JP2009042814A - Image evaluation device and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To execute image evaluation including the detection of a linear noise without providing any image as a reference for inclination correction. <P>SOLUTION: This image evaluation device acquires image data showing an image to be evaluated (S1). The image to be evaluated is provided with a side formed at an angle determined to a sheet. It is desired that there are several sides. The image evaluation device specifies the sides from the acquired image data (S4), and calculates the inclination of the side (S5). Afterwards, the image evaluation device corrects the inclination of the image data based on the calculated inclination (S6), and calculates an evaluation value showing the evaluation result of the image (S7). <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image evaluation apparatus and a program.

  An image formed by the printer may have a linear (streaky) image defect (hereinafter referred to as “noise” for convenience). The cause of the occurrence of linear noise differs depending on the recording method of the printer. And so on. Since the linear noise is continuous as compared with the dot noise or the like, it is easy for the human eye to recognize it as noise even if the color difference from the surroundings is slight. On the other hand, linear noise is difficult to mechanically recognize as noise when the color difference from the surroundings is slight.

  As a method of mechanically recognizing linear noise, there is a method of averaging colors in a direction along the noise straight line by utilizing the fact that the noise is linear. FIG. 11 is a diagram for explaining this method, and shows the gradation value of each pixel of certain image data. In the figure, a rectangular area in which a value of “254” or “255” is described is a pixel, and a value described in the pixel is a gradation value. Further, in the same figure, the linear noise is represented by pixels whose gradation value is “254” continuous in the direction indicated by the arrow A.

  In the image data shown in FIG. 11, when the gradation values are added in the direction indicated by the arrow A and averaged, the average value has a significant difference between a portion that is linear noise and a portion that is not. For example, the average of the gradation values in the fifth column is “254”, but the average of the gradation values in the other first to fourth columns and the sixth to ninth columns is “254.9” or “255”. In this way, the process of detecting linear noise by averaging the gradation values of pixels in a certain direction is hereinafter referred to as “averaging process”.

  The averaging process is an effective process when the direction of the noise straight line is clear. However, when an image formed by a printer is evaluated, a document may be tilted when it is read by a scanner, which may cause a shift in the noise direction. In such a case, even if the deviation is slight, there is a possibility that linear noise cannot be detected by the averaging process, so it is desirable to correct the inclination of the image.

As a technique for correcting the inclination of an image, a technique is known in which a reference line or mark is included in an image and read to correct a deviation from the reference (see, for example, Patent Document 1).
JP-A-2-35869

However, when evaluating an image formed by a printer, it may be desired to evaluate the entire image area. In such a case, if an image serving as a reference is provided, there is a problem that it is impossible to evaluate that portion.
Therefore, an object of the present invention is to perform image evaluation including detection of linear noise without providing an image serving as a reference for inclination correction.

  An image evaluation apparatus according to the present invention is an image to be evaluated formed on a sheet-like object, and has an edge having an angle determined with respect to a predetermined direction of the sheet-like object, and at least an area to be detected Acquisition means for acquiring image data representing an image to be evaluated formed so that uniform colors are continuous in the predetermined direction; and the side of the image to be evaluated from the image data acquired by the acquisition means And calculating means for calculating the inclination of the side from the determined angle, correction means for correcting the inclination of the image data based on the inclination calculated by the calculation means, and inclination by the correction means Detecting means for detecting, as noise, an area that is continuous in the predetermined direction and has a color different from the formed color from the image to be evaluated represented by the image data corrected for To.

In the image evaluation apparatus according to the present invention, the image to be evaluated has a plurality of the sides, the calculation unit calculates inclinations of the plurality of sides, and the correction unit calculates the plurality of inclinations. A configuration for correcting the inclination of the image data based on the image data may be employed.
In this case, the calculation unit further includes an exclusion unit that excludes a part of the plurality of sides included in the evaluated image, and the correction unit includes the plurality of the plurality of sides included in the evaluated image. You may employ | adopt the structure which correct | amends the inclination of the said image data based on the inclination calculated from the edge except the edge excluded by the said exclusion means among the edges.
In this case, the excluding unit may further include a slope calculated by the calculating unit that is greater than a value obtained by adding a standard deviation to an average value of the slopes of the plurality of sides, or a slope of the plurality of sides. If the average value is smaller than the value obtained by subtracting the standard deviation, a configuration may be adopted in which the side that becomes the slope is excluded.

