JP3615308B2 - Image processing apparatus and method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus and method having a function of determining a halftone period of an image.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, various apparatuses for switching processing such as filtering by detecting a period of a document, particularly a halftone dot, have been proposed. For example, the technique described in Japanese Patent Application Laid-Open No. 1-133470 is an example of an image processing method for detecting a halftone dot period by pattern matching and switching filter coefficients. In addition, the method described in Japanese Patent Laid-Open No. 3-62355 uses pair length information (the number of pairs of white bit length and black bit length that are continuously developed adjacent to each other) as a feature value of halftone dot period detection. Thus, the period is detected.
[0003]
The present applicant has previously proposed a color image processing apparatus (Japanese Patent Application No. 8-64538) that performs filtering suitable for an output image by an error diffusion method, and further, image area separation according to the period of the document. Have proposed an image processing method (Japanese Patent Application No. 8-84185) for changing the determination criteria of the above, all of which detect periodicity using DCT (Discrete Cosine Transform).
[0004]
[Problems to be solved by the invention]
All of the above techniques use the characteristics of the halftone part of the halftone dot, but have the following problems. That is,
(1) Misidentified highlight and dark areas;
Regardless of which feature amount is used, the period of the halftone dot is not correctly detected in the highlight portion and the dark portion. This is because in the case of the highlight portion, a halftone dot is missing and a desired gradation image is created. That is, in the example in which the period is detected by pattern matching, there is a possibility that an image including many highlight portions may be determined to have a coarser line number, that is, a lower line number side, depending on the pattern to be prepared.
[0005]
(2) Misidentification of halftone dot portion of secondary color or higher;
The signal to be determined is generally a luminance signal for a monochrome reproduction machine, and min (R, G for the purpose of positively detecting a green signal or a yellow halftone dot for a color reproduction machine. , B) [when the maximum density is 0] or the like is used. A halftone dot is an image created by superimposing four color dots. For example, if a green signal is used to pattern match a portion made up of cyan and magenta secondary colors, it will not match exactly. In other words, there is a possibility that the denser number of lines, that is, the higher number of lines is determined. This is because the characteristic of the green signal is that both cyan and magenta have a response, and conversely, it is not a filter that can select only magenta.
[0006]
The present invention has been made in consideration of the above circumstances,
An object of the present invention is to provide an image processing apparatus and method for determining the halftone dot period of an image more accurately by excluding an area where the halftone dot period cannot be accurately detected from the period detection target.
[0007]
[Means for Solving the Problems]
In order to achieve the above object, according to the first aspect of the present invention, means for reading a document as image data;
An average value calculating means for dividing the image data into a plurality of blocks and calculating an average value of pixel values of the blocks;
Primary color area detection means for determining the block as a primary color area when the average pixel value of the block is within a predetermined range;
Average DCT calculating means for calculating an average DCT coefficient by calculating a DCT coefficient of a block determined to be a primary color area and dividing a value obtained by adding the DCT coefficients at the same position in each block by the number of primary color areas ,
A period calculation means for detecting a peak position in the block by performing a predetermined calculation on the average DCT coefficient;
And a filter control means for performing a filter process using a filter coefficient corresponding to the calculation result of the period calculation means .
In the invention of claim 2 , the step of reading the document as image data;
An average value calculating step of dividing the image data into a plurality of blocks and calculating an average value of pixel values of the blocks;
A primary color region detection step of determining the block as a primary color region when the average value of the pixel values of the block is within a predetermined range;
An average DCT calculating step of calculating an average DCT coefficient by calculating a DCT coefficient of a block determined to be a primary color area and dividing a value obtained by adding the DCT coefficients at the same position of each block by the number of primary color areas; ,
A period calculation step of detecting a peak position in the block by performing a predetermined calculation on the average DCT coefficient;
And a filter control step of performing a filter process using a filter coefficient corresponding to the calculation result of the period calculation step .
[0008]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.
