JP5166158B2 - Image processing apparatus, image reading apparatus, image reading system, image processing method, and image processing program - Google Patents

Description

  The present invention relates to an image reading apparatus that photoelectrically converts light from an original to read an image of the original, and an image processing method for an image forming apparatus that forms an image based on the read image data.

  There is a conventional apparatus that reads an image of a document using an image sensor and performs image processing of the document by digital processing. In addition, this type of conventional image reading apparatus and image processing apparatus may have a function of preventing image data of a document from being stored when a blank document that does not need to be stored is read. . Since such an apparatus does not store unnecessary image data, it is useful for saving capacity when storing an image on a recording medium such as a hard disk.

Conventionally, such a function that does not store image data of a blank original has been realized by a combination of the following means as an example. One is binary image creating means for creating a binary image by replacing a read image of a read original with black pixels when the pixel density of the image is below a certain threshold. The other is to repeatedly scan the binary image generated from the scanned image in the horizontal direction from the upper left to the lower right end and count black pixels (see Patent Document 1), or white pixels are black. A blank document determination unit that counts the number of places where pixels have changed and a place where white pixels have changed to black pixels, and determines that the document is blank when the number of black pixels or the number of changed places is equal to or less than a certain threshold. is there. As another example, a histogram creation unit that creates a density histogram of a scanned image, and a blank document that determines that the document is not a blank page when the created density histogram has more than a certain percentage of pixels darker than the density threshold value There is an apparatus realized by a combination of determination means (see Patent Document 2). By these methods, unnecessary blank original image data is discarded.
JP-A-10-229484 Japanese Patent Laid-Open No. 06-141126

However, in the conventional blank document determination method for counting the number of times the pixel value has changed in the above binary image or the conventional blank document determination method based on the ratio of black pixels, even if the read document is a blank sheet, the blank document In this case, a time-consuming process of always creating a binary image used to determine whether the Further, another conventional blank document determination unit that creates a density histogram and performs the determination based on the image density erroneously determines that a document having no characters or images but having a dark background color is not blank. In other words, the conventional technique has a problem that the speed of reading and storing a document image is reduced and a problem that the blank document determination accuracy is low, and the present invention aims to solve this problem.

The image processing apparatus of the present invention has been made in view of the above problems, and includes a first determination method including a case where it is not determined whether or not the document is a blank document based on image data read from the document. Thus, the first determination method for determining whether or not the original is a blank original and the original for which the first blank original determination means has not determined whether or not the original is a blank original. And a second blank page document determination unit that determines whether or not the document is a blank page document , and the first blank page document determination unit is based on multi-value image data corresponding to the document. A histogram generating means for generating a histogram corresponding to the original; and a means for determining whether the original is a blank original based on the generated histogram, wherein the second blank original determining means comprises: The original Wherein means for creating the binary image data from the multivalued image data corresponding, on the basis of the binary image data created in the above, the document is a Rukoto a means for determining whether the blank document page in to.

The image processing method of the present invention is a first determination method including a case where it is not determined whether or not the document is a blank document based on image data read from the document, and whether or not the document is a blank document. And a second determination method different from the first determination method for a first blank document determination step that determines whether or not a blank document is determined by the first blank document determination step. A second blank document determination step for determining whether or not the document is a blank document , wherein the first blank document determination step creates a histogram corresponding to the document based on the multi-value image data corresponding to the document. And a step of determining whether or not the document is a blank document based on the generated histogram, wherein the second blank document determination step is performed by adding 2 from the multi-value image data corresponding to the document. Value image data A step of creating, based on the binary image data created in the above, the document is characterized Rukoto a step of determining whether a blank document.

The program according to the present invention determines whether or not the document is a blank document by a first determination method including a case where it is not determined whether or not the document is a blank document based on image data read from the document. The original is determined by a second determination method different from the first determination method for an original that has not been determined whether the original is a blank original by the first blank original determination step and the first blank original determination step. An image processing program for causing a computer to execute a second blank document determination step for determining whether or not a blank document is present, wherein the first blank document determination step is based on multi-value image data corresponding to the document. Generating a histogram corresponding to the original, and determining whether the original is a blank original based on the generated histogram, The paper document determination step includes a step of creating binary image data from the multi-value image data corresponding to the document, and determining whether or not the document is a blank document based on the created binary image data. And a step .

