JP2012222822A - Image processing device and image processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing device capable of detecting a specific mark from image data of a variety of magnifications at a low cost.SOLUTION: An image processing device according to an embodiment comprises: reduction means for reducing, when an input magnification of input image data to be printed exceeds a reference magnification, the input image data to reduced image data of the reference magnification; and detection means for detecting a specific mark from the reduced image data when the input magnification exceeds the reference magnification and detecting the specific mark from the input image data when the input magnification does not exceed the reference magnification.

Description

  Embodiments described herein relate generally to an image processing apparatus and an image processing program.

  For example, an image processing apparatus such as a color multifunction peripheral (Multi-Functional Peripheral (MFP)) scans a document and detects a specific mark (for example, a copy prohibition mark) from image data obtained by reading the document. Cancel.

  A scanner mounted on an MFP or the like, for example, reads a document two-dimensionally by moving a line sensor with a carriage. In order to change the scanning magnification, the moving speed of the carriage is changed to change the aspect ratio of the image. For this reason, if a specific mark is to be detected when changing the scanning magnification, it is necessary to change the detection parameter of the specific mark.

  However, the scan magnification of the MFP is generally wide from several tens to several hundreds per 1%, and it is difficult to prepare detection parameters corresponding to all of them. In addition, when an enlargement scan with a magnification of 100% or more is performed, the resolution of the image is increased, so that the number of pixels to be processed increases, and as a result, the calculation cost may increase.

  Therefore, after reducing the image to a fixed size by thinning out the image data in the sub-scanning direction (direction in which the carriage moves) according to the magnification so that the aspect ratio and size of the image data read by the line sensor are always constant. A method for performing recognition processing of a specific mark or the like has been proposed.

JP-A-6-237379

  However, if the read image data is always thinned out in the sub-scanning direction so that the aspect ratio and size of the image data read by the line sensor are always constant, the amount of information necessary to detect the specific mark is lost. Therefore, the detection accuracy of the specific mark may be reduced.

  An object of the present invention is to provide an image processing apparatus and an image processing program capable of detecting a specific mark from image data of various magnifications at low cost.

  The image processing apparatus according to the embodiment includes a reduction unit that reduces the input image data to reduced image data having the reference magnification when the input magnification of the input image data for printout exceeds the reference magnification, and the input magnification is the reference magnification. Detecting means for detecting a specific mark from the reduced image data, and detecting the specific mark from the input image data when the input magnification does not exceed a reference magnification.

It is a figure which shows an example of a structure of the process block of the image recognition apparatus (image processing apparatus) which concerns on embodiment. It is a flowchart which shows an example of the scaling process in an image correction part. It is a figure which shows an example of schematic structure of a binarization process part. It is a flowchart which shows an example of the pixel coefficient process of a pixel counting part. It is a figure which shows an example of the pattern feature-value of the determination target calculated | required from the input image. It is a figure which shows an example of the pattern feature-value registered previously for referring in the collation of a pattern. It is a figure which shows an example of the absolute value of the difference for every element of the determination target pattern feature-value of FIG. 5A, and the reference pattern feature-value of FIG. 5B. It is a figure which shows an example of the value corresponded to the similarity for every element converted using the look-up table called the membership function which inputs the difference value of a feature-value. An example of the membership function used for the conversion from FIG. 5C to FIG. 5D is shown. It is a figure for demonstrating the effect of this embodiment. FIG. 2 is a diagram illustrating an example of a schematic system configuration of a digital multifunction peripheral (MFP) equipped with the image recognition apparatus illustrated in FIG. 1. FIG. 2 is a cross-sectional view illustrating an example of a schematic configuration of a digital multifunction peripheral (MFP) in which the image recognition apparatus illustrated in FIG. 1 is mounted.

  Hereinafter, embodiments will be described with reference to the drawings.