<Example 1>
FIG. 1 shows the configuration of Embodiment 1 of the present invention. In the figure, 1 is an image input device such as a scanner, 2 is a log conversion circuit that converts a linear signal of reflectance into a linear signal of density, 3 is a filter circuit comprising a smoothing filter, an edge enhancement filter, and 4 is a cmy signal. Color correction circuit for converting to complementary color YMC, 5 is a UCR circuit that performs a process of subtracting K color signals from the signal of each color material, 6 is a dither circuit that expresses halftones using text and picture dithers, and 7 is An image output device such as a printer.
[0009]
Reference numeral 8 denotes a periodicity determination circuit that operates during pre-scanning, and includes a periodicity detection circuit 9 and an intermediate region detection circuit 10. Reference numeral 11 denotes a coefficient switching circuit that switches filter coefficients according to the determination result of the periodicity determination circuit.
[0010]
First, the outline of FIG. 1 will be described. Rgb data (8 bits each) that is substantially linear with respect to the reflectance is output from the image input device 1 (resolution is 400 dpi) such as a scanner. Next, log conversion is performed in the Log conversion circuit 2 as preprocessing for color correction in the subsequent stage, and density linear cmy data is generated.
[0011]
The filter circuit 3 is configured by a circuit in which a smoothing filter and an edge enhancement filter as shown in FIG. 2 are connected in series, and the coefficient of the smoothing filter is determined according to the determination result from the periodicity determination circuit 8 described later. Switch.
[0012]
The color correction circuit 4 considers the characteristics of the image input device 1 and the image output device 7 and converts the cmy signal into a complementary color YMC signal. As a color correction method, a linear approximation so-called masking method, tetrahedral interpolation method, triangular prism interpolation method, and the like have been proposed, and color correction is performed using these methods. At the same time, the K signal is calculated from min (C, M, Y).
[0013]
The UCR circuit 5 subtracts the CMY signal based on the following equation.
K ′ = 0.6 × K
C ′ (M ′, Y ′) = C (M, Y) −K ′
The dither circuit 6 performs data conversion using a dither table (dither matrix) and sends the data to the image output device 7.
[0014]
FIG. 3 shows the configuration of the periodicity detection circuit. Here, periodicity is detected using DCT (Discrete Cosine Transform), and it is composed of an average DCT coefficient calculator 21 and a period calculator 22. Details of DCT are described in, for example, Japanese Patent Laid-Open No. 4-21265.
[0015]
The DCT is performed in units of blocks of 8 pixels × 8 pixels. For example, if the number of pixels is 4752 × 3360 pixels when a general A4 size is read at 400 dpi, processing for 594 × 420 blocks is performed. Since the periodicity of the original appears mainly in the intermediate portion, the intermediate area detection circuit 10 determines that the effective block is only when the average pixel value in the block is within a certain range (th1, th2). The periodicity detection circuit 9 adds the DCT coefficients and finally obtains the average. That is, for the reasons described above, the highlight portion and dark portion of the image are not subject to period detection.
[0016]
FIG. 4 is a process flowchart of the average DCT coefficient calculation unit. FIG. 5 is a diagram showing how the DCT coefficients are averaged.
Image data for 8 lines is set (step 101), and an average value Ave of pixel values in the block is calculated (step 102). That is, the periodicity of the document appears mainly in the intermediate portion (that is, all the pixels in the block are not black pixels or white pixels), and therefore the average of the pixel values in the block is within a certain range (th1, th2). Only in this case, the effective block is determined (step 103), and a two-dimensional DCT coefficient is calculated (step 104). Addition is performed for each coefficient (step 105). For example, as shown in FIG. 5, DCT coefficients A00, B00, C00 at the same position in each block are added.
[0017]
The above processing is performed in the main scanning direction (step 107), and when the processing in the main scanning direction is completed (step 106), the same processing is performed in the sub scanning direction (step 109), and the processing in the sub scanning direction is completed (step 107). 108) The average DCT is calculated (step 110). In the example of FIG. 5, the average DCT for three blocks, F00 = (A00 + B00 + C00) / 3,. . . , F77 is obtained.
