JP2004007265A - Apparatus for eliminating creasing noise in image - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To eliminate creasing noise created in an original image in the vertical and horizontal directions based on the remains of creases in manuscript paper. <P>SOLUTION: A scanner 6 reads a landscape form (or a portrait form) original and outputs an original image with binary values. A CPU 1 creates a means for setting a searching range for creasing noise of a predetermined width in the creasing direction of the original image at the creasing point formed based on the predetermined creasing way for the manuscript paper, a means for extracting a square area in which run-lengths with more than predetermined lengths coupling respectively pixels with predetermined pixel values in the horizontal direction for a lateral direction searching range and in the vertical direction for a longitudinal searching range in the searching range are integrated, and a means for eliminating predetermined pixels in the square area by a creasing noise elimination program downloaded through a CD-ROM drive 3 and stored in a RAM 2, so as to eliminate the creasing noise through these means. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a fold noise removing device for an image, an image forming apparatus provided with the noise removing device, a fold noise removing program, and a recording medium.
[0002]
[Prior art]
2. Description of the Related Art A document such as a drawing created in a portrait form or a landscape form on a large size manuscript sheet is always folded and stored in a four-fold, six-fold, eight-fold, or the like.
FIG. 10 is a diagram illustrating an example of folding a document created in a portrait form into six and eight folds. In the case of six fold, the document is folded into two vertically and three horizontally (see FIG. 10A )), In the case of octal folding, it is folded into two vertically and four horizontally (FIG. 10B). Although not shown, when a landscape-shaped original is folded in six, it is folded in three in the vertical direction and two in the horizontal direction. In the case of eight-fold, four in the vertical direction and in the horizontal direction. Folds in two.
If the folding of such a document is repeated many times, the traces of the folds 21 and 22 become strong. When the document having a strong fold trace is read by an image forming apparatus such as an electrophotographic copying machine, the fold traces are read. In some cases, line noise is formed on transfer paper as a line noise composed of a collection of black fine linear noises. Although this line noise causes a decrease in the image quality of a print image and is unnecessary, it has not been removed.
[0003]
Conventionally, an invention disclosed in JP-A-9-238208 is known as an invention for removing line noise. According to the present invention, in character recognition of an image sent from a facsimile device, a defect of a transmission-side facsimile device, for example, a failure of an image sensor or a dust attached to an image sensor causes vertical line noise. It removes the input noise, detects noise from all the pixels on the vertical line of the document image, and removes the vertical line noise.
As described above, the present invention aims to remove vertical line noise generated due to a defect of a facsimile apparatus, and therefore, this vertical line noise may be generated in various parts of an input image. For this reason, noise is detected from all pixels constituting the vertical line of the input document image. For this reason, the detection device itself has a complicated configuration, which not only takes a long time for detection but also inevitably does not detect horizontal line noise. In addition, the necessary information in the vertical line noise portion is not removed, and the removed required information is not supplemented or displayed separately from the vertical line noise and other noise.
[0004]
[Problems to be solved by the invention]
SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and a first object of the present invention is to remove vertical and horizontal fold noise generated in a document image based on fold marks of a document sheet. The second object is to prevent the necessary image information from being removed at that time. The third object is to interpolate the necessary image information that has been removed, the fourth object is to remove the fold noise while leaving the necessary image information, and the fifth object Is to allow the fold noise to be removed to be recognized separately from other noise.
[0005]
[Means for Solving the Problems]
The invention according to claim 1 is an apparatus for removing, as fold noise, a fold image formed as a binary document image based on a folding trace of a document sheet, wherein a document sheet is provided for each landscape-type document image or portrait-type document image. Means for setting a fold noise search range of a predetermined width in the fold direction of the original image at the position of the fold formed based on the predetermined fold method, and within the fold noise search range, Means for extracting a rectangular area integrating run lengths equal to or longer than a predetermined length to which pixels having predetermined pixel values are connected in the horizontal direction and in the vertical direction in a vertical search range, and a predetermined pixel in the rectangular area; And an image fold noise elimination device, comprising: means for removing an image.
[0006]
According to a second aspect of the present invention, in the image fold noise removing apparatus according to the first aspect, the means for setting the fold noise search range sets a noise search range having a different width for each size of the document paper. This is a device for removing fold noise of an image to be folded.