In the present invention, before creating a binary image and making a blank document determination, a histogram of the document is created and a blank document determination is performed based on the histogram. The histogram can be created with a smaller memory capacity than creating a binary image, and the determination method is simple, so that a blank document can be determined quickly. For this reason, blank document determination is performed using a histogram before a binary image is created, and a binary image is created only for a document that cannot be easily determined to be a blank sheet, from the start of image processing to image storage. Time can be shortened. Furthermore, when it is determined in the histogram blank page determination that it is unclear whether a blank document having a dark background color is blank or not, a blank document determination with high accuracy can be performed by a blank document determination based on a binary image.

Although an image reading apparatus such as a scanner embodying the present invention will be described, the present invention can also be applied to an image forming apparatus such as a copying machine or a printer having a similar image processing method.

Exemplary embodiments of the present invention will be described below with reference to the drawings. However, the image processing described in this embodiment should be changed as appropriate according to the configuration of the apparatus to which the invention is applied and various conditions, and the scope of the present invention is limited to the following embodiment. Is not.

FIG. 2 is a sectional view schematically showing the overall configuration of the image reading apparatus according to the first embodiment of the present invention. As shown in FIG. 2, the image reading apparatus 1 includes a pickup roller 2 that picks up a document D, a feed roller 3 that supplies the document D picked up by the pickup roller 2, and a picked-up document D. A separation roller 4 that separates the sheets one by one and a document detection sensor 10 that detects whether a document is stacked or not are provided. The image reading apparatus 1 also includes a registration roller pair 5 for conveying the document D. Further, the image reading unit 33 that reads the surface of the document includes a line image sensor 7a that reads the image information of the document D through the contact glass 6a. The image reading unit 34 that reads the back side of the document includes a line image sensor 7b that reads image information of the document D through the contact glass 6b, and is controlled by the scanner control unit 301 together with the image reading unit 33. The white plate as the facing member 9 of the line image sensor 7b is supported by the contact glass 6a. The image reading apparatus 1 includes a pair of conveying rollers 8 disposed on the downstream side of the image reading units 33 and 34.

The image reading apparatus 1 takes the document D into the apparatus by the pickup roller 2 and the feeding roller 3 and separates the original D one by one by the separation roller 4. The document D is nipped and conveyed by the registration roller pair 5 and the conveyance roller pair 8, the image on the front side of the document D is read by the image reading unit 33, and the image on the back side of the document D is read by the image reading unit 34. Since the opposing member 9 of the line image sensor 7b is a white plate, the periphery of the read document image is a white background image. If the opposing member 9 is a black plate and the background of the original is a black background image, the blank page determination method using the histogram described below may reduce the ratio of the area of the read original image to the area of the read image. May not be able to judge well. Therefore, in this embodiment, the facing member 9 is a white plate. While the image is read, the document D is discharged outside the apparatus while being nipped and conveyed by the conveying roller pair 8.

Next, the scanner control unit 301 of the image reading unit described with reference to FIG. 2 will be described with reference to FIG. FIG. 3 is a block diagram illustrating a schematic configuration of the scanner control unit 301 of FIG. The scanner control unit 301 includes an image reading unit 11 corresponding to the image reading units 33 and 34 in FIG. 2, an A / D conversion unit 12, an image processing unit 13, a CPU 14 that functions as a control center, a memory 15 that temporarily stores image data, an external It is comprised by the communication part 16 for connecting with an apparatus. The host computer 17 functions as an external control unit.

Image data captured by the image reading unit 11 of the scanner control unit 301 is sent to the A / D conversion unit 12. The A / D converter 12 performs A / D conversion for converting an analog electrical signal into digital data. The image data processed by the A / D conversion unit 12 is sent to the image processing unit 13. Here, the sensitivity unevenness of the line image sensor is corrected for each pixel cell of the line image sensor, and the brightness of the entire image is increased. Processing for equalization (hereinafter, shading correction), creation of a histogram, blank document determination, and the like are performed. Thereafter, the image data is transferred to the host computer 17 that is an external control unit via the communication unit 16 that is a general-purpose USB or SCSI.