[0018]
The period calculation unit 22 detects a peak position using a 3 × 3 mask because a peak appears in the DCT coefficient when the document has a specific period. Specifically, a mask as shown in FIG. 6 is applied to the entire average DCT coefficient of 8 × 8, and M is calculated at each position of 8 × 8 according to the equation shown in FIG. Then, the period of the document of interest is determined from the peak position having the maximum M value. Here, the determination is made into four stages of approximately 50 lines, approximately 100 lines, approximately 150 lines, and approximately 200 lines depending on the position where the peak position exists. FIG. 7 is a diagram for explaining such determination of the number of lines. If no peak is detected or the maximum M is smaller than a predetermined value, it is determined that there is no periodicity.
[0019]
As described above, periodicity detection using DCT has been described. However, the present invention is not limited to this, and as described in Japanese Patent Laid-Open No. 1-133470 described above, for example, locality is detected by pattern matching. It is also possible to provide a means for detecting a portion having a period, and to determine the most correlated one among the detected periods as the period of the document. However, even in this case, for example, when matching is performed in units of blocks, the average of the target block (for example, an 8 × 8 mask) is calculated, and only when the average value is within a certain range [thp1, thp2] It is necessary to make it an effective area for period detection.
[0020]
Hereinafter, the operation of the first embodiment will be described.
(During prescan)
In the configuration of the first embodiment illustrated in FIG. 1, a processing unit (periodicity determination circuit) that operates during pre-scanning is surrounded by a dotted line. The pre-scan here means not only when the color scanner actually scans the document, but also includes the first use when the input image is stored in the memory and used multiple times. Before the pre-scan is completed and before the main scan is started, it is known whether the original is non-periodic / 50-line halftone dot / 100-line halftone dot / 150-line halftone dot / 200-line halftone dot. .
[0021]
(During main scan)
A one-drum type color copying machine performs a total of four scans for each plate of CMYK, and a four-drum type reproduces a color image by performing one scan. However, as described above, when all images are stored in the memory and used multiple times, there is no operation of the scanner.
[0022]
Now, the periodicity of the document has been found by the pre-scan described above. The determination result is input to the coefficient switching circuit 11, and the coefficient switching circuit 11 sets a coefficient corresponding to the cycle in the filter circuit 3. For example, the coefficient is switched as described in FIG. 7 of the above-mentioned Japanese Patent Application Laid-Open No. 1-133470 (basically, the higher the number of lines, the higher the smoothness, the larger the smoothness, and the halftone dot image. To remove moire).
[0023]
It should be noted that the periodicity determination result of this embodiment can also be applied to the image area separation parameter control of the previous application (Japanese Patent Application No. 8-84185). In addition, when it is determined that the number of lines is low and achromatic using, for example, chromatic / achromatic determination of an image in combination with other information, a process that presumes a black-and-white newspaper manuscript and actively removes the background. It is also possible to perform.
[0024]
<Example 2>
FIG. 8 shows a configuration of the second embodiment of the present invention. Since the image reproduction system (that is, from the image input device to the image output device) is the same as that of the first embodiment, the description thereof is omitted. The periodicity determination circuit according to the second embodiment includes a primary color region detection circuit 12 and a periodicity detection circuit 9.
[0025]
In the periodicity determination circuit according to the second embodiment, a case where the cycle is determined by performing DCT processing on the green signal will be described as an example. Generally, a halftone original is composed of four color inks of yellow, magenta, cyan, and black. However, in an area where two or more colors overlap, there is a possibility that the original period of the original to be obtained cannot be detected. This is because the band of the green filter is wide and the response is not limited to magenta, but other colors are taken in. Such a problem may occur in the same manner even if period detection is performed by, for example, pattern matching as described in JP-A-1-133470.