[0007]
According to a third aspect of the present invention, in the image fold noise removing apparatus according to the first or second aspect, the means for removing a predetermined pixel in the rectangular area includes a pixel value of all pixels having a predetermined pixel value in the rectangular area. , Or inverts only a run-length pixel value having a predetermined length or more connected to a pixel having a predetermined pixel value.
[0008]
According to a fourth aspect of the present invention, in the image fold noise removing apparatus according to any one of the first to third aspects, when the line segment image traversing the fold noise search range is removed by removing the fold noise, the image is removed. An image fold noise removing device comprising means for interpolating a line segment image.
[0009]
According to a fifth aspect of the present invention, in the image fold noise removing device according to any one of the first to fourth aspects, the search range of the fold noise and / or the length of a run length equal to or longer than a predetermined length is variable. An image fold noise removing device characterized by the following.
[0010]
According to a sixth aspect of the present invention, there is provided an image forming apparatus including the image fold removing device according to any one of the first to fifth aspects.
[0011]
According to a seventh aspect of the present invention, there is provided the image forming apparatus according to the sixth aspect, further comprising means for displaying a fold noise formed on the document image in a display format different from that of other noises. It is.
[0012]
The invention according to claim 8 is a method for removing, as fold noise, a fold image formed in a binary original image based on a folding trace of the original paper, wherein the landscape original document image or the portrait original image is replaced with the original paper image. Setting a fold noise search range of a predetermined width in the fold direction of the original image at the position of the fold formed based on the predetermined fold method, and within the fold noise search range, Horizontally and vertically in a vertical search range, extracting a rectangular area integrating run lengths equal to or longer than a predetermined length, each of which is connected to a pixel of a predetermined pixel value; and And a method for removing fold noise of an image.
[0013]
According to a ninth aspect of the present invention, there is provided a program for causing a computer of an image forming apparatus to execute the procedure of each step described in the eighth aspect.
[0014]
According to a tenth aspect of the present invention, there is provided a computer-readable recording medium storing the program according to the ninth aspect.
[0015]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, a method for removing fold noise according to the present invention will be described.
FIG. 1 is a block diagram of a main part of an image forming apparatus in which a fold noise removing method according to an embodiment of the present invention is performed. In the figure, a CPU (Central Processing Unit) 1 controls the entire apparatus. In addition, a central processing unit (ROM or RAM 2) for removing fold noise and performing interpolation processing of the removed image is provided with a fold noise downloaded from a CD-ROM (not shown) by the overall control program of the image forming apparatus and the CD-ROM drive 3. A recording device for storing various programs in addition to the removal program, and a work area RAM 4 temporarily store an image read as a binary image by the scanner 6 for removing fold noise or an image already read and stored in the memory 5. The recording device for storing, the display 7, displays an image read by the scanner 6, or is included in the image. It is a device for displaying the eye noise Ri.
[0016]
FIG. 2 is a diagram illustrating an overall flow of a process for removing noise according to the embodiment of the present invention.
The processing for removing fold noise will be described with reference to FIGS.
First, when the operator instructs image processing such as copying or fold noise removal processing from the keyboard 8, image input (not shown) is performed by a software module downloaded from the CD-ROM drive 3 and stored in the RAM 2 in the image forming apparatus. A fold noise elimination device including a fold noise search range determination unit, a fold noise region extraction unit, a fold noise removal unit, and an image output unit is configured, and the noise removal processing starts.
By starting this processing, the CPU 1 inputs an image read by the scanner 6 or an image already read by the scanner 6 and recorded in the memory 5 to the image input unit. The input image is passed to the fold noise search range determination unit, and the fold noise search range is determined (S1). The image for which the fold noise search range has been determined is next passed to the fold noise region extraction unit, and a fold noise region is extracted from the already determined fold noise search range (S2). Further, the image from which the fold noise region is extracted removes the fold noise in the fold noise region by the fold noise removal unit (S3), and outputs the image via the image output unit.
[0017]
Next, the processing for determining the fold noise search range will be described in detail.
FIG. 3 is a diagram illustrating a fold noise search range in an input image of a portrait-type six-fold original. Since the document is of a portrait type, it is folded in two in the horizontal direction and three in the vertical direction. Therefore, when an image forming process such as copying or a process of removing fold noise is instructed from the keyboard 8 and the input image of the original is passed from the image input unit to the fold noise search range determination unit, the CPU 1 displays the input image 11 on the input image 11. ± a with respect to a line segment h1 at a bisecting position in the X-axis direction of the XY coordinates with the upper left corner of the input image 11 as the origin 0, and line segments h2 and h3 at a trisected position in the Y-axis direction. The band-like portion having the width of the pixel is determined as the fold noise search range 12.