The memory 15 stores a control program to be executed by the CPU 14, and the CPU 14 controls the apparatus while reading program instructions from the memory 15. The memory 15 also has a role of temporarily storing image data processed by the image processing unit 13.

From here, the operation | movement in a present Example is demonstrated using FIG. In this embodiment, a blank document determination method using a histogram is used as the first blank document determination method, and a blank document determination method using a binary image is used as the second blank document determination method.

In this embodiment, the image information of the captured original is converted into image data by A / D conversion, and then shading correction is performed by the image processing unit 13 to create a histogram. Here, the first blank page original is determined based on the histogram. The image data determined to be blank is discarded from the memory 15. The image data of the document determined not to be blank is transferred to the host computer 17 via the communication unit 16. If it is determined whether it is blank or not, the image processing unit 13 generates a binary image. The generated binary image is temporarily stored in the memory 15. Thereafter, the CPU 14 refers to the binary image stored in the memory 15 and performs second blank page original determination based on the binary image.

Next, the blank document determination operation will be described with reference to the flowchart of FIG. This process is performed by the scanner control unit 301. 1b and 1c are flowcharts of blank page determination for each embodiment, and the description will be given later.

When receiving a scan start instruction from the host computer 17 or the like, the CPU 14 determines whether or not a document is stacked on the document table by examining the output signal of the document detection sensor 10 of the image reading apparatus 1 (step ST10). If it is determined that there is no document, the process is terminated without performing the subsequent scanning operation. When it is determined that the document is loaded on the document table, the above-described scanning operation is controlled (step ST11). The image data read by the image reading unit 11 is temporarily stored in the memory, and it is determined whether or not the image data is binary image data (step ST12). If the image data read here is binary image data, blank page determination is performed using the binary image data without creating a histogram (step ST32). If it is determined in step ST12 that the read image data is not binary image data, the image processing unit 13 creates, for example, a pixel density histogram from the document image data (step ST13).

Next, based on the created histogram, the CPU 14 determines whether the document is a blank document, a blank document, or is not clear, and if it is determined as a blank document (YES in step ST14), the image. Data is discarded from the memory 15 (step ST15). If it is determined that the document is not a blank document (NO in step ST14), the image processing unit 13 is caused to execute image processing on the image data (step ST16). The image processing includes, for example, processing for correcting distortion of image data generated due to the document being conveyed at an inclination, processing for detecting a document frame and deleting a background portion outside it. Further, the image data subjected to the image processing is temporarily stored in the memory 15 (step ST17). Thereafter, the image data is output via the communication unit 16 (step ST18), and the process returns to step ST10.

If it is determined in step ST14 whether the document is blank or not, binary image data is generated from the image data (step ST31). It should be noted that the binary image data generated here is preferably discarded after the determination described later is performed. Next, a blank page original determination method based on binary image data is performed, and it is further determined whether the page is blank or not (step ST32). Details of the histogram creation means and the blank page determination means based on the histogram will be described later.

If the binary image data is used to determine a blank original (YES in step ST32), the multi-value image data is discarded from the memory 15 (step ST33). Details of the binary image data generation method and the blank document determination method based on the binary image data will also be described later.

When it is determined from the binary image data that the document is not a blank document (NO in step ST32), image processing in the image processing unit 13 is executed on the multi-valued image data recorded in the memory 15 ( Step ST16). The image processed image data is temporarily stored in the memory 15 (step ST17). Thereafter, the image data is output via the communication unit 16 (step ST18). The output image data of the original is stored in a hard disk in the host computer 17. The operations from step ST11 to step ST18 are repeated until there are no stacked originals.

In this description, it is described that the image data determined not to be a blank document is transmitted to the host computer via the communication unit 16 and stored in the hard disk of the host computer. However, the image data created in the present invention is described. Is not limited to a hard disk, and storage media such as CD-R, CD-RW, DVD-R, DVD-RAM, DVD-RW, DVD + RW and other optical disks, magnetic tape, and non-volatile memory cards The storage device used may be used.