In the image evaluation apparatus according to the present invention, the image data includes a plurality of images to be evaluated having different colors, and the acquisition unit generates a plurality of pieces of separated image data having different color components from the image data. The calculation means may adopt a configuration for specifying the side of the image to be evaluated from each of a plurality of pieces of separation image data generated by the color separation means.
In this case, the image processing apparatus further includes a reduction unit that subtracts the number of gradations of the plurality of separated image data generated by the color separation unit by a threshold value corresponding to a color component, and the calculation unit includes the number of gradations by the reduction unit. A configuration may be adopted in which the side is specified from each of the decomposed image data in which is reduced. Here, the reduction means typically performs a so-called binarization process, but may be any process as long as it reduces the number of gradations.
Furthermore, it is more desirable that the subtracting means further comprises a threshold value calculating means for calculating the threshold value for each of a plurality of color components from a plurality of separated image data generated by the color separating means.

  In the image evaluation apparatus according to the present invention, the calculation unit includes an extraction unit that extracts a region near the side from the image data, and the inclination of the side based on the image data of the region extracted by the extraction unit. A configuration may be adopted in which

  The present invention can also be provided as a program for realizing such an image evaluation apparatus or a storage medium storing this program. For example, the program according to the present invention should detect a computer that is an image to be evaluated formed on a sheet-like object and has a side having an angle determined with respect to a predetermined direction of the sheet-like object. An acquisition unit that acquires image data representing an image including an image to be evaluated formed so that a uniform color is continuous in at least the predetermined direction for the region; an image of the image to be evaluated from the image data acquired by the acquisition unit Calculation means for specifying the side and calculating the inclination of the side from the determined angle, correction means for correcting the inclination of the image data based on the inclination calculated by the calculation means, and the correction means Function as detection means for detecting an area that is continuous in the predetermined direction and has a color different from the formed color from the image to be evaluated represented by the image data whose inclination is corrected by It is intended to.

  According to the present invention, it is easier to detect linear noise without providing an image serving as a reference for tilt correction in an image area to be evaluated, compared to a case without this configuration.

[First Embodiment]
FIG. 1 is a block diagram showing a configuration of an image evaluation apparatus 100 according to the first embodiment of the present invention. As illustrated in FIG. 1, the image evaluation apparatus 100 includes an image reading unit 10, a control unit 20, and a notification unit 30.

  The image reading unit 10 optically reads a sheet-like object (for example, paper) on which an image is formed by a printer, that is, a document, and generates image data representing an image on the surface of the document. The image data generated by the image reading unit 10 is, for example, data including pixels having a gradation value of 8 bits (that is, 256 gradations) for each color component of R (red), G (green), and B (blue). It is. As the image reading unit 10, for example, a so-called flatbed scanner that reads an original placed on a platen glass can be used, but a drum scanner, a colorimeter, a digital still camera, or the like may be used.

  The control unit 20 includes an interface connected to the image reading unit 10 and the notification unit 30 in addition to an arithmetic device such as a CPU (Central Processing Unit) and a memory such as a ROM (Read Only Memory) and a RAM (Random Access Memory). The control unit 20 acquires image data, performs various arithmetic processes on the acquired image data by executing a program stored in the ROM, and detects linear noise. When detecting linear noise, the control unit 20 outputs detection information, which is information indicating that, to the notification unit 30. Note that the ROM of the control unit 20 stores an angle of an image to be evaluated which will be described later.

  The notification unit 30 includes means for notifying that a document includes linear noise by generating sound or light, and performs notification according to the acquired detection information. As the notification unit 30, for example, a liquid crystal display or a speaker may be used, or a lamp or a buzzer may be used more simply.

  Here, the document in the present embodiment will be described. In the present embodiment, a document refers to a sheet on which an image is formed by an electrophotographic printer. Here, an electrophotographic printer is a printer that has a drum-shaped photoconductor and forms an image on a conveyed sheet through processes such as charging, exposure, development, transfer, and fixing. In the embodiment, it refers to a color printer that forms an image using toner of each color of C (cyan), M (magenta), Y (yellow), and K (black).