[0018]
When determining the fold noise search range 12, instead of setting the width of the search range to a fixed width such as ± a pixel centering on the fold line segment, the width and height of the input image or the resolution is obtained. , Different widths are determined for each input image, that is, for each document size. By making the width of the search range variable in accordance with the document size in this manner, even when the document is placed obliquely on a document table (not shown) of the scanner 6, the fold image is within the fold noise search range 12. , And the erroneous extraction of the fold noise region described later is eliminated.
[0019]
Next, the fold noise region extraction processing will be described in detail.
The extraction of the fold noise is performed in such a manner that a black pixel (having a pixel value of 1) is connected to the fold noise search range 12 in the horizontal direction with respect to the fold in the X-axis direction of the XY coordinates within the fold noise search range 12. This is a process of extracting run lengths equal to or more than the length and sequentially extracting the run lengths for each line, extracting a region where run lengths are integrated adjacent to each other as a rectangular region, and setting the rectangular region as a fold noise region. . The same processing is performed in the vertical direction for the fold in the Y-axis direction.
FIG. 4 is a diagram illustrating a fold noise area within the fold noise search range. When the process of extracting the fold noise region starts, the CPU 1 extracts a run length having a pixel value of 1 and a length equal to or more than a predetermined length from the left end to the right end of the fold noise search range 12 within the fold noise search range 12. This extraction processing is further performed sequentially for each line within the fold noise search range 12, thereby extracting a rectangular area where the run lengths of the pixel values 1 are integrated adjacent to each other to be a fold noise area.
[0020]
FIG. 5 is a diagram illustrating extraction of a fold noise region. In FIG. ₀ , A run length n equal to or longer than a predetermined length of the (n + 1) th line ₁ , A run length n equal to or longer than a predetermined length of the (n + 2) th line ₂ , N + 3, a run length n equal to or more than a predetermined length of the third line ₃ Are adjacent run lengths, so run length n ₀ (The left end of the n line) and the run length n ₃ By connecting the coordinates of the last pixel (right end of the (n + 3) th line), a rectangular area with integrated run lengths is extracted as a fold noise area 13.
At this time, the threshold for making the run length equal to or longer than the predetermined length is (1) the ratio of the width and height of the input image corresponding to the document size, or (2) a somewhat large size is set as the fold noise area, or ( 3) In order to make the elongated area a fold noise area, it is appropriately determined according to the aspect ratio of the rectangular area.
[0021]
Next, in the processing for removing the fold noise, the fold noise is removed by inverting all the black pixels (pixel value 1) in the fold noise area extracted as described above into white pixels (pixel value 0). .
[0022]
When removing fold noise, image information that must not be removed, such as characters, may be present in the fold noise area.
FIG. 6 is a diagram showing an example in which unnecessary information, that is, necessary information, is included in the fold noise area. A circle W in FIG. 6A indicates a search for a fold noise in the vertical direction of the input image 11 of the six-fold original. FIG. 6B is an enlarged view of the fold noise region extracted from the range 12, and shows a state in which the necessary information ABC exists in the fold noise region 13.
When necessary image information ABC that must not be removed enters the fold noise area 13 and the black pixels in the fold noise area 13 are uniformly inverted to white pixels, the image information C also includes black pixels. It will be removed.
Therefore, in the present embodiment, black pixels are inverted to white pixels in run-length units in the fold noise region 13 so that such necessary information is not removed.
[0023]
When removing fold noise, ruled lines as necessary information may intersect in the fold noise area, and a part of the ruled line may be removed by the fold noise removal processing.
[0024]
Therefore, in the present embodiment, a part of the ruled line removed at the time of the fold removal is interpolated.
FIG. 7 is a diagram showing a state of removal and interpolation of a part of the ruled line intersecting the fold noise. In the figure, the ruled line 14 intersecting the fold noise n is shown in FIG. Since the ruled line portion 15 crossing the fold noise n is removed (FIG. 7B), this is interpolated (FIG. 7C). For this purpose, a process of interpolating the ruled line image is performed after the fold noise removal process.