  FIG. 1 is a diagram illustrating an example of a configuration of a processing block of an image recognition apparatus (image processing apparatus) according to an embodiment. The image recognition apparatus shown in FIG. 1 can be applied to, for example, a digital multi-function peripheral (MFP) shown in FIGS. The MFPs in FIGS. 8 and 9 will be described in detail later.

  As shown in FIG. 1, the image recognition apparatus includes a magnification reference unit 101, a magnification correction unit 102, a binarization unit 103, a pixel coefficient unit 104, a feature amount extraction unit 105, a reference feature amount storage unit 106, and a pattern matching unit 107. Etc.

  The magnification reference unit 101 is a processing unit that reads data of an image reading magnification (input magnification) designated by a user on the control panel of the MFP from a predetermined memory.

  When the image reading magnification read by the magnification reference unit 101 is larger than a predetermined value (100% (standard magnification) in the present embodiment), the magnification correction unit 102 has a predetermined magnification ( It is a processing unit that outputs image data that has been subjected to scaling processing (reduction processing) so as to be equivalent to (reference magnification). However, when the image reading magnification is not more than a predetermined value, the read image is output as it is. That is, when the image reading magnification is equal to or smaller than a predetermined value, the read image is output as it is without being enlarged. Therefore, in the present embodiment, the contents of the scaling process are necessarily reduced processes, and pixel thinning and averaging are performed according to the magnification ratio.

  FIG. 2 is a flowchart illustrating an example of the scaling process in the image correction unit 102.

  The image correcting unit 102 determines whether or not the image reading magnification M is 100% or more. If M exceeds 100% (ACT 201, YES), the image size is set to {(100 / M) × 100}. % Is processed (ACT202), and when M is 100% or less (ACT201, NO), no processing is performed and the image data is output to the subsequent process (ACT203). Since the present embodiment assumes that the image recognition apparatus according to the present invention is installed in an MFP, scaling of the input image is in the sub-scanning direction (reading magnification change by line sensor speed control).

  FIG. 3 is a diagram illustrating an example of a schematic configuration of the binarization processing unit 103.

  The binarization unit 103 is a processing unit that binarizes each pixel of the image data output from the magnification correction unit 102 depending on whether or not it has a predetermined color or gradation value. Specifically, when the read image is a color image composed of three components R (Red), G (Green), and B (Blue), binarization may be performed by a processing block as shown in FIG. it can. That is, each of the color component values (R, G, B) is compared with the upper limit value and the lower limit value by the comparators 301 to 306, respectively, and true / false judgment is performed. The logical product of these judgment results is obtained by the logical product 307. By calculating, only pixels belonging to a predetermined partial color space in the RGB color space can be extracted.

  The pixel counting unit 104 is a processing unit that adds the binarization results in each partial region obtained by space-dividing the image region without overlapping. Further, the size of this partial area is variable in accordance with the image reading magnification, and the result of adding the binarization results is normalized with the partial area size when the image reading magnification is 100%. Specifically, when adding the binarization results, the number of added pixels is counted, and the addition result is divided by the number of counted pixels and multiplied by the number of pixels at an image reading magnification of 100%.

  FIG. 4 is a flowchart illustrating an example of the pixel counting process of the pixel counting unit 104.

  The pixel counting unit 104 calculates the magnification ratio R by dividing 100 by the image reading correction magnification M ′ (%), and resets the first line counter C1 and the sub-scanning coordinate value y to 0 (ACT 401). Here, the image reading correction magnification M ′ is M ′ = 100 when the image reading magnification M is 100 or more, and M ′ = M when M is less than 100.

  The pixel counting unit 104 resets the accumulated line memory L (x) to 0 in the main scanning direction, and also resets the second line counter C2 to 0 (ACT 402).

  The pixel counting unit 104 adds the binarized image data to the cumulative line memory L (x), adds the magnification ratio R to the first line counter, and sets 1 to each of the second line counter and the sub-scanning coordinate value. Add (ACT403).