  Further, the image represented by the document according to the present embodiment includes images to be evaluated for CMYK colors. In the present embodiment, the image to be evaluated refers to a rectangular image formed in a predetermined direction on a rectangular sheet. Each side of the image to be evaluated is assumed to be parallel (or perpendicular) to each side of the sheet-like object. Further, the image to be evaluated is formed so that the color inside thereof is uniform.

  FIG. 2 is a diagram illustrating an example of a document according to the present embodiment. As shown in the drawing, on the document D, evaluation images ImC, ImM, ImY, and ImK for each color of CMYK are formed. Note that the size of the document D is, for example, A4 size (297 mm × 210 mm). The evaluated images ImC, ImM, ImY, and ImK are formed over the entire area where the printer can form an image. In addition, the document D has a blank area in which an image to be evaluated is not formed.

  Note that the document D is an image on which an image is formed while being conveyed in the direction indicated by the arrow A in the drawing. That is, the direction indicated by the arrow A is a direction orthogonal to the axial direction of the photoconductor during image formation (see FIG. 3). In FIG. 3, the photoconductor is a cylindrical member indicated by a symbol PC.

  The image to be evaluated is an image formed so as to have a uniform color as a whole, but what is actually formed on the paper may be microscopically non-uniform. For example, in the electrophotographic system, the toner is scattered or the transfer or fixing is incomplete, so that the color becomes darker or lighter than originally intended. Further, the above-described linear noise may occur in the image to be evaluated.

In the present embodiment, linear noise is noise that is generated in parallel with the paper transport direction (the direction indicated by arrow A in FIG. 2). There are multiple causes of this noise. For example, the toner remaining on the photoconductor may not be completely removed due to a defective cleaning member such as a blade, or the surface may be scratched in the rotation direction when the photoconductor rotates. This is due to being attached. In such a case, the linear noise is formed by a defect indicated by a symbol B in FIG. In the present embodiment, such noise is a detection target.
In the electrophotographic system, linear noise can also occur during transfer and fixing.

  With the above configuration, the image evaluation apparatus 100 reads the paper on which the image to be evaluated is formed, determines the presence or absence of linear noise, and notifies that when linear noise is detected. To do. The processing executed by the image evaluation apparatus 100 at this time is as follows.

  FIG. 5 is a flowchart showing the operation of the image evaluation apparatus 100 of the present embodiment. As shown in the figure, first, the control unit 20 of the image evaluation apparatus 100 acquires image data from the image reading unit 10 (step S1). Next, the control unit 20 extracts an end region from the image data (step S2). Here, the “end region” is a region including the side of the image to be evaluated and its vicinity. Here, four regions ArU, ArD, ArL, and ArR shown in FIG. 6 are extracted as end regions. Note that the end regions ArU, ArD, ArL, and ArR may partially overlap.

  Next, the control unit 20 decomposes the image data into R, G, and B color components to obtain single-color (plane) image data (step S3). This monochrome image data is hereinafter referred to as “decomposed image data”. When the decomposed image data is obtained, the control unit 20 specifies the side of the image to be evaluated using each of the decomposed image data (step S4). In the present embodiment, the sides of the image to be evaluated are specified by combining binarization processing and Hough transformation. The binarization process is performed in order to further increase the specific accuracy of the straight line (that is, the side of the image to be evaluated) by the Hough transform. When a straight line is specified by the Hough transform, it is desirable to use only the end region in consideration of the processing speed and accuracy. A known method may be used as a method for identifying a straight line by Hough transform.

Here, the purpose of the binarization process and its specific method will be described.
FIG. 7 is a schematic diagram illustrating an enlarged side of an image to be evaluated. As shown in the figure, the boundary between the region where the image is not formed and the region where the image is not formed is often not a straight line as originally intended. This is because, for example, in the case of an electrophotographic system, toner is scattered or transfer or fixing is incomplete. Even if the Hough transform is performed on such an image as it is, there is a high possibility that the accuracy of specifying a straight line is lowered. Therefore, in this embodiment, the binarization process is performed before the Hough transform.

