JP3479161B2 - Image processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [0001] BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention reads an image of an original,
Image processing of the read digital image data
Plate making equipment (digital printing) that perforates hot stencil paper
Machine), forms a latent image on the photoreceptor by electrophotographic technology and forms it on paper
Transfer device (digital copier) or thermal paper
Etc. related to the copying machine. In particular, the read image
To an image processing device that binarizes and outputs as binary data
Involved. [0002] 2. Description of the Related Art Binary images such as characters and line drawings and gradations of photographs and the like
Use an image processing device such as the one described above
For best results when printing with
Image area is either maximum density or minimum density according to a single threshold
And input / output device characteristics in the gradation image area
It is necessary to perform density conversion to save the input / output density in consideration of
There is. To do this, each part of the image is a binary image area
Or a grayscale image area, and it is necessary to separate the two areas.
It is necessary. [0003] Conventionally, whether a binary image area or a gradation image area is determined.
Divide the image into blocks of nxn pixels to do different things
Feature extraction for each block and use the feature extraction result
(Japanese Patent Laid-Open No. 3-15316)
7) Perform feature extraction using the pixel of interest and its surrounding pixels
And makes a decision for each pixel using the feature extraction result.
Method (JP-A-1-227573). [0004] However, in the former method,
Is used for each block.
Or the border between the binary image area and the gradation image area
There is a problem that the shape of the lock appears. The latter method
In the case of using
Incorrect, but incorrectly identified and correctly identified
And a step in the density occurs, giving a sense of discomfort. [0005] In addition, a thick line drawing or black border in a binary image area is used.
It is difficult to distinguish the data from the high-density part of the photograph in the gradation image area
It is. Thick line drawings and black solids can be distinguished from binary images
Adjusting the discriminating parameter in the image causes the image to collapse
Part can be. High density part of photograph can be distinguished from gradation image
Adjusting the parameters for discrimination as
Concentration decreases. The present invention has been made in order to solve the above problems.
For documents in which value image areas and gradation image areas are mixed,
In the binary image area, the contrast is strong, and in the gradation image area,
Preserves the gradation, and furthermore, the
Provided is an image processing device that outputs an image density signal that does not occur.
The porpose is to do. [0007] [Means for Solving the Problems]To solve the above problems
According to the present invention, an original image input by the original reading means is provided.
Identify the image as a binary image and a gradation image, and perform the corresponding density conversion
Image processing device that outputs an image signal converted by a curve
In (1), the pixel of interest is fine according to the density of the neighboring pixels.
Fine line pixel detecting means for detecting whether or not the pixel is a line pixel,
The target pixel is an edge pixel according to the density of the neighboring pixel.
Edge pixel detecting means for detecting whether or not
Pixel and the edge pixel detection means closest to the pixel of interest.
Counting device that counts the distance to edge pixels detected by
Having a step, whether or not the thin line pixel, the edge pixel
Is based on the information according to whether or not
The state of the eye pixel is changed in multiple steps between the binary image and the gradation image.
Means for judging the state of the target pixel, (2) the target pixel
The input density to the maximum or minimum
Density change for binary images with discontinuous points that tend to binarize
Conversion curve, gradation of input density is stored in output density
Density conversion curve for image and density conversion curve for binary image
One or more discontinuities that interpolate the density conversion curve for the gradation image
Appropriate density conversion curve from density conversion curves with points
And a density conversion curve selecting means for performing density conversion by selecting
An image processing device is assumed. [0008] Attention image used in the apparatus of the present invention
There are various methods as means for identifying the state of the device.
(1) The sharpness of the edge of the pixel of interest according to the density of the neighboring pixels
Is detected, and the state of the pixel of interest is determined. (2) Neighboring painting
Whether the pixel of interest is an edge pixel or not according to the density of the element
Detects the target pixel and the edge pixel closest to the target pixel.
Is counted, and the state of the pixel of interest is determined. (3)
Whether the pixel of interest is a thin line, whether the pixel of interest is an edge pixel
Whether or not there is, the sharpness of the edge of the pixel of interest,
Distance from element edge pixel, length of line segment sandwiching target pixel
Is determined according to the characteristic of each pixel obtained by the above. [0009] According to the present invention, the density can be reduced in the main scanning direction or the sub-scanning direction.
A line segment that scissors the pixel of interest at the rising and falling edges
If there is, count its length. If the length of the line segment is short
It is likely that this is a line that constitutes a character. Also, the present invention
Measure the distance between the target pixel and the edge pixel closest to the target pixel.