  The pixel counting unit 104 determines whether or not the first line counter C1 is greater than or equal to a predetermined integer value N and whether or not the sub-scanning coordinate value y exceeds the maximum value. Is true (ACT 404, YES), the data corrected by multiplying the value of the accumulated line memory L (x) by (N / C2) is output, and the integer N is subtracted from the value of the first line counter C1. Then, refresh is performed (ACT 405), and it is determined whether or not the sub-scanning coordinate value y exceeds the maximum value. If the determination result is true, the process is terminated (ACT 406).

  If the condition of ACT 404 is not satisfied (ACT 404, NO), the pixel counting unit 104 repeats the process of ACT 403. Further, when the condition of ACT 406 is not satisfied (ACT 406, NO), the image counting unit 104 repeats the processes of ACT 402 to ACT 405.

  The feature amount extraction unit 105 holds the output data of the pixel counting unit 104 in a buffer memory, and further sequentially counts data for each predetermined partial area (that is, integrates within the partial area), whereby the pattern feature in each attention area It is a processing unit for calculating the quantity.

  The reference feature amount storage unit 106 is a memory for storing one or more sets of pattern feature amounts (template corresponding to the specific mark) calculated in advance for the specific mark to be detected. In response, one set of pattern feature values is output.

  The pattern matching unit 107 calculates the degree of coincidence (or degree of inconsistency) between the pattern feature amount calculated by the feature amount extraction unit 105 and the pattern feature amount read from the reference feature amount storage unit 106, and the calculated match When the degree is larger than a predetermined threshold (or when the degree of mismatch is smaller than the predetermined threshold), the processing unit outputs a detection signal for a specific mark.

  As a method of calculating the degree of coincidence (or degree of inconsistency) obtained by the pattern matching unit 107, there are various methods such as similarity (degree of coincidence) and Euclidean distance (degree of inconsistency) by fuzzy inference.

  5A to 5D are diagrams for explaining an example of a similarity calculation method based on fuzzy inference. FIG. 5A is a diagram illustrating an example of a pattern feature amount to be determined obtained from an input image. Here, the “element number” on the horizontal axis is the order given to each area obtained by segmenting the pattern area. In the present embodiment, the number of elements is set to 4 and the maximum number of each element for ease of explanation. The value is 5.

  FIG. 5B is a diagram illustrating an example of pattern feature amounts registered in advance for reference in pattern matching. FIG. 5C is a diagram illustrating an example of an absolute value of a difference for each element between the determination target pattern feature quantity of FIG. 5A and the reference pattern feature quantity of FIG. 5B. FIG. 5D is a diagram illustrating an example of a value corresponding to the similarity for each element, which is converted using a lookup table called a membership function that receives the difference value of the feature amount. FIG. 6 shows an example of the membership function used for the conversion from FIG. 5C to FIG. 5D.

  FIG. 7 is a diagram for explaining the effect of the present embodiment. Here, as an example, three cases where the reading magnification of the document is 100%, 200%, and 50% are compared. Since the size of the partial area for calculating the pattern feature amount needs to be proportional to the reading magnification, the area of the partial area is doubled when the reading magnification is 200% compared to when the reading magnification is 100%. The memory capacity required for the feature amount calculation based on the integration as described above is also doubled. Therefore, by performing magnification correction to an image corresponding to a reading magnification of 100% with respect to an input image having a reading magnification of 200%, it is possible to reduce the cost of memory required for calculation.

  On the other hand, when the reading magnification is 50%, the size of the partial area for calculating the feature amount is half that when the reading magnification is 100%, and the memory capacity required for the feature amount calculation can be reduced (however, the information amount of the feature amount itself) Therefore, it is not necessary to perform magnification correction on the input image.

  For example, when the reading magnification M exceeds 100% (for example, when the reading magnification M is 200%), the pattern matching unit 107 determines each area (for example, 4) of the reference size that constitutes the image data that has been reduced and corrected to be equivalent to 100%. The feature amount is calculated from (4 pixels are regarded as one pixel), and the specific mark is detected based on the feature amount calculation result. Further, when the reading magnification M does not exceed 100% (for example, when the reading magnification M is 50%), the pattern matching unit 107 sets a reduced size smaller than the reference size based on the difference between the reading magnification M and 100%. Then, the feature amount is calculated from each reduced-size region (for example, 4 × 2 pixels are regarded as one pixel) constituting the input image data, and the specific mark is detected based on the feature amount calculation result.