The control unit 20 calculates a threshold value in order to perform binarization processing. Further, the control unit 20 calculates a frequency distribution for the gradation values of the decomposed image data in order to calculate a threshold value.
FIG. 8 is a diagram illustrating the decomposed image data ImR, ImG, and ImB, and histograms (frequency distribution diagrams) HgR, HgG, and HgB of the gradation values. As shown in the figure, each histogram has two peaks when the image to be evaluated as in this embodiment is used. The control unit 20 determines, as a threshold value, a gradation value that is between the two peaks and has a frequency that is minimal between the peaks. For example, in the case of the histogram illustrated in FIG. 8, the threshold values are ThR, ThG, and ThB, respectively.

  Returning to the flowchart of FIG. When specifying the side of the image to be evaluated, the control unit 20 calculates the inclination of this side (step S5). At this time, the control unit 20 compares the angle of the side specified from the decomposed image data with the angle of the corresponding side stored in advance in the ROM, thereby determining how much the side of the decomposed image data is inclined from the determined angle. To identify. In addition, when there are a plurality of sides specified from the decomposed image data, the control unit 20 calculates the inclination of each side.

  Subsequently, the control unit 20 corrects the inclination of the image represented by the image data based on the calculated inclination (step S6). Specifically, for example, the control unit 20 corrects the inclination of the image data by executing rotation processing of a predetermined correction angle. When the inclination is calculated from a plurality of sides, the control unit 20 corrects the inclination of the image to be evaluated by an angle obtained by averaging the values of the plurality of inclinations. For example, when the inclinations of the two specified sides are “1.4 °” and “1.6 °” (both are clockwise as positive), the control unit 20 rotates the image data counterclockwise. Rotate “1.5 °” to correct tilt.

  When the inclination of the image data is corrected in this way, the control unit 20 refers to the corrected image data and calculates an evaluation value of the image data (step S7). The “evaluation value” here is an average value of gradation values for pixels in each column, which is obtained by performing the above-described averaging process on the corrected image data. The average value is preferably calculated separately for each area corresponding to the image to be evaluated for each color, but may be calculated for all areas except the blank area.

  Subsequently, the control unit 20 determines whether or not linear noise is detected from the decomposed image data (step S8). This determination is made based on, for example, whether or not there is a region (column) in which the above-described average value differs from the other region (column) by a predetermined value or more. Note that the linear noise detected here may be a so-called white streak (noise lighter than the surroundings), a so-called black streak (noise darker than the surroundings), or both. There may be.

  When linear noise is detected (step S8: YES), the control unit 20 outputs detection information and ends the process shown in the flowchart of FIG. 5 (step S9), and linear noise is detected. If not detected (step S8: NO), the process of step S9 is omitted (skip), and the process shown in the flowchart of FIG. 5 is terminated. Instead of outputting the detection information, the control unit 20 outputs the evaluation value described above, or outputs information indicating at which position of the image data the linear noise is generated. May be.

As described above, according to the image evaluation apparatus 100 of the present embodiment, since the inclination of the image data is corrected based on the inclination of the side of the image to be evaluated, an image serving as a reference for inclination correction is not separately provided. It becomes possible to detect linear noise for the entire evaluation image.
In addition, the image evaluation apparatus 100 can improve the speed and accuracy of specifying a side by specifying the side of the extracted end region, compared to the case of specifying the side using the entire image data. it can.

  Further, the image evaluation apparatus 100 can specify the side angle more accurately than the case where the image evaluation apparatus 100 does not have this configuration by specifying the side from the decomposed image data. For example, when binarization processing is performed without separating image data into single colors, the binarization threshold may differ for each color component. As a result, binarization is performed more than when image data is decomposed. There is a possibility that the accuracy of processing is inferior. In particular, this tendency is remarkable in an image to be evaluated whose color approximates the color of the blank area (usually white). For example, in this embodiment, the image to be evaluated is yellow.

[Second Embodiment]
Subsequently, a second embodiment of the present invention will be described. In the present embodiment, a part of the operation of the first embodiment described above is changed. Therefore, in the present embodiment, this change will be mainly described. In the following description, the points common to the first embodiment will be omitted as appropriate, and the same reference numerals as those in the first embodiment will be used.