To count. Here, the shorter the distance from the edge, the more the character image
The longer the length, the more “tone image likeness”. The present invention is suitable for binary images.Has discontinuities
DoA density conversion curve for a binary image and a floor
In addition to the density conversion curve for gradation images that preserves tonality, both curves
Of the appropriate number to interpolateHave discontinuitiesConcentration conversion curve
It has. [0011] Based on several feature values estimated for each pixel,
When the feature value indicates a strong binary imageDiscontinuous
Have a pointSelect the density conversion curve for the binary image, and
Density conversion for gradation image when showing strong gradation image likeness
Select a curve. And if neither is
Indicates the “feature of a binary image” or “gradation image
The binary image and the gradation image according to the degree of
Several lines interpolating both curves for the imageHave discontinuitiesDensity change
An appropriate curve is selected from the conversion curves and the density is converted. this
The “feature of binary image” or “floor
Tone image-likeness ”,
Due to erroneous area discrimination caused by viewing only the image
It can reduce the discomfort of sudden changes in the density of the output image.
it can. Also, the closer to the edge,Have discontinuities
Density conversion for binary imageHave discontinuitiesDensity conversion song
Select a line and move away from the edge, the density conversion for gradation image
LeaningHave discontinuitiesTo select a density conversion curve
Thicker letters and solid parts can be shown more darkly,
The tone characteristics of the dark part of the toned image can be saved.
You. [0013] Embodiments of the present invention will be described below.
You. (Example 1) FIG. 1 is a block diagram of an image processing apparatus according to an embodiment of the present invention.
You. The original is irradiated with light, and the reflected light from the original is
It is converted to an electric signal (image density signal) by a line sensor and output.
Power. Read by reading means not shown
Image density signal d taken_{i. j}Goes to line memory 2
Is entered. An image density signal output from the reading means
Is input directly to the line memory 2 or
Dotted line to correct for image density signal close to image
Input to the line memory 2 via the MTF correction circuit 1
You. FIG. 2 shows an example in the case of passing through an MTF correction circuit.
The MTF correction coefficient matrix shown as
Pixel density corresponding to each coefficient position for each coefficient of Trix
A comb that multiplies the value and adds each result
The operation of the resolution is performed, and the corrected image signal d 'is obtained.
_{i. j}Is input to the line memory 2. FIG. 3 shows an image stored in the line memory 2.
The position of the pixel corresponding to the density signal on the document is shown. Reading
Pixel corresponding to the image density signal output from the device
To p_{i. j}It expresses. Here, j is on main scanning, i is sub-scanning
Pixel number above, p_{i. j}Is the i-th in the sub-scan,
This is the j-th pixel on scanning. Pixel p_{i, j}The concentration of d
_{i, j}It expresses. Line memory 2 is hatched
Two main scans of the portion and image density signals for two pixels are stored.
You. Image density output from line memory 2
Signal (d_{i, j}, D_{i, j-1}, D_{i, j-2}, D
_{i-1, j}, D_{i-1, j-1}, D_{i-1, j-2}, D
_{i-2, j}, D_{i-2, j-1}, D_{i-2, j-2}) Is
Pixel p_{i-1, j-1}3x3 matrix centered on
Is composed. As shown in FIG. 1, from the line memory 2
Output image density signal d_{i-1, j-5}Is the density conversion
Input to the circuit 3. Output from line memory 2
Image density signal (d_{i, j}, D_{i, j-1},
d_{i, j-2}, D_{i-1, j}, D_{i-1, j-1}, D
_{i-1, j-2}, D_{i-2, j}, D_{i-2, j-1}, D
_{i-2, j-2}) Is input to the edge detection circuit 27.
The edge detection circuit 27 determines that the target pixel is adjacent to the target pixel.
This circuit detects whether or not it is the conversion point of the density with the pixel.
You. FIG. 4 is a diagram showing the edge detection circuit 27.
You. In the arithmetic circuit 13, (a), (b), (c),
Horizontal, vertical, diagonal left, diagonal right directions shown in (d)
Convolution by the edge detection coefficient matrix of
Of the absolute values of the four values obtained
The largest value is the edge signal e_{i-1, j-1}Output as
And input to the comparator 14, the comparator 15, and the comparator 25.