Next, as an example of a method of performing blank page document determination shown in steps ST13 to ST14 in FIG. 1a, a histogram of differences between the density of a pixel of interest and the density of neighboring pixels from image data of the document (hereinafter referred to as difference histogram). ) Is created and a blank document determination method is described.

First, the method for creating the difference histogram will be described as an example. FIGS. 4a, 5a, and 6a represent difference histograms created by the histogram creation method described later. In this description, the horizontal axis represents the difference in density between the pixel of interest and its neighboring pixels. The vertical axis represents frequency. Here, it is assumed that the larger the density value is, the brighter it is, but the present invention is not limited to this. 4b, FIG. 4c, FIG. 5b, and FIG. The density difference between the pixel of interest and its neighboring pixels is the image density data X [i] of the pixel of interest on the image data of the document and the image density of a pixel that is m pixels away from the pixel of interest in the main scanning direction. This is illustrated using the difference from the data X [i + m] (see FIG. 11). If the pixel density is in the range of 0 to 255, for example, the density difference is 0 to 255. If the difference in density is t, it can be expressed by the following formula (1).
Concentration difference t = abs (X [i] −X [i + m]) (1)

The calculation is continued by sequentially shifting the target pixel in the main scanning direction, the density difference t is calculated until there is no pixel that is m pixels away from the target pixel in the main scanning direction, and the above calculation is repeated by moving in the sub scanning direction. Thus, a difference histogram is created. The density data of the pixels when the original is read as a color image is, for example, a value obtained by averaging the density values of R (red), G (green), and B (blue) of each pixel by some calculation. . When calculating the density difference, a pixel in the vicinity of the target pixel in the sub-scanning direction may be used, or a pixel in the vicinity of the main sub-scanning direction and a pixel in the vicinity of the sub-scanning direction. Both pixels may be used. For image density data when a document is read as a color image, three histograms may be created using the density values of RGB, or a histogram for only one of the colors may be created. .

Subsequently, a blank document determination method for determining whether the blank histogram or the blank sheet is unclear based on the shape of the created difference histogram will be described with reference to FIGS. 1 b, 4, and 4. 5 and FIG. 6 will be described as an example.

In the difference histogram creation method described above, a threshold value P is defined for the horizontal axis of the created difference histogram, and a blank original is determined depending on how much pixels having a density difference of P to 255 occupy the entire histogram. Make a decision. In this description, the ratio of pixels having a density difference equal to or higher than the threshold P to the total of all histograms is a [%] (step ST41 in FIG. 1b), and the vertical axis of the difference histogram is Y. It can be represented by the following formula (2).
a [%] = {Σ (i = P, 255) Yi / Σ (i = 0,255) Yi} * 100 (2)

Subsequently, two threshold values are prepared for a [%] in order to determine whether the sheet is blank, not blank, or unclear. These threshold values are constants that are obtained by experiment and are embedded in the memory 15 in advance, and are a ratio L [%] that is a threshold value for determining as blank paper and a ratio H that is a threshold value for determining that the sheet is not blank paper.
[%]. Here, L [%] <H [%]. That is, if a [%] is equal to or less than L [%], the document is determined to be blank (YES in step ST42), and if it is equal to or greater than H [%], it is determined not to be blank (YES in step ST43). ), Exceeding L [%], H
If it is less than [%], it is determined that it is unclear (NO in step ST43). The judgment formula is the following formula (3).
a [%] ≦ L [%]: Blank paper a [%] ≧ H [%]: Non-blank paper
L [%] <a [%] <H [%]: It is unclear which is ... (3)

Hereinafter, taking the case where the density difference threshold value P is set to 50%, the threshold value L is set to 3%, and the threshold value H is set to 10% using the equations (2) and (3), FIGS. Is explained in detail. Normally, when a blank original with a light ground color as shown in FIG. 4b is read or when a blank original with a dark background color as shown in FIG. 4c is read, it is created because there is almost no difference in density between adjacent pixels. The difference histogram has a shape as shown in FIG. Since there is no density difference exceeding 50 in this histogram, the ratio a [%] is 0% according to the equation (2), which is 3% or less, and is determined as a blank original.