  FIG. 8 is a diagram showing an example of a schematic system configuration of a digital multifunction peripheral (MFP) equipped with the image recognition apparatus shown in FIG.

  The image reading device 801 generates digital image data by A / D conversion of an electrical signal obtained by optically reading a document.

  The image recognition device 802 corresponds to the image recognition device shown in FIG. 1 and determines whether or not a specific image exists in the image data generated by the image reading device 801. A detection signal is generated.

  The image processing device 803 performs a filtering process and a screening process on the image data generated by the image reading apparatus 801, thereby smoothing the gradation of the photographic image and enhancing the contrast of the character image to improve the visibility. Or

  The image output device 804 outputs the image data that has been subjected to predetermined processing by the image processing device 803 to a paper medium or the like. That is, the image output device 804 prints out an image based on the image data on a paper medium.

  The system control device 805 performs overall control of the above four devices, centralized management of the operation state of each device, and the like. For example, the image reading magnification data is acquired from the image reading device 801 and set as an operation condition in the image recognition device 802, or the operation of the image processing device 803 is stopped when a detection signal is output from the image recognition device 802. The operation of the image reading device 801 is also stopped when a printout (copying) by the image output device 804 is prohibited or a trouble such as a paper jam occurs in the image output device 804.

  For example, the system control device 805 restricts the output of image data based on a detection signal indicating that a specific mark has been detected. More specifically, the system control device 805 prohibits printing out image data based on a detection signal indicating that a specific mark has been detected.

  FIG. 9 is a cross-sectional view showing an example of a schematic configuration of a digital multi-function peripheral (MFP) equipped with the image recognition apparatus shown in FIG.

  The MFP 100 shown in FIG. 9 outputs image information as, for example, hard copy or printout, and outputs the image information as an image output in a state where toner is fixed on, for example, plain paper or an OHP sheet that is a transparent resin sheet. (Corresponding to the image output device 804), a sheet supply unit 3 that supplies a sheet of an arbitrary size used for image output to the image forming unit 1, and image information that is a target for image formation in the image forming unit 1 An image reading unit 5 (corresponding to the image reading device 801) that captures image data from a reading object (hereinafter referred to as a document) holding information is included.

  The image reading unit 5 includes a document table (document glass) 5a that supports a document and an image sensor, for example, a CCD sensor, that converts image information into image data. The image reading unit 5 converts the reflected light obtained by irradiating the illumination light from the illumination device onto the document set on the document table 5a into an image signal by the CCD sensor.

  Further, when the document is a sheet, the image reading unit 5 discharges the read document from the discharge unit, and at the reading position following the image formation or image capture (hereinafter referred to as reading) of the next document. An automatic document feeder (ADF) 7 for guiding is integrally provided. A document cover may be used in place of the ADF 7.

  In addition, the CCD sensor of the image reading unit 5 may be positioned in the document movement path of the document table 5 a to which the document is moved by the ADF 7. The CCD sensor of the image reading unit 5 is positioned in the document movement path of the document table 5a to which the document is moved by the ADF 7, so that the image information included in the document can be read without positioning the document on the document table 5a. .

  An operation panel (operation unit) 9 that is an operation input unit that gives an instruction to start reading image information of a document by the image reading unit 5 and start image formation to the image forming unit 1 is on the left side or right side of the image reading unit 5. Are positioned on the support 9a and the swing arm 9b (fixed to the image forming unit 1).