You. In the comparator 15, the input edge signal e
_{i-1, j-1}Is valid if is greater than or equal to the first edge threshold T2
(Signal indicating edge pixel), invalid if less than T2
(Signal indicating that the pixel is not an edge pixel)
No. e1_{i-1, j-1}Is output. First edge threshold T
2 is a sharp (large density difference) that hardly appears in the gradation image
The value is set so that only the edge can be detected. In the comparator 14, the edge signal e
_{i-1. j-1}Is effective if is greater than or equal to the second edge threshold T3,
If it is less than T3, it is invalid and the second edge detection signal e2
_{i-1. j-1}Is output. The second edge threshold T3 is
It is smaller than one edge threshold value T2 and slightly sharp (slightly
(Large) Edge is set to a value that can be detected In the comparator 25, the input edge signal e
_{i-1. j-1}Is valid if is greater than or equal to the third edge threshold T8,
If less than T8, invalid third edge detection signal e3
_{i-1. j-1}Is output. The third edge threshold T8 is
It is smaller than 2 edge threshold value T3 and is dull.
I) The edge is set to a value that can be detected. As shown in FIG. 1, an image of a pixel to be processed
Density signal d_{i-1. j-5}Conversion of concentration to time
To match the edge, the edge detection signal is
A circuits 8, 9, and 26. Output from the dot delay circuits 8, 9, 26
The applied first edge detection signal e1_{i-1. j-5}, Second
Edge detection signal e2_{i-1. j-5}And third edge inspection
Outgoing signal e3_{i-1. j-5}Is input to the feature determination circuit 28.
It is. The feature determination circuit 28 detects the first edge inputted
Signal e1_{i-1, j-5}, The second edge detection signal e2
_{i-1, j-5}, The third edge detection signal e3
_{i-1, j-5}Selects density conversion according to the rule shown in FIG.
Signal g_{i-1, j-5}Is determined and output. This density conversion selection signal g
_{i-1. j-5}Is 1. Edge pixels mainly exist in the binary image area 2. Edge sharpness (density difference) is strong (large)
The more the image looks like a binary image, the less sharp the edge
The smaller (smaller) the more like a gradation image Density conversion curve is selected in consideration of image characteristics such as
It is configured as follows. In FIG. 6, the binary signal has a circle.
Effective, × means invalid,-means ignored. Signals with three or more values
For symbols, the numbers represent signal values. Where the density conversion
Selection signal g_{i-1, j-5}The value of g_{i-1, j-5}= 0, binary image g_{i-1, j-5}= 3, gradation image g_{i-1, j-5}= 1, it looks like a binary image g_{i-1, j-5}= 2, it looks like a gradation image It has a meaning. As shown in FIG. 7, the density conversion circuit 3
Binarize concentration to either maximum or minimumHas discontinuities
DoDensity conversion curve A for binary image (shown by solid line)
And for gradation images that store the gradation of the input density in the output density
The density conversion curve B (shown by a dashed line)
Interpolate curvesHave discontinuitiesDensity conversion music for binary image
From the lineHave discontinuitiesDensity conversion curve C (shown by dotted line)
) And density conversion curve for gradation image.Discontinuities
HaveFrom the density conversion curve D (shown by the two-dot chain line)
TwoHave discontinuitiesA total of four density conversion curves
Has a density conversion curve. Density conversion input from the characteristic determination circuit 28
Selection signal g_{i-1, j-5}In response to the, g_{i-1, j-5}= 0,Have discontinuitiesBinary
Image density conversion curve A, g_{i-1, j-5}= 3, the gradation density conversion curve B, g_{i-1, j-5}If = 1, interpolate 2 curvesDiscontinued
Have a continuation pointDensity conversion curve C, g_{i-1, j-5}= 2, interpolate two curvesDiscontinued
Have a continuation pointDensity conversion curve D, The density conversion curve is selected according to the following rule, and the image density signal d
_{i-1, j-5}Is performed, and the converted density signal cd
_{i-1, j-5}Is output. In the density conversion circuit 3, the above four density
The degree conversion curves correspond to four density conversion curves.
A data conversion table is provided, and the feature determination circuit 28
Density conversion selection signal g output from_{i-1, j-5}According to
Image density while referring to the selected data conversion table.
Degree signal d_{i-1, j-5}Conversion density signal
cd_{i-1, j-5}Is output. The converted density signal converted by the density conversion circuit 3
Number cd_{i-1, j-5}Is input to the binarization circuit 4.
