JPH09307768A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a solid image with apparently uniform density by applying density correction in response to the distance from an edge line up to a present line position to an image signal in an area in the vicinity of the edge line of the solid image. SOLUTION: After image information is once stored in an image information storage device 11, the information is given to an edge detection means 12 as required from the image information storage device 11. The edge detection means detects an edge line where areas of an image with different density are in contact. Then the image signal is given to a density correction means 13. The density correction means 13 corrects the density of an area in the vicinity of the edge line of one area in two areas whose density differs from each other and in contact with the edge line detected by the edge detection means 12 so as to be closer to the density of the other area when the line is closer to the edge line. The corrected image signal is fed to an image recorder 15, from which the signal is recorded on a recording medium.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus for correcting the density of an image signal carrying image information to form an image having a desired density distribution.

[0002]

2. Description of the Related Art In an image forming apparatus such as a laser printer or an ink jet printer, a density gradation of image information is expressed by a multi-valued image signal, and toner or ink within a unit area is expressed according to the multi-valued image signal. A method of forming a density gradation image by controlling the adhesion amount is widely adopted.

By the way, when an image formed by these image forming apparatuses is observed by human eyes, it is an image having a uniform density distribution on the image signal, but the human eyes have a non-uniform density distribution. May look like. This phenomenon is generally called the Mach phenomenon or Mach effect, and has been introduced in detail in the literature, for example, "Psychophysics of vision" (Mitsuo Ikeda, published by Morikita Publishing).

FIG. 15 is an explanatory diagram of the Mach phenomenon. FIG. 15A shows a solid image having five different densities (a set of pixels having the same density distribution). This image is output from the image forming apparatus with image information having a completely uniform density distribution on the image signal. When the density distribution of this image is mechanically measured and made into a graph, five horizontal lines are displayed. A stair-step graph consisting of a straight line and four vertical straight lines connecting it should be drawn. However, when the apparent density of this image when viewed by a human is plotted, as shown in FIG. 15B, human eyes have a low density near the edge line at the boundary of each solid image. The side looks like a lower density and the high side looks like a higher density.
As described above, even if the density distribution is physically uniform, the density difference near the edge line is emphasized by the human eye, and it looks like a non-uniform density distribution.

[0005]

This Mach phenomenon is a phenomenon based on a visual correction function when a person detects an image edge, but information is obtained by solid images having different densities, such as a business graph. In the case of image information to be transmitted, it may be inconvenient for a solid image, which originally should have uniform density, to have apparently uneven density.

In view of the above circumstances, the present invention can correct an apparent nonuniformity of the density distribution near the edge line of a solid image and form a solid image having an apparently uniform density. The purpose is to provide.

[0007]

According to an image processing apparatus of the present invention that achieves the above object, an edge line in which regions having different densities on an image are in contact with each other based on an image signal carrying image information. And at least one first area of the two areas having different densities, which border the edge line detected by the edge detection means and the edge line detected by the edge detection means. A density correction unit that corrects the image signal in the edge line vicinity area so that the density of the area near the edge line approaches the density of the other second area of the two areas as it approaches the edge line. It is characterized by having.

Here, the density correction means may correct the image signal so that the density continuously changes over the two regions, and the density correction means may perform the density correction. The image signal may be corrected so that the discontinuity of the density on the edge line remains.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below. FIG. 1 is a schematic diagram of an embodiment of an image processing apparatus of the present invention. In the following description, the density of each pixel of the image signal carrying the image information will be simply referred to as the image signal. In the present embodiment, the density gradation information is expressed as an 8-bit image signal. When the image signal is 0, the pixel coverage is 0% (white), and when the image signal is 255,
The pixel coverage is 100% (black).

As shown in FIG. 1, after the image information is once stored in the image information storage device 11, it is input from the image information storage device 11 to the edge detection means 12 as needed. The edge detection unit 12 detects an edge line where areas of different densities on the image are in contact with each other. Next, the image signal is input to the density correction unit 13. Density correction means 1
In 3, first, it is determined whether the area surrounded by the edge line is a solid image, and the image signal is corrected for the area determined to be the solid image. That is, with respect to one of the two regions having different densities that are bordered by the edge line detected by the edge detection means 12, the density of the region in the vicinity of the edge line in the region becomes the edge line. The image signal in the area near the edge line is corrected such that the closer it is, the closer the density is to the other area. When correcting the image signal, the correction is performed using the conversion table stored in the conversion table storage device 14 and representing the relationship between the image signal and the density value of the output image.