FIG. 8 is an overall flowchart of the fold noise removal processing including the interpolation processing. The ruled line interpolation processing (S4) is performed after the noise removal processing (S3) in the fold noise removal processing flow shown in FIG.
In FIG. 8, steps (S1) to (S3) of the processing are the same as those in FIG. In the present embodiment, the interpolation processing of the ruled line (S4) is performed by configuring an interpolation processing unit (not shown) in the image forming apparatus for the interpolation processing.
[0025]
Hereinafter, the interpolation processing of the ruled line image will be described in detail.
FIG. 9 is a diagram illustrating interpolation of a ruled line image of a horizontal ruled line.
Referring to FIG. 9, first, (1) the fold noise n is extracted again from the pixel from which the fold noise has been removed. (2) Left and right rectangles that are in contact with the extracted fold noise n are extracted and classified into a rectangle L that contacts the left side of the fold noise n and a rectangle R that contacts the right side of the fold noise n (FIG. 9A). At this time, a rectangle not in contact with the fold noise n is not a ruled line and is not extracted. Next, (3) the positional relationship between the extracted rectangles L and R is checked, and the likelihood of a ruled line is determined from this positional relationship. The position Y of the middle point of the Y coordinates of the rectangles L and R extracted for this purpose ₀ Ask for. (4) A threshold value ± t is provided at the left and right rectangles at which the midpoint positions match or at the midpoint position, and the difference between the midpoints of the Y coordinates of the left and right rectangles is compared with the threshold value t. If it is within the value, the combination (pair) of the left and right rectangles L and R is determined as a ruled line (FIG. 9B). Then, (5) the pixels in the area X between the rectangles forming these combinations are inverted to black pixels. This processing is performed from the upper end to the lower end of the fold noise.
In the above, the interpolation processing of the horizontal ruled line traversing the fold noise formed in the vertical direction has been described. Interpolation can be performed in the same form by obtaining the position of the point.
In this embodiment, the case where the ruled line is interpolated has been described. However, the present invention is not limited to the ruled line, and may be, for example, a line segment forming a part of a diagram.
[0026]
Further, in the case of removing the fold noise, a line segment forming a part of the drawing as necessary information may enter the fold noise area. When the fold noise is removed by the above-described method, such a line segment is removed.
[0027]
Therefore, in the present embodiment, the fold noise is removed while leaving a necessary line segment. At this time, the range in which the fold noise is removed is changed by selecting the sensitivity of strong or weak noise removal.
Here, the weak sensitivity is a sensitivity at which the fold noise is weakly eliminated (leave the noise as much as possible). At the weak sensitivity, the search range of the fold noise is narrow and / or the size of the fold noise area is large, or the fold noise area is vertically and horizontally. Each parameter is set so that the ratio becomes large. That is, by narrowing the fold noise search range and / or increasing the run length to which the black pixels are connected, the line segment image that enters the fold noise region is left.
The high sensitivity is a sensitivity that does not leave fold noise (eliminates noise as much as possible). Therefore, in the case of the strong sensitivity, the parameter is set opposite to the weak sensitivity. The selection of the strength is performed by the user using a strength mode key (not shown) on the keyboard 8. The parameter is selected based on the instruction of the key operation, and the CPU 1 performs a noise removal process according to the setting of the parameter.
As another example of the strong and weak sensitivity, interpolation is performed such that the ruled line removed by increasing the image interpolation value (difference between the positions of the right and left rectangles) when the ruled line is interpolated with low sensitivity is not considered as a ruled line as much as possible. There is an example in which the image interpolation value is reduced with high sensitivity so that the image is interpolated as a ruled line as much as possible. These may be configured so that a strong mode key can be selected.
[0028]
As described above, according to the present embodiment, since the fold image (fold noise) n formed from the document having the fold mark is removed, the document is folded for storage or the like, and the document having the fold is printed out. As a result, a high-quality print result from which the fold image has been removed can be obtained. At this time, the fold noise search range 12 is determined at the position of the fold corresponding to the folding method based on the folding method, so that it is not necessary to specify the search range for each fold.
[0029]
Further, since the width of the fold noise search range 12 is automatically changed and determined according to the size of the original, even if the original paper is placed diagonally on the original platen of the image forming apparatus, the fold to be removed is removed. Is less likely to protrude from the search range. Therefore, the fold noise is removed without any omission in extraction.