  FIG. 9 is a flowchart showing the operation of the image evaluation apparatus 100 of this embodiment. This flowchart is the same as the flowchart shown in FIG. 5 except that there is an exclusion process (step SA) between steps S5 and S6. In short, the exclusion process is a process of excluding, from the target, a side whose slope is significantly different from the other sides.

  FIG. 10 is a flowchart showing specific contents of the exclusion process. In this exclusion process, first, the control unit 20 calculates the average value (Ave) of the slope calculated in step S5 (step SA1). Note that when calculating the average value of the slope, either clockwise or counterclockwise is positive, and a positive or negative (or 0) average value is calculated. After calculating the average value, the control unit 20 calculates the standard deviation (σ) using the slope value used to calculate the average value (step SA2).

Next, the control unit 20 uses the calculated average value (Ave) and standard deviation (σ) to determine the gradient g (i) (where i = 1 to n, where n is the total number of calculated gradients). Whether or not the following inequality (1) is satisfied. That is, the control unit 20 determines whether or not each value of the calculated slope is equal to or larger than a value obtained by subtracting the standard deviation from the average value and less than a value obtained by adding the standard deviation from the average value ( Step SA3).
Ave−σ ≦ g (i) ≦ Ave + σ (1)

  If there is a slope g (i) that does not satisfy the above inequality (1) (step SA3: YES), the controller 20 excludes the value (step SA4). Here, excluding a certain value may be, for example, an operation of erasing from the RAM, but it is sufficient if the value does not affect the subsequent processing. Note that if the slope g (i) does not satisfy the above inequality (1) (step SA3: NO), the control unit 20 omits the process of step SA4.

  Thereafter, the control unit 20 executes the process of step S6, that is, the process of correcting the inclination of the image data. At this time, the control unit 20 calculates a correction angle by using only the slope value satisfying the inequality (1) and excluding the slope value not satisfying the inequality (1). The subsequent processing executed by the control unit 20 is the same as that in the first embodiment.

  As described above, according to the image evaluation apparatus 100 of the present embodiment, it is possible to calculate a correction angle by excluding a side whose inclination is significantly different from other sides. For example, when the document is read at an angle, there is a high possibility that all sides are inclined at substantially the same angle. On the other hand, if another side is tilted at almost the same angle, but one side is tilted at a different angle, there is a possibility that there was an abnormality in image formation or that the side is not a side of the image to be evaluated And so on. The image evaluation apparatus 100 according to the present embodiment can calculate the correction angle more accurately by excluding such a side from the reference for calculating the correction angle.

[Modification]
The present invention is not limited to the above-described embodiment, and can be implemented by other aspects. Hereinafter, other embodiments of the present invention will be described. Note that the following modifications may be combined as appropriate.

(1) Modification 1
The image to be evaluated need not be rectangular as long as the angle of the side is determined in advance. However, since the image to be evaluated is for the purpose of detecting linear noise, it is necessary that the image to be evaluated be formed so that the colors are continuously continuous with respect to the direction of the noise to be detected.
Further, the number of images to be evaluated formed on the document is not limited to that described above. For example, if the printer that forms the document is a monochrome printer, it is sufficient that one image to be evaluated is formed on the entire surface of the document. In such a case, the image data does not have to be decomposed image data.

  Further, the color of the image to be evaluated is arbitrary. In the above-described embodiment, the toner color of the printer, that is, the primary color is used. However, an image to be evaluated such as a secondary color may be used. For example, a yellow image to be evaluated may be mixed with cyan to make green because it is difficult to determine noise with a single color.

(2) Modification 2
When specifying the side of the image to be evaluated, the process of extracting the edge region (step S2) and the process of generating the decomposed image data (step S3) are not essential. This is because the sides can be specified from the entire image data, and the sides can be specified from image data that has not been decomposed. However, it can be said that it is more desirable to perform these processes in order to specify the sides accurately.

(3) Modification 3
The image evaluation apparatus according to the present invention may be an apparatus independent of a printer or the like, or may be an apparatus provided integrally with the printer. The image evaluation apparatus according to the present invention may be implemented by a general-purpose information processing apparatus such as a personal computer. In this case, what is necessary is just to make an information processing apparatus run the program which described the content of the process which the image evaluation apparatus which concerns on this invention performs.
The program describing the contents of processing executed by the image evaluation apparatus according to the present invention may be provided in a form stored in a recording medium such as an optical disk, or provided in a form downloaded from a network such as the Internet. May be.