The image signal corrected by the density correcting means 13 is sent to the image recording device 15 and recorded on the recording medium. In this embodiment, 400 dpi by electrophotography is used.
Although an image recording device having a resolution of (dpi: the number of dots per inch) was used, an image recording device such as a thermal transfer system, a silver salt photographic system, or an inkjet system other than the electrophotographic system may be used. Further, even if an image recording apparatus having a resolution different from that of this embodiment is used, the same effect as that of this embodiment can be obtained.

As a method of detecting an edge line in the edge detecting means 12, for example, the document "Image Analysis Handbook" (supervised by Mikio Takagi and Yohisa Shimoda; The University of Tokyo Press)
P. A known method such as the Labrian operator method disclosed in 550 is used. After the edge line is detected, in order to facilitate the subsequent processing, p. Pixels forming an edge line (hereinafter referred to as edge pixels) are ordered by the method described in 579.

FIG. 2 is an explanatory diagram of the ordering of the edge pixels. In FIG. 2, each pixel of the image signal read from the image information storage device 11 (see FIG. 1) is arranged two-dimensionally, that is, on a plurality of points 20 arranged in the arrow x-axis direction, A plurality of columns of these points 20 are arranged and displayed in the arrow y-axis direction. The points 20 of each pixel arranged in this manner are sequentially scanned to find an edge start point. The edge start point means a point to which the traced mark is not yet attached, and is stored in the image information storage device 11 together with the density information of each pixel. For example, point 21
If it is found that a (hatched portion) is a point to which a traced mark has not been added yet, the point 21a is stored as an edge start point (tracked to the corresponding part of the image information stored in the image information storage device 11). Marked already). Next, paying attention to eight pixels adjacent to the edge start point 21a in four directions in the vertical and horizontal directions and in four diagonal directions, the next edge pixel following the edge pixel 21a is traced. By advancing such a tracking operation one after another, each edge pixel 21 forming the annular edge line 22 starting from the edge start point 21a and returning to the edge start point 21a from the edge start point 21a as shown by the arrow.
A, 21b, 21c, 21d, ..., 21o can be ordered.

Simultaneously with the above ordering process, the total extension distance of the edge line 22 is calculated. This calculation is performed by setting the interval between the adjacent edge pixels forming the edge line 22 in the x-axis direction or the y-axis direction as one unit distance and integrating the unit distances. For example, the edge pixel 21
The distance between a and the edge pixel 21b is 2 unit distance, and the distance between the edge pixel 21b and the edge pixel 21c is 1 unit distance. Even if the edge line runs diagonally, it is not measured diagonally, Measure only in the x-axis direction or the y-axis direction and integrate it. In addition, here, the total number of pixels forming the edge line 22 is stored, and hereinafter, this will be referred to as N.

When the processing in the edge detecting means 12 is completed, the density correcting means 13 corrects the image signal. FIG. 3 is a flow chart showing a processing procedure in the density correction means of this embodiment. First, the edge detection means 12
Regarding the edge line detected by (see FIG. 1),
It is determined whether the area surrounded by the edge line is a solid image (step S301). Whether or not it is a solid image is determined as follows. That is, when the pixel density in the area surrounded by the edge line is constant, the area of the area is 100 mm ² or more, and the shape factor S (described later) of the area is 0.5 or more, The area is determined to be a solid image. The reason why the area of the region is used as the determination factor is that the effect of the correction is small even if the correction processing is applied to the region having a small area, and the shape factor S
Is used as a determination factor because the appearance of the Mach phenomenon becomes gentle when the shape is too complicated, and the significance of the correction is small.