In the binarization circuit 4, the input converted density signal cd
_{i-1, j-5}Is binarized by an error diffusion method, and a binary signal
Output as In this embodiment, the edge detection coefficient matrix
3 × 3. The size of the matrix is nxm
(N> 0, m> 0: n and m are integers)
By changing the number or increasing the number of matrices,
It is possible to improve the edge detection accuracy by the edge detection circuit.
it can. In this embodiment,Have discontinuitiesBinary painting
Interpolate density conversion curve for image and density conversion curve for gradation image
Have discontinuitiesAlthough two density conversion curves are used,
The edge detection circuit 27 may be a book.
By increasing the number of comparators with different thresholds,
By increasing the classification of the sharpness of the edge, both curves
InterpolateHave discontinuities3 or more density conversion curves
It can be extended. (Embodiment 2) Hereinafter, another embodiment of the apparatus of the present invention will be described. Figure
FIG. 8 is a block diagram of an image processing apparatus according to an embodiment of the present invention.
You. Image density output from line memory 2
Signal (d_{i, j}, D_{i, j-1}, D_{i, j-2}, D
_{i-1, j}, D_{i-1, j-1}, D_{i-1, j-2}, D
_{i-2, j}, D_{i-2, j-1}, D_{i-2, j-2}) Is
As in the case of the first embodiment, the pixel p_{i-1, j-1}Centered around
To form a 3 × 3 matrix. As shown in FIG. 8, from the line memory 2
Output image density signal d_{i-1, j-5}Is the density conversion
Input to the circuit 3. Output from line memory 2
Image density signal (d_{i, j}, D_{i, j-1},
d_{i, j-2}, D_{i-1, j}, D_{i-1, j-1}, D
_{i-1, j-2}, D_{i-2, j}, D_{i-2, j-1}, D
_{i-2, j-2}) Is input to the edge detection circuit 29. FIG. 9 is a diagram showing an edge detection circuit. Performance
In the arithmetic circuit 13, as in the case of the first embodiment, FIG.
(B), (c), (d) horizontal, vertical, diagonally left
Direction and right diagonal direction
The calculation of the evolution is performed and the four
The largest value among the absolute values of the values is the edge signal e.
_{i-1, j-1}And output to the comparator 15
You. In the comparator 15, the input edge signal e
_{i-1, j-1}Is effective if is greater than or equal to the first edge threshold T2
(Signal indicating edge pixel), invalid if less than T2
(A signal indicating that the pixel is not an edge pixel)
No. e1_{i-1, j-1}Is output. First edge threshold T
2 is a sharp (large density difference) that hardly appears in the gradation image
The value is set so that only the edge can be detected. As shown in FIG. 8, a dot delay circuit
9 is the first edge detection signal e1
_{i-1. j-5}Is input to the feature determination circuit 30. distance
The classification circuit 11 measures the distance from the pixel of interest to the edge pixel.
Count and classify. FIG. 10 shows details of the distance classification circuit 11.
You. The line memory 23 stores the edge of each pixel for one line.
距離 Distance data from the pixel is stored. Each pixel stored in the line memory 23
The pixel on the original corresponding to the distance data to the edge pixel
The position is shown in FIG. First edge detection signal el
_{i-1. j-5}Is input to the distance counting circuit 22, and the pixel p
_{i-1. j-5}Distance from edge de_{i-1. j-5}
Is the left pixel p_{i-1, j-6}, The pixel p on
_{i-2, j-5}, Upper right pixel p_{i-2, j-4}Edge of
Distance from each, de_{i-1, j-6}, De
_{i-2, j-5}, De_{i-2, j-4}Use the following
Counted by territory. E1_{i-1. j-5}If = 1, de
_{i-1. j-5}= 0 e1_{i-1. j-5}If = 0, de_{i-1. j-5}=
min (de_{i-1, j-6}, De_{i-2, j-5}, D
e_{i-2, j-4}) +1 The above de_{i-1, j-6}, De
_{i-2, j-5}, De_{i-2, j-4}Is line memory
Taken out of 23 and de_{i-1, j-5}Is newly
Is stored in the memory 23. Distance from edge de
_{i-1, j-5}Is input to the comparator 24 and de
_{i-1, j-5}<T6, then f_{i-1, j-5}= 0 T6 ≦ de_{i-1, j-5}<T7, then f
_{i-1, j-5}= 1 T7 ≦ de_{i-1, j-5}If <T8, f
_{i-1, j-5}= 2 T8 ≦ de_{i-1, j-5}Then f_{i-1, j-5}=
3 It is classified according to the following rules. Distance classification signal f
_{i-1, j-5}Is input to the feature determination circuit 30. In the feature determination circuit 30, the first input
Edge detection signal e1_{i-1, j-5}, Distance classification signal f
_{i-1, j-5}, The density according to the rule shown in FIG.