The original as shown in FIG. 5b has many characters and images on the original and the density difference between adjacent pixels is large. Therefore, a difference histogram having a shape as shown in FIG. 5a is created. From FIG. 5a and Formula 2, the ratio a [%] is calculated as 12%. Accordingly, since the ratio a [%] of the value equal to or greater than the threshold value exceeds 10%, it is determined that the document is not blank.

In addition, the difference histogram of a document which is not a blank sheet as shown in FIG. 6B but has few characters and images has a shape as shown in FIG. 6A. If the ratio a [%] of the value that is equal to or greater than the threshold value P is calculated from the histogram by the formula (2), it becomes 5%, and the ratio a [%] exceeds 3% and is less than 10%. Therefore, it is determined that the original is unclear.

In the above-described method for determining a blank document using the difference histogram, the density difference threshold P is set to 50, for example, as shown in FIGS. 4, 5, and 6. Any value may be used as long as it is considered appropriate, and the user may arbitrarily set the value. The ratio L [%] for determining as blank paper and the ratio H [%] for determining that the paper is not blank can be arbitrarily set to values that are considered appropriate for the user to determine blank paper. May be. In this embodiment, the maximum value of the image density is expressed as, for example, 255. However, the present invention is not limited to this and can be applied. Further, in the above-described blank document determination method, the determination is performed based on the ratio of pixels having a density difference equal to or higher than the threshold P. However, the determination is performed using an absolute amount such as the number of pixels having a density difference equal to or higher than the threshold P. May be. In the determination of a blank document based on the difference histogram, the threshold value H is not used, and two determinations, that is, a determination that it is a blank sheet and a determination that it is unclear, may be used as options. Further, in the determination of a blank document based on the difference histogram, the threshold L may not be used, and two determinations, that is, a determination that it is not blank and a determination that it is unclear, may be used as options.

The image data of the document determined to be blank by the first blank document determination method based on the above difference histogram is discarded from the memory 15. If it is determined that the sheet is not blank, the image is temporarily stored in the memory 15 and is processed by the image processing unit 14 according to the image processing command, and then transmitted to the host computer 17 via the communication unit 16. The If the result of the first blank page original determination is unclear, the second blank page original determination described later is performed.

Next, a second blank original determination method for generating a binary image of a document and determining a blank original as shown in steps ST31 to ST32 in FIG. 1a will be described. First, the image data of the document is subjected to main scanning in the horizontal direction and sub-scanning in the vertical direction, and each pixel of the image data of the document is changed to a white pixel when the pixel density is equal to or higher than the threshold value W. If it is less than the threshold value W, it is converted to a black pixel, and a binary image of the document is created (step ST31). Thereafter, the created binary image is scanned for all pixels from left to right and from top to bottom, for example, and a pixel that becomes a change point from a white pixel to a black pixel and a pixel that becomes a change point from a black pixel to a white pixel And count. If the number of pixels serving as the change point is equal to or less than a certain threshold value N, the document is determined to be blank (step ST32).

A method of determining a document as a blank sheet when the ratio of the number of black pixels to the number of pixels of the entire binary image created by the above method or the ratio of the number of black pixels to the number of white pixels is equal to or less than a certain threshold value. It may be used as a two-blank document determination method.

In the first embodiment, as another example of the first blank page original determination process in step ST14 shown in FIG. 1a, a first blank page original is created using a pixel density histogram (hereinafter referred to as a density histogram) created from image data of the original. A method for performing the determination will be described. The image reading apparatus uses the same image reading apparatus 1 as in the first embodiment, and the control system has the same configuration as that shown in FIG. Further, as in the first embodiment, the pixel is exemplified as a brighter pixel as the density value is larger. That is, in the second embodiment, the first blank page original determination method using the difference histogram in the first embodiment is replaced with the following first blank page original determination method using the density histogram.

A method for creating a density histogram will be described. 7a, FIG. 8a, FIG. 9a, and FIG. 10a represent density histograms created by the histogram creation method described later. In these drawings, the horizontal axis represents the density, and the density histogram is created so that the vertical axis represents the number of pixels having the density represented by the horizontal axis. A value of 0 to 255 on the horizontal axis indicates that the pixel is darker as it is closer to 0 and brighter as it approaches 255. The density when the original is read as a color image is a value obtained by averaging the density values of R (red), G (green), and B (blue) by some calculation. In this embodiment, the density when an original is read as a color image uses the average of the RGB density values, but three histograms may be created using the density of each RGB, A histogram for only one of the colors may be created. 7b, FIG. 8b, FIG. 9b, and FIG. 10b each represent a document to be read.