The shape factor S is obtained as follows. That is, when the total extension distance of the edge line obtained by the edge detecting means 12 is L and the number of pixels in the area (including the edge pixels) surrounded by the edge line is A, S = 4πA / L ² The value defined by is the shape factor S. Originally, the shape factor S should have a maximum value of 1 when the area surrounded by the edge lines is a perfect circle, and should have a smaller value as the area surrounded by the edge lines has a complicated shape. The total extension distance L of the edge line in the present embodiment is obtained by measuring the distance between the adjacent edge pixels in the x-axis direction or the y-axis direction as described above, that is, the pixels adjacent in the oblique direction are directly measured obliquely. Since it is measured as the sum of the distances in the x-axis direction and the y-axis direction, the true length of the edge line does not match, and the shape factor S may be a value larger than 1. However, since it is certain that the value of the shape coefficient S will decrease as the area surrounded by the edge line has a complicated shape, it is used as one element for determining whether or not the image is a solid image in the present embodiment. .

Returning to FIG. 3, the description of the flow chart of the density correction means will be continued. If it is determined in step S301 that the area is not a solid image, the image signal is sent to the image recording apparatus without performing the density correction processing of the image signal (processing after step S302). When it is determined in step S301 that the area is a solid image, the process proceeds to step S302, and the density correction processing of pixels outside the edge line from step S302 to step S306, and step S307 to step S310.
The density correction processing for the pixels inside the edge lines up to is performed. In the following description, for convenience, the side with the smaller image signal value (that is, the side with the lower pixel density) is referred to as the outer side of the edge line with the edge line as a boundary. The larger side (that is, the side with higher pixel density) is called the inside of the edge line.

In the density correction processing of pixels outside the edge line, the counter a is set to the initial value 1 in step S302.
After that, the extraction of the a-th contour pixel (step S303) and the density correction of the a-th contour pixel (step S304) are executed, and the counter a is executed every time the steps S303 and S304 are executed once. Is incremented by 1 (step S306), and the counter a is added in step S305.
Until the number exceeds 48, the extraction of the contour pixel in step S303 and the density correction of the contour pixel in step S304 are repeatedly executed. Detailed procedures of the contour pixel extraction and the contour pixel density correction will be described later.

Next, the density correction processing of the pixels inside the edge line is started. First, after the counter a is set to the initial value 0 in step S307, extraction of the a-th contour pixel (step S308) and density correction of the a-th contour pixel (step S309) are executed, and then step S30.
8 and step S309 are executed once, 1 is subtracted from the counter a (step S310), and step S3
Step S30 until the counter a reaches −48 at 10
The extraction of the contour pixel of 8 and the density correction of the contour pixel of step S309 are repeatedly executed. Thus, the processing of pixels outside the edge line and the processing of pixels inside the edge line are each performed 48 times.

Therefore, in this embodiment, the area having a total width of 96 pixels across the edge line is corrected. 96
The width of a pixel corresponds to about 6 mm in this embodiment. Hereinafter, the width of 96 pixels will be referred to as a correction area. When the correction processing in the correction areas outside and inside the edge line of the solid image is completed, the corrected output image signal is sent to the image recording device 15 in the subsequent stage.

Next, detailed procedures of the contour pixel extraction in step S303 and the contour pixel density correction in step S304 will be described with reference to FIGS. FIG.
7 is a flowchart showing a detailed procedure of a contour pixel extraction step S303 and a contour pixel density correction step S304 in FIG. First, as a preparation, the density dO of the image signal outside the edge line of the solid image and the density dI of the image signal inside the edge line are stored.

Next, a pixel (a-th contour pixel) forming a contour line outside by a pixels from the edge line of the solid image is extracted, and a density correction process (to be described later) of the a-th contour pixel is executed. FIG. 5 is an explanatory diagram of contour pixels in the present embodiment. FIG. 5A shows step S3 of FIG.
An example of a solid image determined to be a solid image in 01 is shown, and FIG. 5B shows a partially enlarged view of FIG. 5A.

