the load on the electronic watermark information histogram only on the transmitting side. However, in this method, although the file can be made one, only the information that can be generated from the received image data is embedded (the same thing is included twice), so the image quality in the secondary target device is guaranteed. I can't do it.

JP 2001-127985 A

  In order to solve the above problems, the image processing apparatus of Japanese Patent Application No. 2002-358430 of the applicant of the present application provides necessary and sufficient images for both the primary purpose (main purpose) and the secondary purpose (secondary purpose). Therefore, when the monitor display is the primary purpose and the print output is the secondary purpose, additional image information is embedded in the data converted into the resolution data sufficient for the monitor display. A technique is proposed in which simple image information is extracted and restored (higher resolution) using this information.

  However, since the image processing apparatus of Japanese Patent Application No. 2002-358430 performs uniform processing on the entire image region, it is difficult to say that it is an effective image information addition method. For example, the resolution is like a photographic image. A high resolution is not necessary in a large image area, and a high resolution is required only in a character image area. Therefore, embedding information in the photographic image area increases the image file size.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and when transmitting image data from a transmission-side image processing apparatus to a reception-side image processing apparatus, the reception-side image processing apparatus does not increase the size of the image data. An object is to provide an image processing apparatus, an image processing method, and an image processing program capable of reproducing an image with high image quality.

In order to solve the above-described problems and achieve the object, an image processing apparatus according to an aspect of the present invention includes an attribute determination unit that determines whether image data is a pattern area in units of a predetermined area, and the image data The first conversion means for generating the first image data by performing the process for expanding the color space quantization width, and the second process for performing the process for reducing the color space quantization width for the first image data. Second conversion means for generating the image data, arithmetic means for generating difference data by subtracting the image data and the second image data, and compressing the pattern area of the difference data at a low compression rate And compressing means for compressing another area of the difference data at a high compression rate to generate compressed data, and embedding means for embedding the compressed data in the first image data. To do.

An image processing method according to another aspect of the present invention includes an attribute determination step for determining whether image data is a picture area in units of a predetermined area, and a process for expanding a color space quantization width for the image data. A first conversion step of generating first image data, and a second conversion step of generating second image data by performing a color space quantization width reduction process on the first image data When an arithmetic step of generating differential data by performing a subtraction between the image data and the second image data, the picture area of the difference data is compressed at a low compression ratio, high other areas of the difference data The method includes a compression step of generating compressed data by compressing at a compression rate, and an embedding step of embedding the compressed data in the first image data.

An image processing program according to another aspect of the present invention causes a computer to execute the above-described image processing method.

According to the present invention, a large amount of information related to a picture area that has a large effect of improving the image quality with respect to the number of colors is embedded, and the embedding of information about another area that has a small effect of improving the image quality with respect to the number of colors is reduced. In other words, it is possible to restore high-quality image data without increasing the size of the image data more than necessary. Thus, for example, when sending image data to the reception-side image processing apparatus from the transmission-side image processing apparatus, it becomes possible to efficiently transmit the image data, a high-quality image by receiving image processing apparatus It can be played back.

Hereinafter, an image processing apparatus, an image processing method, and an image processing program according to the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to the embodiments. In addition, constituent elements in the following embodiments include those that can be easily assumed by those skilled in the art or those that are substantially the same.

  FIG. 1 is an overall view showing an example of an image processing system according to the present invention. As shown in FIG. 1, the image processing system includes a transmission-side image processing device 1 that transmits read image data, and a reception-side image processing device that receives and visualizes the transmitted image data via a communication network 3. 5 is comprised. The reception-side image processing device 5 includes a primary-purpose reception-side image processing device 5a and a secondary-purpose reception-side image processing device 5b.

  FIG. 2 is a block diagram illustrating a configuration of the transmission-side image processing apparatus 1 illustrated in FIG. As shown in FIG. 2, the transmission-side image processing apparatus 1 performs a first conversion on the data creation unit 7 that creates digital image data and the digital image data created by the data creation unit 7. A first conversion unit 9 that generates the second image data, a second conversion unit 11 that performs the second conversion on the first image data to generate the second image data, the digital image data and the first A subtraction unit 13 that performs subtraction (operation) on the image data of 2 to generate difference data (operation data), a compression unit 15 that generates compressed data (embedding information) obtained by compressing the difference data, and a first Based on the determination result of the attribute determination unit 35 and the attribute determination unit 35 that determines the image attribute of the digital image data generated by the data generation unit 7, compressed data (embedding information) is applied to the first image data. And an embedded portion 17 to be embedded. .