Conversion selection signal g_{i-1, j-5}Is determined. This concentration
Conversion selection signal g_{i-1, j-5}Is 1. Edge pixels mainly exist in the binary image area 2. The distance between the pixel of interest and the edge pixel closest to the pixel of interest
The shorter the separation, the more like a character image, and the longer the separation, the more like a gradation image.
Show Density conversion curve is selected in consideration of image characteristics such as
It is configured as follows. In FIG. 11, there is a circle for binary signals.
Effective, × means invalid,-means ignored. Signals with three or more values
For symbols, the numbers represent signal values. Where the density conversion
Selection signal g_{i-1, j-5}The value of g_{i-1, j-5}= 0, binary image g_{i-1, j-5}= 3, gradation image g_{i-1, j-5}= 1, it looks like a binary image g_{i-1, j-5}= 2, it looks like a gradation image It has a meaning. The density conversion circuit 3 is the same as in the first embodiment.
Has four density conversion curves, and the characteristic determination circuit 3
Density conversion selection signal g input from 0_{i-1, j-5}To
Depending G_{i-1, j-5}= 0,Discontinuity
HavingDensity conversion curve A for binary image g_{i-1, j-5}= 3, gradation density conversion curve B g_{i-1, j-5}If = 1, interpolate 2 curvesDiscontinued
Have a continuation pointDensity conversion curve C (density conversion curve for binary image)
Than) g_{i-1, j-5}If = 2, interpolate two curvesDiscontinuous
Have a pointDensity conversion curve D (from the density conversion curve for gradation image)
R) The density conversion curve is selected according to the following rule, and the image density signal d
_{i-1, j-5}Conversion density signal cd
_{i-1, j-5}Is output. In the density conversion circuit 3, the above four density
The degree conversion curves correspond to four density conversion curves.
A data conversion table is provided, and the feature determination circuit 30
Density conversion selection signal g output from_{i-1, j-5}According to
Image density while referring to the selected data conversion table.
Degree signal d_{i-1, j-5}Conversion density signal
cd_{i-1, j-5}Is output. The density converted and output by the density conversion circuit 3 is
Exchange density signal cd_{i-1, j-5}Is input to the binarization circuit 4.
It is. In the binarization circuit 4, the input converted density signal cd
_{i-1, j-5}Is binarized by an error diffusion method and
And output. In this embodiment, similar to the first embodiment,
The accuracy of edge detection can be improved. Example 1
Similarly,Have discontinuitiesDensity conversion curve and gradation for binary image
Interpolate density conversion curve for imageHave discontinuitiesDensity change
The number of commutation curves may be one, and
Interpolate both curves by increasingHave discontinuities
ToThe density conversion curve can be extended to three or more. (Embodiment 3) Hereinafter, another embodiment of the apparatus of the present invention will be described. Figure
FIG. 12 is a block diagram of an image processing apparatus according to an embodiment of the present invention.
You. Image density output from line memory 2
Signal (d_{i, j}, D_{i, j-1}, D_{i, j-2}, D
_{i-1, j}, D_{i-1, j-1}, D_{i-1, j-2}, D
_{i-2, j}, D_{i-2, j-1}, D_{i-2, j-2}) Is
As in the case of the first embodiment, the pixel p_{i-1, j-1}Centered around
To form a 3 × 3 matrix. As shown in FIG. 12, the line memory 2
Output image density signal d_{i-1, j-5}Is the high concentration line
It is input to the detection circuit 10 and the density conversion circuit 3. line
The image density signal (d output from the memory 2)_{i, j}, D
_{i, j-1}, D_{i, j-2}, D_{i-1, j}, D
_{i-1, j-1}, D_{i-1, j-2}, D_{i-2, j}, D
_{i-2, j-1}, D_{i-2, j-2}) Is a fine line detection circuit
5, the edge detection circuit 7. Fine line detection circuit 5
Is a circuit for detecting whether or not the pixel of interest is a fine line.
In the fine line detection circuit 5, the horizontal direction shown in FIGS.
Line detection in the horizontal direction (main scanning direction) and vertical direction (sub scanning direction)
Convolution operation is performed by the coefficient matrix.
The larger of the absolute values of the two calculated values is
It is compared with the threshold value T1.
Signal if it is less than T1, invalid (signal that it is not a thin line)
No.) Fine line detection signal h_{i-1, j-1}As a process
Pixel image density signal d_{i-1, j-5}Concentration change
Dot delay circuit to match the replacement timing
6 is input. The fine line detection from the dot delay circuit 6
Outgoing signal h_{i-1, j-5}Is output and the feature determination circuit 12
Is input to FIG. 13 shows an edge in the apparatus of this embodiment.