  The image forming unit 1 includes first to fourth photosensitive drums 11a to 11d that hold latent images, and a developing device 13a that supplies a developer, that is, toner, to the latent images held by the photosensitive drums 11a to 11d to develop the latent images. To 13d, a transfer belt 15 for sequentially holding toner images held by the photosensitive drums 11a to 11d, and first to fourth cleaners for removing toner remaining on the photosensitive drums 11a to 11d from the individual photosensitive drums 11a to 11d. 17a to 17d, a transfer device 19 that transfers the toner image held by the transfer belt 15 to a sheet that is a transparent resin sheet such as plain paper or an OHP sheet, and the toner image transferred to the sheet by the transfer device 19 is fixed to the sheet. It includes a fixing device 21 and an exposure device 23 that forms latent images on the photosensitive drums 11a to 11d.

  The first to fourth developing devices 13a to 13d are arbitrary colors of Y (yellow, yellow), M (magenta), C (cyan), and Bk (black, black) used for obtaining a color image by subtractive color mixing. The latent image held by each of the photosensitive drums 11a to 11d is visualized in any one of Y, M, C, and Bk colors. The order of the colors is determined in a predetermined order according to the image forming process and the characteristics of the toner.

  The transfer belt 15 holds the toner images of the respective colors formed by the first to fourth photosensitive drums 11a to 11d and the corresponding developing devices 13a to 13d in the order of toner image formation.

  The transfer belt 15 sets the pressure exerted by the belt facing member 51 that sets the pressure between the photosensitive drums 11 a to 11 d of the image forming unit 1 and the transfer belt 15 and the belt cleaner 53 that cleans the surface of the transfer belt 15. The belt cleaner facing member 55 and the sheet receive a predetermined pressure that is a tension toward the outside from each of the support rollers 57 that sets the pressure on the transfer belt 15 side when contacting the transfer belt 15 by the pressure from the transfer device 19. .

  The paper supply unit 3 supplies a sheet for moving the toner image to the transfer device 19 at a predetermined timing.

  Cassettes not described in detail located in the plurality of cassette slots 31 accommodate sheets of any size, and the pickup roller 33 takes out the sheets from the corresponding cassettes according to the image forming operation. The size of the sheet corresponds to the magnification required for image formation and the size of the toner image formed by the image forming unit 1.

  The separation mechanism 35 prevents the pickup roller 33 from taking out two or more sheets from the cassette.

  The plurality of transport rollers 37 send the sheet separated by the separation mechanism 35 into one sheet toward the aligning roller 39.

  The aligning roller 39 sends the sheet to a transfer position where the transfer device 19 and the transfer belt 15 are in contact with the transfer device 19 at a timing when the toner image is transferred from the transfer belt 15.

  The fixing device 21 fixes a toner image corresponding to image information on a sheet, and as an image output (hard copy or printout), a stock unit 47 located in a space between the image reading unit 5 and the image forming unit main body 1. Send to.

  The transfer device 19 is positioned in an automatic duplex unit (ADU) 41 that replaces both sides of a sheet that is an output image (hard copy / printout) on which a toner image is fixed by the fixing device 21. A bypass tray is attached to the ADU 41.

  In the image forming unit 1, the ADU 41 is jammed between the (final) transport roller 37 and the aligning roller 39, or between the aligning roller 39 and the fixing device 21 or between the transfer device 19 and the fixing device 21. If it is jammed, move to the side (right side). The ADU 41 integrally includes a cleaner 43 that cleans the transfer device 19.

  A media sensor 45 positioned between the conveying roller 37 and the aligning roller 39 detects the thickness of the sheet conveyed to the aligning roller 39. The media sensor 45 includes an optical sensor shown in U.S. Patent Application No. 12/197880 filed on August 25, 2008, U.S. Patent Application No. 12/199424 filed on August 27, 2008, and Alternatively, a type using a shift of the thickness detection roller shown in 61/043801, which was provisionally filed in the United States on April 10, 2008, can be used.

  The configuration of this embodiment will be summarized below.

