JPH05300369A - Picture processing unit - Google Patents

Abstract

PURPOSE:To provide the excellent picture processing unit in which moire is eliminated by detecting an edge in the main scanning direction and the subscanning direction even from an original of a dot photograph as well as a character and a picture. CONSTITUTION:The picture processing unit is provided with a frame memory 4 storing a picture signal obtained from a read section 1 and converted into digital data by an A/D converter 2, a ROM 5 as an area decision means deciding a binarizing area comprising plural picture elements and one or plural peripheral areas adjacent to the binarizing area and comprising picture elements of the same number for those for the binarizing area, a density detection means detecting the density level of the binarizing area and the peripheral area, an edge detection means detecting a picture edge from the density level and a CPU 3 provided with a binarizing processing means implementing different binarizing processing to the binarizing area depending on the result of detection of the edge detection means.

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 processing image data such as an image scanner, a facsimile and a copying machine.

[0002]

2. Description of the Related Art When an image processing apparatus such as an image scanner, a facsimile or a copying machine, which processes image data, inputs an image of a striped original or a dot original,
So-called "moire" occurs. A smoothing process is performed to remove this moire. However, when the smoothing process is performed, a phenomenon occurs in which an edge that is an original contour portion is blurred. Therefore, in the conventional image processing device,
There has been performed a process of emphasizing an edge portion using a spatial filter or determining that there is an edge when the value of the spatial filter exceeds a predetermined threshold value and not performing smoothing processing.

[0003]

In the above-described conventional image processing apparatus, it is effective when the original to be read is a character or a photograph, but when the original is a halftone photograph, each dot is a dot.
However, there is a problem that moire cannot be removed because each dot is emphasized and moire is emphasized, and since each dot is judged to be an edge and smoothing processing is not performed.

A method for solving this problem is disclosed in JP-A-63-6.
It is disclosed in Japanese Patent No. 9376. The method is to measure the distance between the maximum point and the next maximum point from the density waveform of one line of image data, and if the distance is more than a certain threshold and the maximum points are periodic, it is judged as a halftone photograph. , Others are the edges of text or photographs. Next, the difference in the average density between the maximum points is extracted to extract the edges of the halftone dot photograph, and the smoothing process is performed on the other than the edges.

However, this method has a problem that the edge in the main scanning direction can be detected but the edge in the sub scanning direction cannot be detected.

The present invention solves the problems of the disclosed techniques described in the above-mentioned conventional examples and gazettes, and detects the edges in the main scanning direction and the sub-scanning direction not only for characters and photographs but also for halftone dot originals. Then, it aims at providing the outstanding image processing apparatus which can remove a moire.

[0007]

In order to achieve the above object, the present invention comprises a binarized area consisting of a plurality of pixels, the same number of pixels as the binarized area, and being adjacent to the binarized area. Area determining means for determining one or a plurality of peripheral areas; density detecting means for detecting density levels of the binarized area and peripheral area; edge detecting means for detecting edges of the image from the density levels; Binary processing means for performing different binarization processing on the binarized area according to the detection result of the edge detection means.

[0008]

Therefore, according to the present invention, the density levels of the binarized area composed of a plurality of pixels and the peripheral area composed of the same number of pixels are compared to detect whether or not it is an edge. Of course, even if it is a manuscript of halftone pictures,
It is possible to remove moire in the main scanning direction and the sub scanning direction.

[0009]

Embodiments of the present invention will now be described in detail with reference to the drawings.

FIG. 1 is a schematic block diagram of an embodiment of an image processing apparatus according to the present invention. In FIG. 1, reference numeral 1 denotes an image reading unit for reading an image of a document, which includes a CCD sensor (not shown) and converts optical information obtained from the document into an electric signal and outputs an analog image signal. 2 is 8-bit A
The D / D converter converts an analog image signal into digital image data of 256 gradations of black (0) to white (255) (hereinafter, referred to as image data). Reference numeral 3 denotes a CPU that performs various calculations, transfers the image data to a memory, performs a calculation process for binarizing the image data, detects the density level of the image data in a predetermined area, and detects the edge of the image. Performs arithmetic processing, etc. Reference numeral 4 is a frame memory for storing at least one frame of image data of 256 gradations. Reference numeral 5 stores an image processing program executed by the CPU 3, and stores threshold data required for binarization.
OM. The ROM 5 includes a rewritable ROM that can be erased by ultraviolet rays or electrically.