In FIG. 5B, edge pixels (0th pixel) forming the edge line 22 (white polygonal line) of the solid image are shown.
Contour pixels) are shown. The contour pixel extraction of FIG. 3 is performed so as to wrap the outside of these edge pixels (step S30).
The first contour pixel extracted by 3) is shown, the second contour pixel is shown so as to wrap the outside thereof, and the third contour pixel is shown so as to further wrap the outside thereof. The fourth contour pixel and the subsequent pixels are not shown. Inside the edge pixel (0th contour pixel), the -1st contour pixel is shown side by side along the contour of the edge pixel. The second
The contour pixels and the subsequent pixels are not shown.

Returning to FIG. 4, the detailed procedure for extracting each of these contour pixels will be continued. First, a counter n indicating the order of edge pixels forming an edge line of the solid image is set to an initial value 1, and a counter i for ordering contour pixels adjacent to the outside or inside of the edge pixel is set to an initial value 1. Set (step S40
1).

Next, pay attention to the (a-1) th contour pixel whose order is the n-th ordered by the above-described edge pixel ordering process (see FIG. 2) (step S40).
2) Next, the parameter m indicating the vertical and horizontal positions of the four adjacent pixels above and below and to the left and right of the edge pixel is set to 1 (step S403). Here, the parameters m = 1, 2, 3, are located at the lower, left, upper and right positions of the edge pixel.
4 respectively.

Next, the x-coordinate and y-coordinate of the m-th position are obtained (step S404), and then the x-coordinate and y-coordinate are obtained.
It is determined whether the pixel density of the coordinates is equal to the previously stored density dO (step S405). If the result of determination in step S405 is equal to the density d0, step S405
Proceed to 406, and if the density is not equal to dO, step S
Proceed to 410.

In step S406, the x coordinate and y
A pixel at the coordinate is newly registered as an outer contour pixel for one pixel, and then the process proceeds to step S407 to perform a pixel density correction process (described later). Next, 1 is added to the counter i (step S408), and next, whether or not the parameter m is 4, that is, the bottom, left, and top of the edge pixel,
It is determined whether or not the processing for the four adjacent positions on the right has been completed (step S409). If the result of determination in step S409 is that processing for all adjacent positions has been completed, processing advances to step S412, and if processing for all adjacent positions has not completed, processing advances to step S410. In step S410, 1 is added to the parameter m, that is, the processing for the next adjacent position is prepared, and then the process proceeds to step S411. In step S411, it is determined whether or not the parameter m exceeds 4 as a result of the addition. If the result of determination in step S411 is that the parameter m exceeds 4, then step S411
412, if the parameter m is 4 or less, the process returns to step S404 to repeat the processing from step S405.

In step S412, 1 is added to the counter n indicating the order of edge pixels, and then step S41.
At 3, it is determined whether or not the counter n has exceeded the total number N of pixels of the edge line obtained by the edge detecting means 12. If the result of determination in step S413 is that the counter n does not exceed the total number of pixels N, step S402.
Then, the process is repeated for the edge pixel whose order is the nth. As a result of the determination in step S413,
If the counter n exceeds the total number N of pixels, the processing has been completed for all edge pixels of the solid image, and the process returns to step S305 in FIG. Step S30
In 5 as described above, as long as the counter a does not exceed 48, 1 is added to the counter a in step S306, and the contour pixel extraction for the added a is performed (step S
303) and contour pixel density correction (step S304),
That is, steps S401 to S41 in FIG.
The processes up to 3 are repeatedly executed. In this way,
As a result of the determination in step S305, the counter a is 4
When it exceeds 8, that is, when the processing of the pixels outside the edge line is completed, the process proceeds to the processing of the pixels inside the edge line (step S307 in FIG. 3).

Steps S307 to S3 in FIG.
The processing from step S302 to step S3
It is almost the same as the processing up to 05. However, step S
The initial value of the counter a in 307 is 0, the counter a is decremented by 1 in step S311, and the end determination condition is that the counter a is −48 or less in step S311. Further, in this case, the density dI of the pixel inside the solid image is used instead of the density dO used for the density comparison in step S405 of FIG.