  (1) An image recognition apparatus according to the present embodiment is an image recognition apparatus that detects a specific mark from input multi-value image data, and includes an input magnification reference unit that refers to an input magnification of image data, and an input magnification is Magnification correction means for reducing and correcting input image data corresponding to a predetermined magnification when a predetermined magnification is exceeded, binarization means for binarizing the magnification-corrected image data based on pixel values, and binarization A pixel counting unit that counts the number of pixels of the image data within an area of a size corresponding to the input magnification, a feature amount extracting unit that calculates a feature amount based on the pixel count result, and a feature amount that is calculated in advance for a specific mark Whether or not a specific mark exists in the input image data based on a comparison between the reference feature quantity storage means to be stored and the feature quantity calculated by the feature quantity extraction means and the feature quantity read from the reference feature quantity storage means Or Comprising a determining pattern matching means.

  (2) Further, in addition to the above (1), the image recognition apparatus according to the present embodiment includes a magnification setting unit for setting an input magnification by a user, and a reading unit for reading a document with a size corresponding to the set magnification. When the image recognition apparatus detects a specific mark from the input image data, the input image data is not faithfully output.

  With the above-described configuration, the image recognition apparatus according to the present embodiment does not decrease the detection accuracy of a specific mark at a normal image reading magnification even when the minimum reading magnification of the detection target is reduced, and even when the reading magnification is large. An increase in memory can be prevented.

  The image processing apparatus according to the present embodiment has been described with respect to a case where a function (program) for implementing the present embodiment is preset in the apparatus. However, the present invention is not limited to this, and a similar function (program) is provided from the network. It is also possible to download to the device, or to install the same function (program) stored in a recording medium in the apparatus. The recording medium may take any form as long as it can store a program such as an optical disk and a USB memory and can be read by the apparatus. Further, the function obtained by installing or downloading in advance may be realized in cooperation with an OS (operating system) or the like inside the apparatus.

  Although several embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 101 ... Magnification reference part, 102 ... Magnification correction part, 103 ... Binarization part, 104 ... Pixel coefficient part, 105 ... Feature quantity extraction part, 106 ... Reference feature-value memory | storage part, 107 ... Pattern matching part

Claims (9)

A reduction means for reducing the input image data to reduced image data of the reference magnification when the input magnification of the input image data for printout exceeds the reference magnification;
Detecting means for detecting a specific mark from the reduced image data when the input magnification exceeds a reference magnification, and detecting the specific mark from the input image data when the input magnification does not exceed a reference magnification;
An image processing apparatus comprising:   The detection means calculates a feature amount from each region of the reference size constituting the reduced image data when the input magnification exceeds a reference magnification, detects the specific mark based on a feature amount calculation result, and inputs the input When the magnification does not exceed a reference magnification, a reduction size smaller than the reference size is detected based on a difference between the reference magnification and the input magnification, and a feature amount is calculated from each region of the reduction size constituting the input image data. The image processing apparatus according to claim 1, wherein the specific mark is detected based on a feature amount calculation result.   The detection means detects the specific mark based on a predetermined template corresponding to the specific mark in both cases where the input magnification exceeds a reference magnification and when the input magnification does not exceed a reference magnification. 2. The image processing apparatus according to 2.   The image processing apparatus according to claim 1, further comprising a restriction unit that restricts an output of the input image data when the specific mark is detected.   The image processing apparatus according to claim 4, wherein the restricting unit prohibits printing of the input image data when the specific mark is detected.   The image processing apparatus according to claim 1, further comprising an image reading unit that reads a document image at the input magnification and outputs the input image data.   The image processing apparatus according to claim 1, further comprising an image output unit that prints out the input image data. When the input magnification of the input image data for printout exceeds a reference magnification, a procedure for reducing the input image data to reduced image data of the reference magnification;
Detecting the specific mark from the reduced image data when the input magnification exceeds a reference magnification, and detecting the specific mark from the input image data when the input magnification does not exceed the reference magnification;
An image processing program for causing a computer to execute.   The image processing program according to claim 8, wherein, when the specific mark is detected, the computer executes a procedure for restricting the output of the input image data.

Images