Subsequently, FIG. 1c, FIG. 7, FIG. 8, FIG. 9, and FIG. 10 are illustrated as examples of a method for determining whether it is blank or not from the shape of the created density histogram. While explaining.
First, in the density histograms shown in FIGS. 7a, 8a, 9a, and 10a, a white pixel determination threshold value K is set in advance for the density, and all pixels having a density exceeding the white pixel determination threshold value K are white pixels. To do. Let Qn be the ratio of the number of pixels having density value n to the total number of pixels. Of the pixels determined not to be white pixels, the ratio r [%] of the pixel having the highest density obtained by the following equation (4) (step ST51 in FIG. 1c) indicates which is not blank or blank. It is determined whether or not it is clear.
r [%] = max (Q0, Q1, Q2,..., QK) (4)

The ratio W [%] for determining as blank paper and the ratios B [%] and HW [%] for determining non-blank are obtained in advance by experiments and embedded in the memory 15. Perform blank page original determination. That is, r [%]
Is equal to or less than W [%], the document is determined to be blank (FIG. 1c, ST52). If it is equal to or greater than B [%], it is determined that the document is not blank (FIG. 1c, ST54). Exceeding B [%] or HW [%]
Is exceeded, it is determined that either is unclear (FIG. 1c, ST53, ST54). The judgment formula can be expressed by the following formula 5.
r [%] ≦ W [%]: Blank paper B [%] ≦ r [%] ≦ HW [%]: Non-blank paper
W [%] <r [%] <B [%], or HW [%] <r [%]: either is unclear ... (5)

From now on, taking the case where white pixel determination threshold value K is set to 180, W is set to 1%, B is set to 8%, and HW is set to 80% using Expressions (4) and (5), FIG. 8, FIG. 9, and FIG. 10 will be specifically described. Normally, when a blank original as shown in FIG. 7b is read, when a density histogram is created, a shape as shown in FIG. 7a is obtained. Since there is no pixel having a density lower than 180 which is the white pixel determination threshold value K, r is 0 and is 1% or less, so that it is determined as a blank original by Expression (5). On the other hand, a density histogram as shown in FIG. 8A is created from a document with many characters as shown in FIG. 8B, and r, which is the largest ratio among pixels having a density of 180 or less, is 10% from the equation (4). Calculated. Since r is 8% or more according to Equation (5), it is determined that the document is not blank.

If the original is not blank as shown in FIG. 9b but has fewer characters and images, a density histogram as shown in FIG. 9a is created. Among the pixels having a density of less than 180, r, which is the largest ratio, is calculated as 2% by Expression (4). Since the calculated r exceeds 1% but is also less than 8%, it is determined as unclear whether the document is a blank document or not by Equation (5). Further, for a blank paper document having a dark background color as shown in FIG. 10B, a density histogram as shown in FIG. 10B is created, and r calculated from Expression 4 is 83%. Therefore, since r exceeds 80%, it is determined that it is not clear whether such a document is a blank document or not.

In the present embodiment, the ratio r [%] of pixels having a density that is determined to be a black pixel and that has the largest ratio of the number of pixels is used for the determination of a blank document. Alternatively, the ratio of the number of pixels determined as black pixels to the total number of pixels may be used. Further, it is appropriate for the user to determine a blank page with respect to the white pixel determination threshold K, the ratio W [%] for determining as blank, and the ratios B [%] and HW [%] for determining that the page is not blank. It may be possible to set a value that seems to be. Furthermore, in this embodiment, the maximum value of the image density is represented as 255, but the present invention is not limited to this and can be applied. Further, in the determination of a blank document based on the density histogram, the determination may be made with two determinations, i.e., the determination that the blank document is not blank, and the determination that it is unclear, without using the threshold W, and are included in the present invention. The image data of the document determined to be blank is discarded from the memory 15 by the blank document determination method using the density histogram as the first blank document determination described above. If it is determined that the image is not blank, the image data temporarily stored in the memory 15 is transmitted to the host computer 17 via the communication unit 16 after the image processing unit 13 performs image processing based on the image processing command. The Further, the above-described second blank page original determination shown in ST31 to ST32 in FIG. 1a is performed for a document that is determined to be unclear whether it is blank or not.