Next, the correction processing of the pixel density in step S407 of FIG. 4 will be described. This correction process is performed using a correction function. FIG. 6 is a diagram showing the density distribution of the solid image area inside and outside the edge line before the correction processing. The horizontal axis of FIG. 6 indicates the distance from the edge line, and the unit thereof is the counter a used to count the number of contour lines in FIGS. 3 and 4. FIG.
The vertical axis of is the density of the pixel at the distance a from the edge line. As shown in FIG. 6, a solid image area having an inner density DI and an outer density D with an edge line as a boundary.
A steep step is formed by the solid image area of O 2.

FIG. 7 is a diagram showing the density distribution of the solid image area inside and outside the edge line after the correction processing. As shown in FIG. 7, the density DI of the inner solid image area begins to decrease to the right within the correction area spanning the edge line, and eventually the density D of the outer solid image area.
Reach O. This density distribution is expressed as a function of the distance from the edge line (hereinafter referred to as a correction function), the shape of the correction function is appropriately selected, and the pixel density on both sides of the edge line is corrected to obtain the apparent appearance due to the Mach phenomenon. The nonuniformity of the above density is canceled and a solid image having an apparently uniform density can be obtained. However, since the shape of the correction function suitable for it varies depending on the image information, it is desirable to prepare some correction functions in advance and select a correction function suitable for each image information from among them to perform correction.

An example of the correction function other than the linear function shown in FIG. 7 will be described. FIG. 8 is an example of a correction function using a quadratic polynomial. As shown in FIG. 8, the density DI of the inner solid image area begins to drop in the correction area by drawing a gentle curve, and eventually reaches the density DO of the inner solid image area with the edge line as an inflection point. In this way, the density distribution in the correction area as the correction function is the nth
You may use the function represented by the polynomial of 1).

FIG. 9 is a diagram showing an example of a correction function in which the density distribution is discontinuous in the edge line portion. As shown in FIG. 9, this correction function is such that the discontinuity of the density on the edge line remains in order to leave the sharpness of the level difference of the density on both sides of the edge line. It may be effective to use a correction function.

FIG. 10 is a diagram showing a correction function in which the density distribution is discontinuous in the edge line portion and the correction area is asymmetric with respect to the edge line. As shown in FIG. 10, a correction function in which the correction area is asymmetric with respect to the edge line is used to change the correction method on the dark side and the dark side, so that it may appear more natural depending on the image information. It may be possible to correct to a solid image.

If the relationship between DI and D0 is the opposite of that shown in FIGS. 7 to 10, a left-right inverted function of the edge line may be used. By the way, generally, in an image processing apparatus, there is a non-linear relationship between the density of an image signal inside the apparatus and the density of an image actually output, and the relationship between the two is different for each apparatus. It is usually different.

FIG. 11 is a graph showing the relationship between the density of the image signal inside the image processing apparatus and the density of the image actually output. Usually, before the image is output from the image processing apparatus, it is stored in the image processing apparatus in order to correct the non-linear relationship as shown in FIG. 11 and to correct the machine difference between the image processing apparatuses. Correction is performed using the conversion table. That is, as shown in FIG. 11, when it is desired to output at the target density DA with respect to the maximum density Dmax of the output image, the density is corrected to the target value dA of the image signal.

In the image processing apparatus of this embodiment, this conversion table is stored in the conversion table storage device 14. Therefore, in the present embodiment, the pixel density in the correction region across the edge line is corrected using the correction function having the functional form as shown in FIGS. 7 to 10, and then stored in the conversion table storage device 14. The density is corrected using the above conversion table to obtain an output image signal.

In the present embodiment, when four solid images having pixel coverages of 80%, 60%, 40%, and 20% as shown in FIG. 12 are adjacent to each other and are corrected using various correction functions. The output image of was evaluated. FIG.
FIG. 4 is an explanatory diagram of an image in which four solid images having a pixel coverage of 80%, 60%, 40%, and 20% are adjacent to each other.

With respect to the image shown in FIG. 12, an output image after being corrected by the following nine kinds of correction functions and an output image without being corrected were compared and evaluated by a sensory evaluation method. As the correction function, the following A type, B type,
A total of 9 types of correction functions were used by further subdividing the 3 types of C type into 3 types.

The correction function belonging to type A is a function represented by a continuous function whose density distribution is n-th (n ≧ 1) in the correction area centered on the edge line as shown in FIG. As described above, the widths of the correction regions (widths in the horizontal axis direction) are A-1: 2 mm, A-2: 5 mm, and A-3: 8 mm, respectively.