  The data creation unit 7 is an application software such as a scanner that reads a document such as printed matter / photographic paper, a digital still camera / video camera that captures a space, an X-ray camera, an MRI, a CT scanner, and an RIP / word processor that artificially creates an image It can be composed of wear.

  The attribute determination unit 35 performs image attribute determination for each predetermined area unit (for example, pixel unit or block unit) for the image data generated by the data generation unit 7 and embeds an attribute determination signal S indicating the image attribute. Output to unit 7. FIG. 3 is a diagram illustrating an example of the relationship between the image attribute and the attribute determination signal S. As shown in FIG. 3, the attribute determination signal S is composed of 2-bit data, and is “01” for the “character / line image area”, “10” for the “halftone image area”, and “photographic paper photograph area”. In the case of “11” in the case, “00” in the case of “other areas (including the white background portion)”.

  As a method for attribute determination, a known digital copying machine image area separation unit may be used for a scanner image, and a bit map is used for a RIP image by a command interpreter when converting to raster data. A shape attribute determination map may be created.

  The first conversion unit 9 and the second conversion unit 11 change the basic specifications of the image data. Here, the basic specifications refer to resolution, number of colors, and sampling frequency. The first conversion unit 9 performs first conversion on the digital image data to reduce these specification values (reduction in resolution, reduction in the number of colors, reduction in sampling frequency) to generate first image data. The second conversion unit 11 performs second conversion (higher resolution, increased number of colors, increased sampling frequency) on the first image data to increase the specification values, and the second image data. Is generated. The first conversion unit 9 and the second conversion unit 11 convert at least one of the resolution, the number of colors, and the sampling frequency among the basic specifications.

  In other words, the process for reducing the resolution is an expansion process for the spatial quantization width, and the process for increasing the resolution is a process for reducing the spatial quantization width. In other words, the process for reducing the number of colors is an expansion process for the color space quantization width, and the process for increasing the number of colors is a process for reducing the color space quantization width. In other words, the sampling frequency reduction process is a temporal quantization width expansion process, and the sampling frequency increase process is a temporal quantization width reduction process.

  Each conversion uses a known image processing method. For example, a simple sampling method or an averaging processing method is used for low resolution conversion, and a nearest neighbor method or a weighted average interpolation method is used for high resolution conversion. Use. In the following embodiment, a case where resolution is converted will be described as an example.

  FIG. 4 is an explanatory diagram for explaining the case of converting the resolution. Specific examples of the first conversion and the second conversion in the first conversion unit 9 and the second conversion unit 11 will be described with reference to FIG. In FIG. 4, (a) is a part of the image data, where one glance represents one pixel, and the number in the gutter represents a pixel value (gradation). In this example, 4 × 4 pixels are shown. . An image obtained by subjecting this image to low resolution conversion by the averaging processing method is shown in FIG. Further, images obtained by performing high resolution conversion on the image of (b) are shown in (c) and (d), (c) is the near-less neighbor method, and (d) is the load average method (load). The matrix is shown in (m). Although (c) and (d) are similar to (a), they are not completely identical. This shows an example in which the original resolution image data cannot be completely restored again from only the image data whose resolution has been reduced once. That is, part of the information is reduced when the resolution is reduced, and the reduced information cannot be restored again.

  Here, an example in which the resolution is [÷ 2] and [× 2] has been shown, but the present invention is not limited to this. For example, the data creation unit 7 creates 300 dpi image data, which is first-ordered. When transmitting to a monitor with a target of 75 dpi and a 150 dpi printer for secondary purposes, the resolution is set to [÷ 4] and [× 2]. Further, the resolution is not limited to two-dimensional, and may be three-dimensional.

  The subtractor 13 outputs difference data obtained by subtracting the second image data converted by the second converter 11 from the image data generated by the data generator 7 to the compressor 14. In the example shown in FIG. 4, (a)-(c) = (f) and (a)-(d) = (f) represent subtraction.

  The compressing unit 15 performs quantization / entropy coding on the difference data input from the subtracting unit 13 and outputs the compressed data subjected to compression processing to the embedding unit 17 as embedding information. In FIG. 4, (g) and (h) are obtained by quantizing (÷ 4, truncating the remainder) of (e) and (f). Here, entropy coding is data compression processing such as arithmetic coding, Huffman coding, and universal coding.