When it is desired to detect and discard a blank document in the image reading apparatus, a binary document determination method based on the histogram of the document is performed as in each of the above embodiments, and only when the blank page determination is unclear, binary is used. By using the method of creating an image and determining a blank document, the case of creating a binary image of the document is reduced. As a result, a high-speed blank original determination process can be performed, and the time from reading the original to storing the image data can be shortened. A document that has been erroneously determined to be non-blank in the conventional blank document determination process in which only a blank document determination based on a histogram has been performed can be used to discard image data without erroneous determination by using a blank document determination based on a binary image. This makes it possible to effectively use the storage area. Note that the image data of the blank original may be invalidated instead of being discarded. There are various methods for invalidating image data, such as changing the file name of image data, registering it in a list that manages invalid files, and deleting information for managing files. .

Furthermore, since the above-described histogram creating means can be realized with a small hardware configuration, it can be mounted with almost no influence on the price of the entire apparatus. Further, the above-described blank document determination unit may be performed by an external information processing apparatus such as a host computer instead of the scanner control unit 301. In this case, the information processing apparatus such as a host computer corresponds to the image processing apparatus of the present invention, and the image processing method and program executed in the information processing apparatus such as the host computer correspond to the image processing method and program of the present invention. In addition, the white paper and the white paper in the present invention are originals such as white paper, paper that has a substantially uniform background color and is not printed, black paper that has almost no characters, etc. It refers to an unnecessary document with a small amount of information on the screen, and is not limited to an actual white document and actual white paper.

The present invention has been described in detail with respect to an image reading apparatus such as a scanner. However, the present invention can also be applied to an image forming apparatus such as a copying machine or a printer complex machine, or other apparatuses having an image reading function. However, the configuration described in the above embodiment should be appropriately changed depending on the configuration of the apparatus to which the invention is applied and various conditions, and is not intended to limit the scope of the present invention to the above embodiment. .

Flow chart showing a series of flows for an image processing method Flowchart of blank document determination processing Flowchart of blank document determination processing Schematic showing the internal structure of the scanner of the image reading apparatus A block diagram illustrating a schematic configuration of a control system of an image reading apparatus A block diagram illustrating a schematic configuration of a control system of an image reading apparatus Difference histogram for non-blank documents Difference histogram of a document that is determined to be unclear whether it is blank Density histogram of blank document Density histogram of a non-blank document Density histogram of a document that is determined to be unclear whether it is blank Density histogram of blank document The figure which showed the method of calculating the difference histogram

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Image reading device 2 ... Pick-up roller 3 ... Feeding roller 4 ... Separation roller 5 ... Registration roller pair 6a ... Contact glass 6b ... Contact glass 7a ... Line Image sensor 7b ... Line image sensor 8 ... Conveying roller pair 9 ... Counter member 10 ... Document detection sensor 33 ... Front side image reading unit 34 ... Back side image reading unit 301 ... Scanner Control unit 11 ... image reading unit 12 ... A / D conversion unit 13 ... image processing unit 14 ... CPU
15 ... Memory 16 ... Communication unit 17 ... Host computer

Claims (11)