It is a block diagram which shows the structure of the image evaluation apparatus which concerns on this invention. It is a figure which shows an example of a manuscript. FIG. 5 is a diagram illustrating a document conveyance direction during image formation. It is a figure which shows the defect of the photoreceptor which causes generation | occurrence | production of a linear noise. It is a flowchart which shows operation | movement of an image evaluation apparatus. It is a figure which illustrates an edge part field. It is the schematic diagram which expanded and illustrated the side with a to-be-evaluated image. It is the figure which illustrated decomposition | disassembly image data and the histogram of the gradation value. It is a flowchart which shows operation | movement of an image evaluation apparatus. It is a flowchart which shows the exclusion process which an image evaluation apparatus performs. It is a figure for demonstrating an averaging process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Image evaluation apparatus, 10 ... Image reading part, 20 ... Control part, 30 ... Notification part

Claims (9)

An image to be evaluated formed on a sheet-like object, having a side having an angle determined with respect to the predetermined direction of the sheet-like object, and a uniform color continuously in at least the predetermined direction for a region to be detected Acquisition means for acquiring image data representing an image including the image to be evaluated formed so as to
Calculation means for specifying the side of the image to be evaluated from the image data acquired by the acquisition means, and calculating the inclination of the side from the determined angle;
Correcting means for correcting the inclination of the image data based on the inclination calculated by the calculating means;
Detecting means for detecting an area that is continuous in the predetermined direction and has a color different from the formed color, from an image to be evaluated represented by image data whose inclination is corrected by the correction means. Image evaluation device. The evaluated image has a plurality of the sides,
The calculating means includes
Calculating the inclination of each of the plurality of sides;
The correction means includes
The image evaluation apparatus according to claim 1, wherein the inclination of the image data is corrected based on the plurality of calculated inclinations. The calculating means includes
An exclusion means for excluding a part of the plurality of sides included in the image to be evaluated;
The correction means includes
The inclination of the image data is corrected based on an inclination calculated from an edge excluding the edge excluded by the exclusion unit among the plurality of edges included in the image to be evaluated. The image evaluation apparatus according to 1. The exclusion means includes
The slope calculated by the calculating means is larger than the value obtained by adding the standard deviation to the average value of the slopes of the plurality of sides, or the value obtained by subtracting the standard deviation from the average value of the slopes of the plurality of sides. The image evaluation apparatus according to claim 3, wherein, when smaller, an edge having the inclination is excluded. The image data includes a plurality of the evaluated images having different colors,
The acquisition means includes
Color separation means for generating a plurality of separated image data having different color components from the image data;
The calculating means includes
The image evaluation apparatus according to claim 1, wherein the side of the image to be evaluated is specified from each of a plurality of pieces of separated image data generated by the color separation unit. A reduction means for reducing the number of gradations of the plurality of separation image data generated by the color separation means by a threshold value corresponding to the color component;
The calculating means includes
The image evaluation apparatus according to claim 5, wherein the side is specified from each of the decomposed image data in which the number of gradations is reduced by the reduction means. The decrementing means is:
The image evaluation apparatus according to claim 6, further comprising a threshold value calculation unit that calculates the threshold value for each of a plurality of color components from a plurality of separated image data generated by the color separation unit. The calculating means includes
Comprising extraction means for extracting a region near the side from the image data;
The image evaluation apparatus according to claim 1, wherein the inclination of the side is calculated based on image data of a region extracted by the extraction unit. Computer
An image to be evaluated formed on a sheet-like object, having a side having an angle determined with respect to the predetermined direction of the sheet-like object, and a uniform color continuously in at least the predetermined direction for a region to be detected Acquisition means for acquiring image data representing an image including an image to be evaluated formed so as to
A calculating unit that identifies the side of the image to be evaluated from the image data acquired by the acquiring unit, and calculates an inclination of the side from the determined angle;
Correcting means for correcting the inclination of the image data based on the inclination calculated by the calculating means;
A program for functioning as a detection unit that detects an area that is continuous in the predetermined direction and has a color different from the formed color from an image to be evaluated represented by image data whose inclination is corrected by the correction unit.

Images