[0030]
Further, since the pixel values of all the pixels of the predetermined pixel value in the fold noise region 13 are inverted, the inversion processing is facilitated. Further, since the inversion process of the pixel value is performed for each run length of the black pixel constituting the fold noise, the character C which is the necessary information existing in the fold noise region is not removed.
[0031]
Furthermore, since the ruled line portion removed by the fold noise removal is interpolated for the ruled line 14 traversing the fold noise n, a part of the ruled line does not remain removed by the fold noise removal process.
[0032]
Furthermore, the user can select the noise removal sensitivity to strongly or weakly remove the fold noise according to the user's intention. If selected, the fold noise can be removed while leaving such a line segment. It is clear that the line segment does not enter the fold noise area, and if it is desired to strongly remove the fold noise, it is possible to obtain a high-quality image output from which the fold noise has been removed by selecting strong sensitivity. it can. In addition, the switching of the parameters with high and low sensitivity is performed not by individual parameter selection but by collective selection using the mode key, so that the user can use the device easily.
[0033]
Next, another embodiment of the present invention will be described.
Other embodiments of the present invention display fold noise in a different color than other noise.
Normally, an area extracted as fold noise (fold noise) is displayed in blue, for example, and an area not extracted as fold noise (other noise) is displayed on the display 7 in black with black pixels. Other noises can be extracted by comparing the original image with the image after noise removal, and extracting an image present in the original image and not present in the image after noise removal as noise. Therefore, the extracted noise may be displayed as black pixels (black).
In the present embodiment, the display color is changed so that the fold noise can be recognized and recognized separately from other noises.However, instead of a method of changing the display color, a display method of a different format such as displaying the fold noise by blinking display is used. Can be adopted.
[0034]
According to the present embodiment, by displaying the fold noise in blue different from other noises, it is possible to perform a checking operation focusing only on the fold noise. The fold noise is similar to the ruled line, so it may be difficult to distinguish between the ruled line and the noise immediately.However, the difference in the display color makes the line segment of the drawing original distinct from the fold noise, and focuses on the fold noise. Work becomes easier.
[0035]
As described above, for each landscape-type original image or portrait-type original image, a fold noise search range having a predetermined width in the fold direction of the original image is set at the position of the fold formed based on the predetermined folding method of the original paper, Within this fold noise search range, a rectangle in which run-lengths equal to or longer than a predetermined length to which pixels of a predetermined pixel value are connected in the horizontal direction in the horizontal search range and in the vertical direction in the vertical search range, respectively. The procedure of processing for extracting an area and removing pixels in the extracted rectangular area has been described. To execute this processing, a procedure for instructing a computer is described in a well-known programming language to create a program. The program is recorded on a well-known computer-readable recording medium such as a CD-ROM, a DVD-ROM, and a flexible disk. Therefore, by downloading the program directly to the computer of the image forming apparatus from the recording medium or via a communication line such as a network or a LAN, a document image is created based on the fold marks of the document sheet described above. Vertical and horizontal fold noise can be removed.
[0036]
【The invention's effect】
According to the first and eighth aspects of the present invention, a fold noise search range is determined at a fold position corresponding to the folding method based on the manner of folding the original paper, and a fold image (fold noise) existing within the search range is determined. Therefore, when image processing is performed on a document that is folded for storage or the like and has creases in the vertical and horizontal directions of the document paper, it is possible to easily obtain a high-quality image output from which crease images have been removed. it can.
According to the second aspect of the present invention: the fold noise search range having a width corresponding to the size of the original paper is automatically determined, so that when the original paper is placed obliquely on the original platen of the image forming apparatus. Also, the possibility that the line of the fold to be removed falls within the search range increases, and the elimination of the fold noise is eliminated.
Effect corresponding to the third aspect of the invention: Since the pixel values of all the pixels of the predetermined pixel value in the rectangular area are inverted, the inversion processing is facilitated. Further, since only the run-length pixel value of the predetermined pixel value constituting the fold noise is inverted, the necessary information existing in the fold noise area is not removed.
Effects corresponding to the fourth aspect of the present invention: Since the removed portion of the line segment traversing the fold noise is interpolated, a part of the line segment does not become defective due to the fold noise removal process.
According to the fifth aspect of the present invention, when a necessary line segment image exists within the fold noise search range, the fold noise is removed by changing the range in which the fold noise is removed. can do.