FIG. 3 is a diagram illustrating a detection circuit. Operation circuit 13 is the same as in the first embodiment.
5 (a), 5 (b), 5 (c) and 5 (d)
Edge detection coefficient in vertical, vertical, diagonal left, and diagonal right directions
Matrix performs convolution operation
And the largest value among the absolute values of the four values obtained is
Edge signal e_{i-1, j-1}Is output as comparator 1
4 and 15 are input. In the comparator 15, the input edge signal e
_{i-1, j-1}Is valid if is greater than or equal to the first edge threshold T2
(Signal indicating edge pixel), invalid if less than T2
(A signal indicating that the pixel is not an edge pixel)
No. e1_{i-1, j-1}Is output. First edge threshold T
2 is a sharp (density difference) that hardly appears in the gradation image.
(Large) The value is set so that only the edge can be detected. In the comparator 14, the edge signal e
_{i-1, j-1}Is effective if is greater than or equal to the second edge threshold T3,
The second edge detection signal e2 which is invalid if less than T3
_{i-1, j-1}Is output. The second edge threshold T3 is
It is smaller than one edge threshold value T2 and slightly sharp (slightly
(Large) Edge is set to a value that can be detected. FIG. 14 shows a high density line detection circuit. High concentration
The line detection circuit recognizes a line having a preset thickness and density.
This is the circuit used for
Between the edge pixel and the falling edge pixel, and
This circuit detects whether the density is high. The first edge detection signal e1
_{i-1, j-1}Is a four dot delay circuit
16 and input to the OR circuit 17. At the same time,
Output e1 of delay circuit 16_{i-1. j-2}, E1
_{i-1. j-3}, E1_{i-1. j-4}, E1
_{i-1. j-5}Is input to the OR circuit 17 and the pixel
Edge signal m_{i-1. j-5}Is output as The comparator 19 receives the input image density signal.
d_{i-1. j-5}Is valid if is greater than or equal to threshold T4, less than T4
Then, the invalid high concentration detection signal c4_{i-1. j-5}Is output
Is done. The OR circuit 21 has a dot delay circuit 16
Output e1_{i-1. j-5}, High concentration line detection signal m '
_{i-1. j-5}Is entered. The edge from the OR circuit 21
ヂ or high density line signal bm_{i-1. j-5}Is output
The edge of the previous pixel which is input to the
Is the high density line detection signal bm_{i-1. j-6}Is output. First pixel edge signal m_{i-1. j-5}, Takano
Degree detection signal c4_{i-1. j-5}, The edge or height of the previous pixel
Density line detection signal bm_{i-1, j-6}To AND circuit 18
Is entered. The high density line detection signal m 'is output from the AND circuit 18.
_{i-1. j-5}Is output and the OR circuit 21 and the feature determination
Input to the road 12. Count distance from target pixel to edge pixel
FIG. 15 shows details of the distance classification circuit 11 for performing classification.
The edge memory of each pixel for one line is stored in the line memory 23.
The distance data from the element is stored. FIG. 16 shows the data stored in the line memory 23.
The pixel of the pixel corresponding to the distance data to the edge pixel of each pixel
Indicates the position on the document. First edge detection signal el
_{i-1. j-5}Is input to the distance counting circuit 22, and the pixel p
_{i-1. j-5}Distance from edge de_{i-1. j-5}
Is the left pixel p_{i-1, j-6}, The pixel p on
_{i-2, j-5}, Upper right pixel p_{i-2, j-4}Edge of
Distance from each, de_{i-1, j-6}, De
_{i-2, j-5}, De_{i-2, j-4}Use the following
Counted by territory. E1_{i-1. j-5}If = 1, de
_{i-1. j-5}= 0 e1_{i-1. j-5}If = 0, de_{i-1. j-5}=
min (de_{i-1, j-6}, De_{i-2, j-5}, D
e_{i-2, j-4}) +1 The above de_{i-1, j-6}, De
_{i-2, j-5}, De_{i-2, j-4}Is line memory
Taken out of 23 and de_{i-1. j-5}Is newly
Is stored in the memory 23. Distance from edge de
_{i-1. j-5}Is input to the comparator 24, d_{i-1. j-5}<T6, then f_{i-1. j-5}=
0 T6 ≦ de_{i-1. j-5}<T7, then f
_{i-1. j-5}= 1 T7 ≦ de_{i-1. j-5}If <T8, f
_{i-1. j-5}= 2 T8 ≦ de_{i-1. j-5}Then f_{i-1. j-5}=
3 It is classified according to the following rules. Distance classification signal f
_{i-1. j-5}Is input to the feature determination circuit 12. In the characteristic determination circuit 12, the input fine line is detected.