Next, in the configuration of the above embodiment, the CPU
The image processing method executed by the method 3 is not limited to one, but various methods are possible. Here, the first and second embodiments will be described.

FIG. 2 is a diagram showing a unit area set in order to carry out the binarization processing in the first embodiment, and a characteristic amount is calculated for each unit area. As shown in FIG. 2, this unit area is composed of a binarized area E ₅ and six areas of its peripheral areas E ₁ to E ₄ and E ₆ . Further, each area is 4 × 4 in the main scanning direction and the sub-scanning direction.
Matrix unit, that is, 16 pixels. FIG. 3 shows the matrix of the area E ₅ . In FIG. 3, I (j, i) (where j is the position of the matrix,
i represents the position of a pixel in the matrix) represents the density of each pixel forming the matrix. The area designating data of the binarized area and its peripheral area, and the matrix configuration are stored in the ROM 5 in advance and read by the CPU 3 as necessary. Therefore this R
The OM 5 cooperates with the CPU 3 to form a region determining means for determining the binarized region and its peripheral region.

Edge detection is performed based on this unit area. For that purpose, the average density of each area is obtained from the density calculation formula shown in the following formula.

[0014]

[Equation 1] FIG. 4 is a diagram showing the process from the average density obtained by the density calculation formula described above to the detection of an edge. FIG. 4A shows the average densities m ₁ , m ₂ , ..., M ₆ of each area.
FIG. 4B shows a combination of areas of average densities to be compared in order to detect a diagonally left edge. Here, assuming that the edge determination threshold value per pixel is TA,
If the condition of the following equation is satisfied, it is determined that the edge is a diagonally left edge.

[0015]

[Equation 2] FIG. 4C shows a combination of average density regions to be compared in order to detect a right slanted edge. If the condition of the following equation is satisfied, it is determined that the edge is a right diagonal edge.

[0016]

[Equation 3] FIG. 4D shows a combination of areas of average density to be compared in order to detect a vertical edge. Whether it is a vertical edge or not is determined by whether or not the following expression is satisfied.

[0017]

[Equation 4] After the edge detection is performed, the binarization processing is performed on the binarized area E ₅ of FIG. 2. At this time, each pixel in the matrix of the area E ₅ is processed by the threshold matrix for performing the dither processing shown in FIG. To binarize.

If the number of pixels in the main scanning direction is H and the number of lines in the sub scanning direction is V, the positions of pixels and lines during binarization are counted by a counter in each direction. When the counter value in the main scanning direction is X and the counter value in the sub-scanning direction is Y, X changes from 0 to H and Y changes from 0 to V. The remainder obtained by dividing the counter value X in the main scanning direction by 8 is represented by X% 8, and the remainder obtained by dividing the counter value Y in the sub-scanning direction by 8 is represented by Y% 8.
Is a value of j = X% 8 and k = Y% 8.

Therefore, in both the main scanning direction and the sub-scanning direction, binarization processing is performed with the same threshold value for every eight pixels. However, the threshold matrix in FIG. 5 is 8 × 8.
And the binarization target matrix of the region E ₅ is 4 ×
4, the threshold matrix is composed of four sub-matrices A whose unit is a 4 × 4 matrix,
B, C, and D are set, and one of these partial matrices is compared with the area E ₅ to perform binarization processing. In this case, assuming that the value at the j-th row and the k-th column of the threshold matrix of FIG. 5 is D (j, k), when it is determined that the region E ₅ is an edge, I
If (5, i) ≧ D (j, k), a white pixel is output and I
If (5, i) <D (j, k), black pixels are output.
When it is determined that the area E ₅ is not an edge, m ₅ ≧ D
In the case of (j, k), a white pixel is output and m ₅ <D (j,
In the case of k), a black pixel is output.

Next, in the first embodiment, the CP of FIG.
A series of operations of the process of binarizing the area E ₅ executed by U3 will be described with reference to the operation flowchart of FIG. In the present embodiment, for ease of explanation, the maximum values (H + 1) and (V + 1) of the count numbers in the main scanning direction and the sub scanning direction are set to multiples of 4. Therefore, H% 4 = 3 and V% 4 = 3.

First, the counter values X and Y in the main-scanning and sub-scanning directions are set to initial values 0 (step S1). X
The data of the density levels of the pixels at the positions of = 0 and Y = 0 are read from the frame memory 4 to the CPU 3 (step S
2). Next, it is determined whether or not the count value X of the pixels of the read data in the main scanning direction is X% 4 = 3, that is, a multiple of 4 (step S3), and it is not a multiple of 4. X is incremented by 1 to X + 1 (step S4), the process proceeds to step S1, the data of the next pixel on the same line is read, and steps S2 to S4 are executed until X% 4 = 3. Each of these steps is a routine for reading the density data of 4 pixels in the main scanning direction of the 4 × 4 target matrix.