[0026]
In view of this, a primary color area circuit 12 for detecting a primary color area (area consisting only of yellow, magenta, and cyan ink) from the original is provided in the periodicity determination circuit, and an effective block only when the target block is the primary color area. And DCT coefficients are added, and finally the average is obtained. That is, for the reason described above, the region of the secondary color or more of the image is not set as a period detection target. The condition is when, for example, the following color is established for the average value of RGB (when the maximum density is 0) in the block.
[0027]
Yellow block;
Rave> TH YR &Gave> TH YG & Bave <TH YB
Magenta block;
Rave> TH YR & Gave <TH YG &Bave> TH YB
Cyan block;
Rave <TH YR &Gave> TH YG &Bave> TH YB
FIG. 9 is a process flowchart of the average DCT coefficient calculation unit according to the second embodiment. In FIG. 9, in step 202, average values Rave, Gave, and Bave of R, G, and B in the block are obtained. In step 203, it is determined whether or not the block is a primary color block (that is, a block in which the above condition is not satisfied). Since the following process is the same as that of Example 1, the description is abbreviate | omitted.
[0028]
【The invention's effect】
As described above, according to the present invention , a color area with ambiguous period information is detected by adding a simple circuit, and the area is excluded from the period detection area. Can be accurately detected.
[0029]
According to the present invention , a simple circuit detects a highlight area or a dark area where period information is ambiguous, and excludes the area from the period detection area, so that the period of a halftone document is accurately detected. It becomes possible.
[0030]
According to the present invention , a simple circuit detects a region of a secondary color or more in which halftone dots with ambiguous period information overlap and excludes the area from the period detection region. It becomes possible to detect accurately.
[Brief description of the drawings]
FIG. 1 shows a configuration of Embodiment 1 of the present invention.
FIG. 2 shows a specific example of a filter circuit.
FIG. 3 shows a configuration of a periodicity detection circuit.
FIG. 4 is a process flowchart of an average DCT coefficient calculation unit according to the first embodiment.
FIG. 5 is a diagram showing how DCT coefficients are averaged.
FIG. 6 is a diagram illustrating calculation of a periodic parameter.
FIG. 7 is a diagram illustrating line number determination.
FIG. 8 shows a configuration of Embodiment 2 of the present invention.
FIG. 9 is a process flowchart of an average DCT coefficient calculation unit according to the second embodiment.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Image input device 2 Log conversion circuit 3 Filter circuit 4 Color correction circuit 5 UCR circuit 6 Dither circuit 7 Image output device 8 Periodicity determination circuit 9 Periodicity detection circuit 10 Intermediate area detection circuit 11 Coefficient switching circuit

Claims (2)

Means for reading a document as image data;
An average value calculating means for dividing the image data into a plurality of blocks and calculating an average value of pixel values of the blocks;
Primary color area detection means for determining the block as a primary color area when an average value of pixel values of the block is within a predetermined range;
Average DCT calculating means for calculating an average DCT coefficient by calculating a DCT coefficient of a block determined to be a primary color area and dividing a value obtained by adding the DCT coefficients at the same position in each block by the number of primary color areas; ,
A period calculation means for detecting a peak position in the block by performing a predetermined calculation on the average DCT coefficient;
An image processing apparatus comprising: a filter control unit that performs a filter process using a filter coefficient corresponding to a calculation result of the cycle calculation unit. Reading the document as image data;
An average value calculating step of dividing the image data into a plurality of blocks and calculating an average value of pixel values of the blocks;
A primary color region detection step of determining the block as a primary color region when an average value of pixel values of the block is within a predetermined range;
An average DCT calculating step of calculating an average DCT coefficient by calculating a DCT coefficient of a block determined to be a primary color area and dividing a value obtained by adding the DCT coefficients at the same position of each block by the number of primary color areas; ,
A period calculation step of detecting a peak position in the block by performing a predetermined calculation on the average DCT coefficient;
And a filter control step for performing a filter process using a filter coefficient corresponding to the calculation result of the period calculation step.

Images