Signal h_{i-1. j-5}, The first edge detection signal e1
_{i-1. j-5}, The second edge detection signal e2
_{i-1. j-5}, High concentration line detection signal
m '_{i-1. j-5}, Distance classification signal f_{i-1. j-5}To
The density conversion selection signal g according to the rule shown in FIG.
_{i-1. j-5}Is determined and output. This density conversion selection signal g
_{i-1. j-5}Is 1. Fine line pixels and edge pixels mainly exist in the binary image area.
Exist 2. Edge sharpness (density difference) is strong (large)
Shows a more binary image quality, with a lower edge sharpness
(Smaller) indicates the likeness of a gradation image. 3. Density rise in main scanning direction or sub scanning direction
Line sandwiching the pixel of interest between the edge pixel and the falling edge pixel
If there is a minute, characters are composed if the density of the line is high and thin
Line (part of the binary image). 4. The distance between the pixel of interest and the edge pixel closest to the pixel of interest
The shorter the separation, the more like a binary image. The longer the separation, the more like a gradation image.
Show Density conversion curve is selected in consideration of image characteristics such as
It is configured as follows. In FIG. 17, the binary signal has a circle.
Effective, × means invalid,-means ignored. Signals with three or more values
For symbols, the numbers represent signal values. Where the density conversion
Selection signal g_{i-1. j-5}The value of g_{i-1. j-5}= 0, binary image g_{i-1. j-5}= 3, gradation image g_{i-1. j-5}= 1, it looks like a binary image g_{i-1. j-5}= 2, it looks like a gradation image It has a meaning. As shown in FIG. 7, the density conversion circuit 3
Binarize concentration to either maximum or minimumHas discontinuities
DoDensity conversion curve A for binary image (shown by solid line)
And for gradation images that store the gradation of the input density in the output density
The density conversion curve B (shown by a dashed line)
From the density conversion curve for binary imageDiscontinuity
HavingDensity conversion curve C (indicated by dotted line) and gradation
From the density conversion curve for imageHave discontinuitiesDensity conversion song
Two density changes consisting of line D (indicated by the two-dot chain line)
And a total of four density conversion curves. Density conversion input from the characteristic determination circuit 12
Selection signal g_{i-1. j-5}Depending on g_{i-1. j-5}= 0,Have discontinuitiesBinary
Image density conversion curve A g_{i-1. j-5}= 3, gradation density conversion curve B g_{i-1. j-5}If = 1, interpolate 2 curvesDiscontinued
Have a continuation pointDensity conversion curve C (density conversion curve for binary image)
Than) g_{i-1. j-5}= 2, interpolate two curvesDiscontinued
Have a continuation pointDensity conversion curve D (density conversion curve for gradation image)
Than) The density conversion curve is selected according to the following rule, and the image density signal d
_{i-1. j-5}Conversion density signal cd
_{i-1. j-5}Is output. The density conversion circuit 3 is the same as in the first embodiment.
is there. In the present embodiment, the fine line detection coefficient matrix, the edge
The detection coefficient matrix is 3 × 3.
Change the size to n × m (n> 0, m> 0: n and m are integers)
, Change the coefficient, and increase the number of matrices.
This makes the edge detection circuit 7 more accurate in edge detection.
You can increase the degree. In addition, the high density line detection circuit
By increasing the number of delay stages in the
The thickness of the line to be detected can be increased. Further,Have discontinuitiesDensity change for binary image
Interpolation curve and density conversion curve for gradation imageDiscontinuities
HaveThe density conversion curve is two, but only one
And different threshold values of the edge detection circuit 7
By increasing the number of comparators
It is possible to increase the degree of classification, and
Density conversion music that interpolates both curves by increasing the number of classes
The line can be extended to three or more lines. [0069] According to the apparatus of the present invention, the erroneous determination of the area can be performed.
Discomfort due to sudden changes in the density of the output image due to
You. Character images such as characters and thin lines and gradation images such as photographs are mixed.
Output an image density signal that is faithful to the original
can do. Also,Have discontinuitiesDensity change for binary image
Interpolation curve and the density conversion curve for gradation image.