The correction function belonging to type B is a function whose density distribution is n-th order (n ≧ 1) and whose density on the edge line is discontinuous, as shown in FIG. As described above, the widths (widths in the horizontal axis direction) of the correction areas are B-1: 1 mm, B-2: 2.5 mm, and B-3: 4 mm, respectively.

As shown in FIG. 10, the correction function belonging to the C-shape has a density distribution of the n-th order (n ≧ 1) and is partially discontinuous in the edge line, and at the same time, in the edge line part, A function whose function form is asymmetric, and in which the width of the correction area (width in the horizontal axis direction) is C-1: 2 mm, C-2: 5 mm, C-3: 8 mm, respectively, as follows: Refers to. However, in the case of a function belonging to the C-shape, the ratio of the widths of the correction areas on the high density side and the low density side across the edge line is set to 1: 3.

As a sensory evaluation method of the output image,
20 evaluators visually compared the Mach phenomenon in the vicinity of the edge line between the image before correction and the image after correction, and the comparison result was expressed by the following evaluation index. That is,
The image after correction was evaluated according to which one of ⊚: extremely preferable, ∘: preferable, ∘: neither can be said, Δ: not preferable, ×: unsightly. Table 1 shows the evaluation results.
Shown in

[0044]

[Table 1]

The density gradient in the correction area is as follows.
In any of the functional forms, the density gradient per 1 mm was 0.1 when the density difference inside and outside the solid image was 1.0. As a result of the evaluation, as shown in Table 1, it can be seen that the width of the correction region is preferably 2 mm or less in the case of the A-shaped function. Further, when the correction area was made larger than 8 mm, the evaluation was either Δ or X in any of the functions. Even when an n-th order (n ≧ 2) polynomial was used as the correction function, the same evaluation result as that in the case of the first order was obtained.

Next, in the functional forms of B type (see FIG. 9) and C type (see FIG. 10) in which the density distribution is partially discontinuous at the edge line, the length of the partially discontinuous portion occurring at the edge line. Table 2 shows the evaluation results when the height ΔD was changed. FIG. 13 is an explanatory diagram of the length ΔD of the discontinuous portion in the correction function in which the density on the edge line is discontinuous.

As shown in FIG. 13, the density inside the uncorrected solid image is DI, the density outside the uncorrected solid image is DO, and the length of the discontinuous portion formed on the corrected edge line (density) When the difference is ΔD (see FIG. 13), Δ
The values of D / | DI-DO | are 0.3, 0.5, and 0.7.
Table 2 shows the results of evaluation of the correction effect on the Mach phenomenon of the output image by changing to. The evaluation index is the same as in Table 1.

[0048]

[Table 2]

It can be seen from Table 2 that the correction may have an adverse effect when the value of ΔD / │DI -DO │ becomes 0.5 or less. Therefore, in the case of B-type and C-type functions, it is desirable that the value of .DELTA.D / .vertline.DI-DO.vertline. Is 0.5 or more. From the above results, when the A type is used as the correction function, the correction area is 2 mm or less, and when the B type and C type are used as the correction function, ΔD / | DI −DO | ≧ 0.5. It can be seen that a correction function satisfying is desirable.

In each of the above-described embodiments, the density correcting means corrects both of the two areas on both sides of the edge line, but it is necessary to correct both of the two areas on both sides of the edge line. However, depending on the image information to be handled, only one of the regions may be corrected and the other region may be left uncorrected.

By the way, when the image information subjected to the density correction processing described above is image-formed by using the electrophotographic image forming apparatus, when the electrostatic latent image on the photoconductor is developed,
The edge effect due to the fringe electric field in the direction parallel to the surface of the photoconductor that occurs near the edge of the solid image is reduced. This edge effect is a defect peculiar to an electrophotographic image forming apparatus, and various prevention methods have been proposed in the past, but there is still no complete solution.