When X% 4 = 3 in step S3, it is determined whether or not the count value Y in the sub-scanning direction is Y% 4 = 3 (step S5), and Y% 4 = 3. If not, X is set to X-3, Y is set to Y + 1 (step S6), and the process proceeds to step S2 to read data. That is, reading of the four pixels of the next line is started. If Y% 4 = 3 in step S5, it means that all the data of the 4 × 4 target matrix of the area E ₅ has been read, and therefore the average of the area E ₅ is calculated according to (Equation 1). The density level is detected (step S7).

In the processing from step S1 to step S7, the average density is detected not only in the binarized area E ₅ but also in other areas E ₁ to E ₆ except E _5. To do.

Based on the average density level calculated in step S7 as the density detecting means, (Equation 2), (Equation 3)
And (Equation 4) are used to determine whether or not it is an edge (step S8). In step S8 as the edge detection means, the process proceeds to the step as the binarization processing means that performs different binarization processing according to the detection result.
That is, if it is not an edge, binarization is performed with the average density (step S9), and if it is an edge, binarization is performed with the same density (step S10).

Next, it is determined whether or not the latest fetched pixel data, that is, I (5,16) in FIG. 3 is the last pixel of the line, that is, whether X = H ( Step S11), unless X = H, X = X + 1, Y = Y
-3 (step S12), the process proceeds to step S2 to read the next pixel data three lines before. Then, the processing from step S2 to step S12 is repeatedly executed until X = H. This is a routine for performing binarization processing for four lines.

When X = H in step S11, it is determined whether or not Y = V, that is, whether or not it is the last line (step S13). X = 0, Y = Y + 1 (step S1
4) Go to step S2 to read the first pixel data of the next line. Then, step S2 is performed until Y = V.
The processing from to step S14 is repeatedly executed. If Y = V in step S13, the binarization process for one frame ends.

As described above, in the first embodiment,
When binarizing a region composed of a plurality of pixels, it is possible to determine whether or not the binarized region is an edge by obtaining an average density level of one or a plurality of peripheral regions composed of the same number of pixels. The binarization processing is performed by a different method based on the determination result. Therefore, not only in characters and photographs but also in halftone photographs, it is possible to remove moire not only in the main scanning direction but also in the sub-scanning direction and enhance the edges.

In the first embodiment, the average density is detected for every 16 pixels, but the average density for every 4 pixels in the main scanning direction is calculated for one line, and the average density for 4 lines is calculated, and then 2 is calculated. The average density of the binarized area may be detected.

Further, although the average density is obtained in the first embodiment, the edge may be detected by obtaining the total density.

Next, the second embodiment will be described. FIG. 7 shows a unit area for binarization. Each area of E ₁ to E ₉ is composed of 16 pixels of a 4 × 4 matrix as in the first embodiment, and E ₅ is a matrix for binarization. The pixel positions of the matrix E ₅ have the structure shown in FIG. 3 as in the first embodiment. FIG. 8 shows the characteristic amount P of the density level of the pixel in each area of FIG.

FIG. 9 shows the CP of FIG. 1 in the second embodiment.
7 is a flowchart of a series of operations of binarizing the region E ₅ executed by U3, which is the same as the flowchart of FIG. 6 according to the first embodiment except step S70 and step S80, and the description thereof will be given. Omitted, different parts will be described.

FIG. 10 shows the extreme value detection subroutine of step S70. The maximum density of the matrix of E ₅ , which is the binarization processing area, is calculated (step S7).
1) The minimum density is calculated (step S72). Next, the minimum maximum value and the minimum maximum value are calculated (steps S73 and S74). Then, it is determined whether P _{0 of the} binarized area E ₅ is an extreme value.

The judgment of the extreme value is made to be the maximum value with respect to the surrounding pixels around P ₀ if all the surrounding pixels satisfy the following condition (Equation 5).

[0034]

[Equation 5] Further, regarding the surrounding pixels around P ₀ , if all the surrounding pixels satisfy the following condition (Equation 6), it is set to the minimum value.