Advantageous Effects Corresponding to the Invention of Claim 6: An image forming apparatus having effects corresponding to the inventions of claims 1 to 5 can be provided.
Advantageous Effect Corresponding to the Invention of Claim 7: In the image forming apparatus according to the invention of claim 6, by displaying the fold noise in a display format different from that of other noises, it is possible to perform a check operation focusing only on the fold noise. .
Effects corresponding to the ninth and tenth aspects: In the computer of the image forming apparatus, the image fold noise removing method of the present invention can be easily implemented.
[Brief description of the drawings]
FIG. 1 is a block diagram of a main part of an image forming apparatus that executes a fold noise removing method according to an embodiment of the present invention.
FIG. 2 is an overall flowchart of a fold noise removing process according to the embodiment of the present invention.
FIG. 3 is a diagram illustrating a positional relationship between an input image of a portrait-type original six-fold and a fold noise search range.
FIG. 4 is a diagram illustrating a fold noise area within a fold noise search range.
FIG. 5 is a diagram illustrating extraction of a fold noise region.
FIG. 6 is a diagram illustrating an example in which necessary information is included in a fold noise region.
FIG. 7 is a diagram illustrating a relationship between removal of ruled lines crossing fold noise and interpolation.
FIG. 8 is an overall flowchart of a fold noise removal process including an interpolation process.
FIG. 9 is a diagram illustrating interpolation of a ruled line image.
FIG. 10 is a diagram illustrating an example of six-fold and eight-fold of a portrait type document.
[Explanation of symbols]
1 CPU (central processing unit), 2 ROM or RAM, 3 CD-ROM drive, 4 work area RAM, 5 memory, 6 ··· Scanner, 7 ··· Display, 8 ··· Keyboard, 11 ··· Input image, 12 ··· Crease noise search range, 13 ··· Crease noise area, 14 ··· -Ruled line, 15 ... Ruled line intersection.

Claims (10)

An apparatus for removing a fold image formed as a binary document image based on a fold mark of a document sheet as fold noise,
Means for setting a fold noise search range of a predetermined width in the fold direction of the document image, at a fold position formed based on a predetermined folding method of the document sheet, for each landscape type document image or portrait type document image,
Within the fold noise search range, a rectangle obtained by integrating run lengths equal to or longer than a predetermined length to which pixels having predetermined pixel values are connected in the horizontal direction in the horizontal search range and in the vertical direction in the vertical search range. Means for extracting a region;
An apparatus for removing fold noise of an image, comprising: means for removing a predetermined pixel in the rectangular area. 2. The image fold noise elimination device according to claim 1, wherein said means for setting the fold noise search range sets a noise search range having a different width for each size of the document paper. . 3. The image fold noise removing device according to claim 1, wherein the means for removing a predetermined pixel in the rectangular area inverts pixel values of all pixels of a predetermined pixel value in the rectangular area, or a predetermined pixel. An image fold noise eliminator characterized by inverting only a run-length pixel value having a length equal to or longer than a predetermined length to which a value pixel is connected. 4. The image fold noise removing apparatus according to claim 1, wherein when the line segment image traversing the fold noise search range is removed by removing the fold noise, the removed line segment image is interpolated. An image fold noise elimination device comprising: 5. The image fold noise removing device according to claim 1, wherein a search range of the fold noise and / or a length of a run length equal to or longer than a predetermined length is variable. Noise removal device. An image forming apparatus comprising the image fold removing device according to claim 1. 7. The image forming apparatus according to claim 6, further comprising: means for displaying the fold noise formed on the document image in a display format different from that of other noises. A method for removing a fold image formed in a binary document image based on a fold mark of a document sheet as fold noise,
A step of setting a fold noise search range of a predetermined width in the fold direction of the original image at a position of a fold formed based on a predetermined folding method of the original paper for each landscape type original image or portrait type original image,
Within the fold noise search range, a rectangle obtained by integrating run lengths equal to or longer than a predetermined length to which pixels having predetermined pixel values are connected in the horizontal direction in the horizontal search range and in the vertical direction in the vertical search range. Extracting a region;
A method for removing fold noise of an image, comprising a step of removing a predetermined pixel in the rectangular area. A program for causing a computer of an image forming apparatus to execute a procedure of each step according to claim 8. A computer-readable recording medium on which the program according to claim 9 is recorded.

Images