  Based on the image attribute determination signal S input from the attribute determination unit 35, the embedding unit 17 replaces the redundant portion of the first image data with the corresponding embedded information using a digital watermark technique. The digital watermark technique is a process of modifying image data based on embedded information within a range where image quality deterioration cannot be perceived. It is a technology that has attracted attention recently, such as embedding copyright information in content images.

  The embedding unit 17 embeds embedding information corresponding to a portion determined to be a character / line drawing area of the first image data when embedding processing of embedding information for improving image quality with respect to resolution is performed. The embedding process is executed, and the embedding process of the corresponding embedding information is not performed for other areas. In addition, if the embedding information is embedded in the entire area of the first image data, an abnormal image is generated in the halftone image area, for example, when the image is reproduced in the receiving image processing apparatus. . Since the halftone image is a gradation image that can express the gradation, only a halftone screen is used to express gradation using ink at the time of printing. That is, in the halftone dot image area, embedding information embedding processing is not necessary for increasing the resolution. If the resolution is increased halfway, a moire pattern is generated due to the location of the quantization error in the compression unit 15. An abnormal image may occur.

  As described above, by embedding the embedding information for increasing the resolution in the character / line image area, a high-quality image can be reproduced when the image is reproduced by the secondary purpose image processing apparatus. Although the example in which the character / line drawing area is determined is shown in the attribute determination unit 35, an edge portion (a portion where the density changes rapidly) such as a boundary portion between the character and the background may be detected. In addition, by embedding embedded information for increasing the resolution in the halftone image area, it is possible to reduce the image data size by not embedding unnecessary information, Image quality deterioration such as moire due to resolution can be suppressed.

  Next, the operation of the transmission side image processing apparatus 1 will be described. The data creation unit 7 creates digital image data. The attribute determination unit 35 determines an image attribute of the digital image data generated by the data generation unit 7 in units of a predetermined area and outputs an attribute determination signal S. The first conversion unit 9 performs first conversion (resolution reduction processing) on the digital image data generated by the data generation unit 7 to generate first image data. The second conversion unit 11 performs second conversion (high resolution processing) on the first image data to generate second image data. The subtraction unit 13 generates difference data (calculation data) by subtracting (calculating) the second image data from the digital image data created by the data creation unit 7. The compression unit 15 generates compressed data (embedded information) obtained by compressing the difference data. The embedding unit 17 embeds compressed data (embedding information) in the first image data based on the attribute determination signal S. Thereafter, image data obtained by converting the first image data in which the embedded information is embedded into a standard format is transmitted to the reception-side image processing device 5 by a format conversion unit (not shown).

  Thus, by converting to a standard format, generality can be given to transmitted / received image data, and use in the receiving image processing apparatus can be guaranteed. Standard formats include international standards such as JPEG, JPEG2000, and MPEG2,4. Since the information is embedded in the image data itself, there is an advantage that the handling of the file is not complicated compared to the second conventional method in which the embedded information is a separate file.

  Here, in the example shown in FIG. 4, the entropy when 16 quantized values are regarded as a non-memory information source is 1.91 bit / pixel in FIG. It becomes 1.31 bits / pixel, and if it is understood by the Markov model, the entropy is even smaller. This temporarily embeds 1 bit / pixel information. If the gradation (number of colors) is 24 bits / pixel, if the image data originally has image quality redundancy of 4/24 = 16%, there is no deterioration in image quality, and this portion can be replaced with embedded information. it can.

  In the above embodiment, the processing of the resolution of the image data has been described. However, when embedding unit 17 performs embedding for improving the image quality with respect to the number of colors, it is determined as a photographic paper photo area or a halftone image area. It is desirable to perform an embedding process on the rendered picture area. Increasing the number of colors for the character / line image area also has an image quality advantage, but the effect is less noticeable than the pattern area. It is preferred to implement.

  In the process for the number of colors, in FIG. 4, the first conversion unit 9 performs quantization for reducing the number of colors (for example, 8 bits → 4 bits), and the second conversion unit 11 performs inverse quantization again (see FIG. 4). 4 bits → 8 bits). The subtractor 13 obtains the image lost in the first converter 9 by taking the difference between the digital image signal generated by the data generator 7 and the second image data converted by the second converter 11. (Gradation) is extracted. The compression unit 15 compresses the data so that the embedding can be easily performed, and the embedding unit 17 executes the embedding process on the image whose number of colors is reduced. The reception-side image processing device 5b performs high-quality reproduction by increasing the number of colors.