[0035]

[Equation 6] Using this maximum calculation method, the extreme value judgment is performed for all the pixels in the area shown in FIG. Finally, the maximum concentration, minimum concentration, maximum minimum concentration and minimum maximum concentration are calculated, and if there is no extreme value, no extreme value is set.

The processing from step S1 to step S70 is performed not only on the binarized area E ₅ but also on each of the areas E ₁ to E ₉ other than E ₅ .

Next, it is detected whether the matrix of E ₅ is an edge based on the obtained feature amount (maximum density, minimum density, maximum minimum density, minimum maximum density) (step S).
80). The conditions for being an edge are: No extremum,
And (maximum density) − (minimum density) ≧ TA, 2.
(Maximum concentration)-(minimum maximum concentration) ≧ TA, 3.
In the case of (maximum minimum density)-(minimum density) ≧ TA, In other cases, it is determined that the edge is not an edge.

The binarization processing based on this judgment result is the same as that of the first embodiment. When it is judged that the area E ₅ is an edge, I (5, i) ≧ D (j, k). ), A white pixel is output, and if I (5, i) <D (j, k), a black pixel is output. When it is determined that the area E ₅ is not an edge, a white pixel is output when m ₅ ≧ D (j, k), and a black pixel is output when m ₅ <D (j, k).

As described above, also in the second embodiment,
Similar to the first embodiment, when binarizing a region including a plurality of pixels, the maximum density, the minimum density, the maximum minimum density, and the minimum maximum density of one or a plurality of peripheral areas including the same number of pixels are determined. By determining, whether or not the binarized area is an edge is discriminated, and the binarization process is performed by a different method based on the discrimination result. Therefore, not only in characters and photographs but also in halftone photographs, it is possible to remove moire not only in the main scanning direction but also in the sub-scanning direction and enhance the edges.

[0040]

As is apparent from the above embodiment, the present invention is a region in which a binarized region including a plurality of pixels and one or a plurality of peripheral regions including the same number of pixels are determined as unit regions. A determination unit is provided, the density detection unit detects the density level of this unit area, and the detection result determines whether or not it is an edge. Since different binarization processing is performed by the and binarization processing means, moiré is removed not only in the main scanning direction but also in the sub-scanning direction and the edges are emphasized, not only in characters and photographs but also in halftone dot photographs. The effect is obtained.

[Brief description of drawings]

FIG. 1 is a schematic block diagram showing a configuration of an exemplary embodiment of the present invention.

FIG. 2 is a diagram showing a binarized unit area in the first embodiment of the present invention.

FIG. 3 is a diagram showing a matrix of pixels in a binarized region in FIG.

FIG. 4A is a diagram showing the average density of the unit area of FIG. (B) to (d) are diagrams showing regions to be compared for edge detection.

FIG. 5 is a diagram showing an 8 × 8 threshold matrix of the present invention.

FIG. 6 is an operation flowchart of a binarization process in the first embodiment of the present invention.

FIG. 7 is a diagram showing a binarized unit area in the second embodiment of the present invention.

8 is a diagram showing feature values in FIG. 7. FIG.

FIG. 9 is an operation flowchart of binarization processing in the second embodiment of the present invention.

10 is a flowchart of an extreme value detection subroutine in FIG.

[Explanation of symbols]

 1 image reading unit 2 A / D converter 3 CPU 4 frame memory 5 ROM

Claims (5)

[Claims] 1. An image processing apparatus for performing binarization processing on image data having gradation obtained by reading an image of an original, the binarization area having a plurality of pixels, and the same number of the binarization areas. Area determining means for determining one or a plurality of peripheral areas each of which is adjacent to the binarized area and density detecting means for detecting the density levels of the binarized area and the peripheral area.
An edge detection unit that detects an edge of the image from the density level; and a binarization processing unit that performs different binarization processing on the binarized region according to a detection result of the edge detection unit. Characteristic image processing device. 2. The image processing apparatus according to claim 1, wherein the density detecting unit obtains an average density or a total density in each of the binarized area and the peripheral area. 3. The density detecting means obtains a maximum density, a minimum density, a maximum minimum value density and a minimum maximum value density in each of the binarized area and the peripheral area. Image processing device. 4. The image processing according to claim 2, further comprising edge detection means for detecting the edge based on a difference in average density or a difference in total density in each of the binarized area and the peripheral area. apparatus. 5. An edge detecting unit for detecting the edge based on a maximum density, a minimum density, a maximum minimum density and a minimum maximum density in each of the binarized area and the peripheral area. Item 3. The image processing device according to item 3.