A first blank document determination unit for determining whether or not the document is a blank document by a first determination method including a case where it is not determined whether or not the document is a blank document based on image data read from the document; ,
A second determination method different from the first determination method is used to determine whether or not the original is a blank original for an original that has not been determined by the first blank original determination unit. 2 blank paper original judging means ,
The first blank original determination unit includes:
A histogram creating means for creating a histogram corresponding to the document based on the multi-value image data corresponding to the document;
Means for determining whether the document is a blank document based on the generated histogram;
With
The second blank page original determination means includes:
Means for creating binary image data from the multi-value image data corresponding to the original;
Means for determining whether or not the original is a blank original based on the generated binary image data;
The image processing apparatus according to claim Rukoto equipped with. By the first blank document determination unit, the document prior to the determination of whether a blank document, image data read from the document is further provided with a judging means for judging whether the binary image data The image processing apparatus according to claim 1. When the determination unit determines that the image data is binary image data , the second blank page document determination unit determines that the document is a blank page document without performing the determination by the first blank page document determination unit. The image processing apparatus according to claim 2 , wherein the image processing apparatus determines whether or not the image processing is in progress. 2. The image processing according to claim 1 , wherein the histogram creating unit creates a difference histogram indicating a distribution of density differences between a pixel of interest of multi-valued image data corresponding to the document and a neighboring pixel. apparatus. The image processing apparatus according to claim 1 , wherein the histogram creating unit creates a density histogram indicating a density distribution of each pixel of the multivalued image data corresponding to the document. Any one of claims 1 to 5, characterized in that it further comprises means for the image data, discarding or disabling of the document is determined to blank original by said first blank document page determination means or said second blank document determination unit The image processing apparatus according to item 1.   The image processing apparatus according to claim 1, wherein the original to be determined by the first blank original determining unit includes paper having a ground color having a substantially uniform density. .   An image reading apparatus for reading an image of a document,
  A first blank document determination unit for determining whether or not the document is a blank document by a first determination method including a case where it is not determined whether or not the document is a blank document based on image data read from the document; ,
  A second determination method different from the first determination method is used to determine whether or not the original is a blank original for an original that has not been determined by the first blank original determination unit. 2 blank paper original judging means and
With
  The first blank original determination unit includes:
  A histogram creating means for creating a histogram corresponding to the document based on the multi-value image data corresponding to the document;
  Means for determining whether the document is a blank document based on the generated histogram;
With
  The second blank page original determination means includes:
  Means for creating binary image data from the multi-value image data corresponding to the original;
  Means for determining whether or not the original is a blank original based on the generated binary image data;
An image reading apparatus comprising:   An image reading system comprising an image reading device for reading an image of a document, and an information processing device connected to the image reading device,
  The information processing apparatus includes:
  A first determination method including a case where it is not determined whether or not the document is a blank document based on image data read from the document by the image reading device. A blank document determination means;
  A second determination method different from the first determination method is used to determine whether or not the original is a blank original for an original that has not been determined by the first blank original determination unit. 2 blank paper original judging means and
With
  The first blank original determination unit includes:
  A histogram creating means for creating a histogram corresponding to the document based on the multi-value image data corresponding to the document;
  Means for determining whether the document is a blank document based on the generated histogram;
With
  The second blank page original determination means includes:
  Means for creating binary image data from the multi-value image data corresponding to the original;
  Means for determining whether or not the original is a blank original based on the generated binary image data;
An image reading system comprising: A first blank document determination step for determining whether or not the document is a blank document by a first determination method including a case where it is not determined whether or not the document is a blank document based on image data read from the document; ,
Whether or not the original is not a blank original by the first blank original determination means is determined by a second determination method different from the first determination method to determine whether or not the original is a blank original. A second blank original determination step ,
The first blank document determination step includes
Creating a histogram corresponding to the document based on multi-valued image data corresponding to the document;
Determining whether the document is a blank document based on the created histogram;
With
The second blank document determination step includes
Creating binary image data from the multi-value image data corresponding to the document;
Determining whether or not the document is a blank document based on the generated binary image data;
Image processing method according to claim Rukoto equipped with. A first blank original determination step for determining whether or not the original is a blank original by a first determination method including a case where it is not determined whether or not the original is a blank original based on image data read from the original; ,
Whether or not the original is not a blank original by the first blank original determination means is determined by a second determination method different from the first determination method to determine whether or not the original is a blank original. An image processing program for causing a computer to execute a second blank sheet original determination step ,
The first blank original determination step includes
Creating a histogram corresponding to the document based on the multi-value image data corresponding to the document;
Determining whether or not the document is a blank document based on the created histogram;
Including
The second blank original determination step includes
Creating binary image data from the multi-value image data corresponding to the document;
Determining whether the document is a blank document based on the generated binary image data;
Image processing program characterized in that it comprises a.

Images