  FIG. 5 is a block diagram showing the configuration of the primary-side receiving image processing apparatus 5a. As shown in FIG. 5, the primary-side reception-side image processing device 5 a includes a visualization unit 21. In other words, since image data that meets the specifications of the primary target device is received here, the primary target device does not need to convert image data and is visualized as it is. The load processing of the main target device is reduced compared to the first conventional method. If information embedding is performed within the image quality redundancy range of the main target device, sufficient image quality can be guaranteed with the main target device. The visualization unit 21 visualizes digital image data and is a display device such as a monitor, a projector, or a printer. The system of the present invention is intended to allow a user to use an image displaying this device.

  FIG. 6 is a block diagram showing the internal configuration of the secondary-side receiving image processing apparatus 5b. As shown in FIG. 6, the secondary-side reception-side image processing device 5b includes a second conversion unit 11 that performs second conversion on the image data transmitted from the transmission-side image processing device 5a, 2, an adder 25 connected to the converter 11, an extractor 27 that inputs image data from the transmission-side image processing device and extracts embedded data, and a decompressor connected to the extractor 27 and the adder 25. And a visualization unit 21 connected to the addition unit 25.

  The second conversion unit 11 largely converts the resolution of the image data transmitted from the transmission-side image processing apparatus 1 as a basic specification and outputs the converted data to the addition unit 25. The extraction unit 27 extracts the embedded information from the image data transmitted from the transmission image processing apparatus 1 and outputs it to the decompression unit 19. The decompressing unit 19 restores the extracted embedded information by inverse quantization after entropy decoding. In the example of FIG. 4 above, (g) and (h) are inversely quantized (× 4) as shown in (i) and (g). The addition unit 25 adds the image data converted by the second conversion unit 11 and the embedded information expanded by the expansion unit 29. In the example of FIG. 4, (c) + (i) = (k), (d) + (j) = (l) represents the addition. The visualization unit 21 is a part that visualizes digital image data, and is a display device such as a monitor, a projector, and a printer, and displays the image data added by the addition unit 25.

  Next, the operation of the secondary-side receiving image processing apparatus 5b will be described. The image data transmitted from the transmission-side image processing apparatus 1 via the communication network 3 is, for example, image data whose resolution has been increased by the second conversion unit 11 in order to conform to the specifications of this device (FIG. 4C, ( d)), and embedded information is extracted by the extraction unit 27 from the image data transmitted from the transmission-side image processing apparatus 1 via the communication network 3, and the extracted embedded information is entropy decoded by the decompression unit 29. After (FIGS. 4 (g) and (h)), inversely quantized data (FIGS. 4 (i) and (j)) is obtained. Then, after the high-resolution image data (FIGS. 4C and 4D) and the inverse quantized data (FIGS. 4I and 4J) are added and corrected by the adding unit 25, the visualization unit 21.

  Here, the deviation from the original image (FIG. 4A) (average of absolute values of pixel value differences) is 3.13 in FIG. 3C and 2 in FIG. On the other hand, after addition correction, the value is 0.63 in FIG. 10 (k) and 1.38 in FIG. This means that the image quality is improved by the addition correction.

  That is, a part of the information deleted when the resolution is reduced by the transmission-side image processing apparatus 1 is embedded within a range that does not affect the main target device, and is extracted / used by the secondary target device. Sufficient image quality can be guaranteed even with the next-purpose equipment.

  In the above embodiment, the case where the first conversion unit 9 and the second conversion unit 11 perform resolution conversion has been described. However, the number of colors and the sampling frequency may be converted. In this case, in principle, the same effect as in the above-described resolution example can be obtained only by changing the quantization space.

  FIG. 7 is a block diagram illustrating the configuration of the transmission-side image processing apparatus 1 according to the second embodiment. In the transmission-side image processing device 1 according to the first embodiment, the embedding unit 17 embeds embedding information in the first image data based on the determination result of the image attribute of the attribute determining unit 35, thereby increasing the image data size. It is the structure which makes small. On the other hand, in the second embodiment, the compression unit 15 compresses the difference data calculated by the subtraction unit 13 based on the determination result of the image attribute of the attribute determination unit 35, and the data size of the obtained embedded information It is the structure which makes small.

In FIG. 7, the attribute determination signal S generated by the attribute determination unit 35 is input to the compression unit 15. The compression unit 15 performs a compression process on the difference data input from the subtraction unit 13 by controlling a compression parameter based on the attribute determination signal S. For example, when embedding information for improving the image quality with respect to the resolution, the compression processing is performed so that the compression rate is lower in the character / line image area of the difference data than in other areas. While a large amount of information is embedded, a compression process that increases the compression rate is performed on the other areas of the difference data so that less embedded information is embedded than in the character area. The embedding unit 17 embeds the compressed data (embedding information) generated by the compression unit 15 into the first image data subjected to the first conversion by the first conversion unit 9.