Become multipleHave discontinuitiesDensity according to the density conversion curve
By providing the degree conversion information, thick characters and solid parts
Even so, the edge part can be thickened, and it looks black
Can be Furthermore, as you move away from the edge,
Since the density can be converted to a gradation, the density of the gradation image
Effects such as the ability to preserve the gradation of dark areas
is there.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of an image processing apparatus according to an embodiment of the present invention. FIG. 2 is a diagram illustrating a coefficient matrix of an MTF correction circuit used in an embodiment of the present invention. FIG. 3 is a diagram showing positions on a document of pixels in a line memory for storing an image density signal according to the embodiment of the present invention. FIG. 4 is a diagram illustrating an edge detection circuit according to an embodiment of the present invention. FIG. 5 is a diagram illustrating a coefficient matrix of an edge detection circuit according to the embodiment of the present invention. FIG. 6 is a table showing a feature determination rule according to the embodiment of the present invention. FIG. 7 is a diagram illustrating an example of a density conversion curve of the density conversion circuit according to the embodiment of the present invention. FIG. 8 is a block diagram of an image processing apparatus according to an embodiment of the present invention. FIG. 9 is a diagram illustrating an edge detection circuit according to an embodiment of the present invention. FIG. 10 is a diagram showing a distance classification circuit according to the embodiment of the present invention. FIG. 11 is a table showing a feature determination rule according to the embodiment of the present invention. FIG. 12 is a block diagram of an image processing apparatus according to an embodiment of the present invention. FIG. 13 is a diagram illustrating an edge detection circuit according to an embodiment of the present invention. FIG. 14 is a diagram illustrating a high density line detection circuit according to an embodiment of the present invention. FIG. 15 is a diagram illustrating a distance classification circuit according to an embodiment of the present invention. FIG. 16 shows a line memory 23 according to an embodiment of the present invention.
Indicates the position on the document of the pixel corresponding to the distance data to the edge pixel of each pixel stored in the document. FIG. 17 is a table showing a feature determination rule according to the embodiment of the present invention. FIG. 18 is a diagram showing a coefficient matrix of the fine line detection circuit used in the embodiment of the present invention. [Description of Signs] 1 MTF correction circuit 2 Line memory 3 Density conversion circuit 4 Binarization circuit 5 Thin line detection circuit 6 Dot delay circuit 7 Edge detection circuit 8 Dot delay circuit 9 Dot delay circuit 10 High density line detection Circuit 11 Distance classification circuit 12 Feature determination circuit 13 Operation circuit 14 Comparator 15 Comparator 16 Dot delay circuit 17 Logical OR circuit 18 Logical product circuit 19 Comparator 20 Dot delay circuit 21 Logical OR circuit 22 Distance counting circuit 23 Line Memory 24 comparator 25 comparator 26 dot delay circuit 27 edge detection circuit 28 feature determination circuit 29 edge detection circuit 30 feature determination circuit

──────────────────────────────────────────────────続 き Continuation of front page (56) References JP-A-4-336874 (JP, A) JP-A-1-169683 (JP, A) JP-A-5-284358 (JP, A) JP-A-4- 220080 (JP, A) JP-A-1-181380 (JP, A) JP-A-7-87319 (JP, A) JP-A-61-157169 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H04N 1/40-1/409 H04N 1/46 H04N 1/60

Claims (1)

(57) Claims 1. A document image input by a document reading means is identified as a binary image and a gradation image, and an image signal converted by a corresponding density conversion curve is output. In the image processing apparatus, (1) a thin line pixel detecting unit that detects whether a pixel of interest is a thin line pixel according to the density of a neighboring pixel, and whether the pixel of interest is an edge pixel according to the density of the neighboring pixel Edge pixel detecting means for detecting whether or not the target pixel and the edge pixel detected by the edge pixel detecting means closest to the target pixel, distance measuring means for counting the distance,
The state of the target pixel is determined in multiple stages between the binary image and the gradation image based on the information according to whether the pixel is the thin line pixel, whether the pixel is the edge pixel, and the distance. (2) There is a discontinuous point for binarizing the input density to either the maximum or the minimum according to the determination of the state of the target pixel.
A density conversion curve for a binary image, a density conversion curve for a tone image for storing the gradation of an input density as an output density, and a density conversion curve for the binary image and the density conversion curve for the tone image 1 An image processing apparatus comprising a density conversion curve selecting means for selecting an appropriate density conversion curve from density conversion curves having more than two discontinuous points and performing density conversion.