FIG. 14 is an explanatory diagram of the edge effect in the electrophotographic image forming apparatus. As shown in FIG. 14A, the photoconductor 4 is used when the electrostatic latent image on the photoconductor is developed.
A line of electric force 44 as indicated by an arrow is formed in the gap 43 between the developer carrier 1 and the developer carrier 42. FIG. 14B shows the surface potential of the photoconductor at this time in association with FIG. 14A. As shown in FIG. 14B, when there is a large potential step in the surface potential of the photoconductor, an electric field 44a near the step portion 45a of the surface potential curve 45 is attracted to the step portion 45a and inclined, A so-called fringe electric field is formed. Therefore, in the vicinity of the edge of the electrostatic latent image corresponding to such a level difference of the potential, a line of the edge is broken, so-called edge effect occurs, and development that is faithful to the potential of the electrostatic latent image cannot be obtained.

However, according to the image processing apparatus of the present invention,
Due to the above-described correction process, a sharp change in density near the edge line becomes gentle.
As shown in (c), the potential change in the step portion 45a of the surface potential curve of the photosensitive member at the time of development also becomes gentle. Therefore, as shown in FIG. 14 (d), in the vicinity of the step portion 45a of the surface potential curve 45. The formation of the fringe electric field at is reduced, and the edge effect is reduced. As described above, the density correction processing in the vicinity of the edge line according to the present invention can reduce the Mach phenomenon, and can reduce the edge effect due to the fringe electric field at the time of electrostatic latent image development.

[0054]

As described above, according to the image processing apparatus of the present invention, in the area near the edge line of the solid image, the density correction according to the distance from the edge line is added to the image signal to detect the human image. The Mach phenomenon based on the visual correction function is canceled, and a solid image with apparently uniform density can be obtained.

[Brief description of drawings]

FIG. 1 is a schematic diagram of an embodiment of an image processing apparatus of the present invention.

FIG. 2 is an explanatory diagram of ordering of edge pixels.

FIG. 3 is a flowchart showing a processing procedure in the density correction means of the present embodiment.

FIG. 4 is a step S303 of extracting contour pixels in FIG.
9 is a flowchart showing a detailed procedure of step S304 of contour pixel density correction.

FIG. 5 is an explanatory diagram of contour pixels according to the present embodiment.

FIG. 6 is a diagram showing a density distribution of a solid image area inside and outside an edge line before a correction process.

FIG. 7 is a diagram showing the density distribution of the solid image area inside and outside the edge line after the correction processing.

FIG. 8 is an example of a correction function using a quadratic polynomial.

FIG. 9 is a diagram showing an example of a correction function in which the density distribution is discontinuous in the edge line portion.

FIG. 10 is a diagram showing a correction function in which the density distribution is discontinuous in the edge line portion and the correction region is asymmetric with respect to the edge line.

FIG. 11 is a graph showing the relationship between the density of the image signal inside the image processing apparatus and the density of the image actually output.

FIG. 12 shows pixel coverages of 80%, 60%, 40.
It is explanatory drawing of the image which four solid images of% and 20% adjoin.

FIG. 13 is an explanatory diagram of the length ΔD of the discontinuous portion in the correction function in which the density on the edge line is discontinuous.

FIG. 14 is an explanatory diagram of an edge effect in an electrophotographic image forming apparatus.

FIG. 15 is an explanatory diagram of a Mach phenomenon.

[Explanation of symbols]

 11 Image Information Storage Device 12 Edge Detection Unit 13 Density Correction Unit 14 Conversion Table Storage Device 15 Image Recording Device

Claims (3)

[Claims] 1. An edge detecting means for detecting an edge line where areas of different densities on the image are in contact with each other, based on an image signal carrying image information, and an edge line detected by the edge detecting means. For at least one first region of two regions having different densities which are adjacent to each other, the closer the concentration of the region near the edge line of the first region is to the edge line, An image processing apparatus comprising: a density correction unit that corrects an image signal in the edge line neighboring area so as to approach the density of the other second area of the two areas. 2. The image processing apparatus according to claim 1, wherein the density correction unit corrects the image signal so that the density continuously changes over the two regions. 3. The image processing apparatus according to claim 1, wherein the density correction unit corrects the image signal so that the discontinuity of density on the edge line remains.