  As a result, embedding information for increasing the resolution can be embedded in the character portion with emphasis, and a higher-quality image can be reproduced by the secondary-purpose receiving-side image processing apparatus. When embedding information is embedded in order to improve the image quality with respect to the number of colors, the compression unit 15 compresses the difference data so that the compression rate is low for the photographic paper photo area and the halftone photo area. The processing is performed so that more embedding information is embedded than in other areas, while the character / line image area of the difference data is subjected to a compression process to increase the compression ratio, Less embedding information is embedded than in a halftone dot photo area.

  Note that the image processing apparatus according to the present invention can be applied to a system including a plurality of devices (for example, a host computer, an interface device, a scanner, a printer, and the like). ) May be applied.

  Another object of the present invention is to supply a recording medium recording a program code of software for realizing the functions of the above-described image processing apparatus to the system or apparatus, and the computer of the system or apparatus (or CPU, MPU, It can also be achieved by the DSP) executing the program code stored in the recording medium. In this case, the program code read from the recording medium itself realizes the functions of the image processing apparatus described above, and the program code or the recording medium storing the program constitutes the present invention. Recording media for supplying the program code include FD, hard disk, optical disk, magneto-optical disk, CD-ROM, CD-R, magnetic tape, non-volatile memory, optical recording medium such as ROM, magnetic recording medium, optical Magnetic recording media and semiconductor recording media can be used.

  Further, by executing the program code read by the computer, not only the functions of the image processing apparatus described above are realized, but also an OS (operating system) operating on the computer based on the instruction of the program code. However, it goes without saying that a case where the function of the image processing apparatus described above is realized by performing part or all of the actual processing.

  In addition, after the program code read from the recording medium is written in a memory provided in a function expansion board inserted in the computer or a function expansion unit connected to the computer, the function expansion is performed based on the instruction of the program code. It goes without saying that the CPU or the like provided in the board or the function expansion unit performs part or all of the actual processing, and the above-described functions of the image processing apparatus are realized by the processing.

As described above, the image processing apparatus, the image processing method, and the image processing program according to the present invention read a paper document or the like with a scanner and transmit the electronic image data, and the transmitted image The present invention can be used in an image processing system configured with a receiving image processing apparatus that receives and visualizes data.

1 is an overall view showing an example of an image processing system according to the present invention. It is an internal block diagram of a transmission side image processing apparatus. It is a figure which shows an example of the relationship between an image attribute and an attribute determination signal. It is explanatory drawing for demonstrating the case where the resolution is converted. It is an internal block diagram of the primary-side receiving-side image processing apparatus. It is an internal block diagram of the secondary-side receiving image processing apparatus. 7 is a block diagram illustrating a configuration of a transmission-side image processing apparatus according to Embodiment 2. FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Transmission side image processing apparatus 3 Communication network 5 Reception side image processing apparatus 5a Primary purpose reception side image processing apparatus 5b Secondary purpose reception side image processing apparatus 7 Data creation part 9 First conversion part 11 Second conversion Unit 13 subtraction unit 15 compression unit 17 embedding unit 21 visualization unit 25 addition unit 27 extraction unit 29 expansion unit 35 attribute determination unit

Claims (3)

Attribute determination means for determining whether image data is a pattern area in a predetermined area unit;
1 conversion means for generating a first image data by performing a color space quantization width expansion process on the image data;
Second conversion means for generating a second image data by performing a color space quantization width reduction process on the first image data;
Arithmetic means for performing subtraction between the image data and the second image data to generate difference data;
Compression means for compressing the pattern area of the difference data at a low compression rate and compressing other areas of the difference data at a high compression rate to generate compressed data;
Embedding means for embedding the compressed data in the first image data;
An image processing apparatus comprising: An attribute determination step of determining whether the image data is a pattern area in a predetermined area unit;
A conversion step of generating first image data by performing color space quantization width expansion processing on the image data;
A second conversion step of generating a second image data by performing a color space quantization width reduction process on the first image data;
A calculation step of subtracting the image data and the second image data to generate difference data;
A compression step of compressing the pattern area of the difference data at a low compression ratio and compressing the other area of the difference data at a high compression ratio to generate compressed data;
An embedding step of embedding the compressed data in the first image data;
An image processing method comprising: An image processing program for causing a computer to execute the method according to claim 2 .

Images