JP2003037734A - Image processing unit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing unit that can output an image of higher image quality by using a simple means to conduct dot image area separation processing, with high accuracy. SOLUTION: A window with a particular pixel number for storing a target pixel to discriminate whether or not it is a part of a dot image area sequentially stores image data, a pattern detection means sequentially detects whether or not this window is formed with a particular pattern unique to the dot image area, and sequentially detects whether or not the target pixel in the window is part of the dot image area, on the basis of the result of detection. By detecting it over the entire image data, the dotted image area can be detected.

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 detecting a halftone dot image area from digital image data and filtering the halftone dot image area to output a high quality image.

[0002]

2. Description of the Related Art An image processing apparatus for performing desired image processing on image data generated by reading an original and outputting the image data is an image input apparatus such as a scanner or an image output apparatus such as a printer. There are various modes such as an image transmitting / receiving apparatus via a network such as a facsimile, or an image forming apparatus combining these.

When an image is output using these devices as described above, the image quality greatly changes due to the difference in the characteristic peculiar to the device, and the image quality of the output image obtained also greatly changes due to the characteristic of the original document to be read. To do. In particular, in order to improve the image quality that changes depending on the characteristics of the document itself to be read, the image data input from the image input device is separated into image areas such as photographs, characters, and halftone dots, and an image suitable for that area is separated. An apparatus that obtains a certain image quality by performing processing has been proposed. For example, an image area separation apparatus disclosed in Japanese Patent Laid-Open No. 05-048891 is related to this apparatus.

This device discloses a means for separating each image area on the premise that a photograph and a halftone dot do not exist on the same original document. The first feature quantity extraction means and the first identification processing means take charge of extracting and distinguishing and separating the halftone dot area and the character area based on the extracted feature quantity, and the distinguishing feature quantity of each of the photograph and the character. The second feature quantity extraction means and the second identification processing means are responsible for identifying and separating the photograph area and the character area based on the extracted feature quantity. Then, one of the above is selected according to the document to be read.

According to this proposal, the image area is separated according to the feature quantity.
The maximum density, the minimum density, the maximum density difference, the minimum density difference, and the number of pulses are calculated from the image data to be identified, compared using many parameters, and the separation processing is performed based on the comparison result.

[0006]

However, in the proposed method, many parameters (threshold values) must be prepared in advance in order to extract the feature quantity. Cause separation, and as a result,
It may deteriorate the image quality. Therefore, the parameters to be preset should not be used or should be minimized even if they are used.

Therefore, an object of the present invention is to provide an image processing apparatus capable of outputting an image of higher image quality by performing highly accurate halftone image area separation processing by a simple means.

[0008]

In order to achieve the above object, the invention of claim 1 has a specific pixel number for storing a pixel of interest for determining whether or not it is a part of a halftone dot image area. , Forming means for forming a window in which a predetermined number of image data are stored by shifting by 1 pixel data so that all input image data are all target pixels, and a window formed by the forming means is a mesh. Pattern detection means for gradually detecting whether or not the pixel is formed of a specific pattern specific to the dot image area, and whether or not the pixel of interest is part of the halftone dot image area based on the detection result of the pattern detection means. An image processing apparatus comprising: a detection unit that makes a gradual determination and detects a halftone image region based on the determination result.

According to a second aspect of the present invention, in the first aspect of the present invention, the pattern detecting means detects peak pixels by comparing adjacent pixels of each pixel in the window formed by the forming means with each other. Means, encoding processing means for binarizing each pixel in the window with a predetermined threshold value capable of extracting the peak pixel obtained by the peak detecting means, and binarization by the encoding processing means. The image processing apparatus is characterized in that it has a comparison means for comparing and collating the created window with a sample window made up of sampled binary data of a halftone dot image area.

According to a third aspect of the present invention, in the second aspect, the peak detecting means subtracts or adds a predetermined number of pixel values of one of the adjacent pixels, and adds or subtracts one of the pixel values. An image processing apparatus is characterized in that a peak pixel is detected by comparing the pixel with another pixel which is not added or subtracted adjacent thereto.

According to a fourth aspect of the present invention, in the second or third aspect, the copying means for copying the pixel value of the peak pixel detected by the peak detecting means to a pixel adjacent to the peak pixel to generate a copying peak pixel. In addition, the encoding processing means sets each pixel in the window including the peak pixel copied by the copying means with a predetermined threshold value capable of extracting the peak pixel and the copied peak pixel. An image processing apparatus characterized by binarization.

According to a fifth aspect of the present invention, in any one of the second to fourth aspects, the peak detecting means compares the adjacent pixels with each other and is larger than the pixel value of the adjacent pixel, or of the adjacent pixel. An image processing apparatus is characterized in that it is determined to be a peak pixel when the pixel value is smaller than the pixel value.

According to a sixth aspect of the present invention, in any one of the second to fifth aspects, the comparison means is arranged so that the comparison is made within the window.
The binarized data and the sampled binarized data in the sample window are AND-operated with respect to the respective pixel numbers, and the arithmetic result and the sample window are compared to determine whether or not they are equivalent. An image processing apparatus characterized by collation.

[0014]

DETAILED DESCRIPTION OF THE INVENTION Referring to the accompanying drawings,
Embodiments of the present invention will be described in detail. The image processing apparatus according to the present embodiment is suitable for detecting a halftone dot image area from digital image data and filtering the halftone dot image area to output a high quality image. The detection of the image area will be described in detail according to the flow of the flowchart of FIG. 1 with reference to FIGS. 2 to 8.
In the present embodiment, a monochrome 256 gradation image is exemplified, but a color image or the like may be applied.
In this embodiment, the logic of black and white is white: 0, black: 256.
Will be treated as

First, an original including a halftone dot area is read by a scanner composed of a photoelectric conversion mechanism such as a CCD sensor to input an image (step S1), and then, FIGS.
As shown in (1), from the image data for one line in the main scanning direction read by the forming means, 1 × 21 pieces of image data centered on the pixel of interest arranged at the pixel number 10 are acquired to obtain the window 1 Are formed (step S2). Although a 1 × 21 window is used here, other sizes may be used.

Next, the peak is detected from the 1 × 21 window 1. By the peak detection means, the pixel value of one pixel adjacent between the pixels arranged in the 1 × 21 window 1 is set to an arbitrary predetermined value (TH1:
In the present embodiment, it is set to 10) (step S3A). Here, by setting TH1 to a value that matches the characteristics of the input device, noise and the like can be dealt with, and more effective peak detection can be performed. In order to facilitate the description, a window 1a showing a state in which pixels in the window are collectively subtracted as shown in FIG. 3B will be described.

If one of the subtracted pixels is compared with the other adjacent non-subtracted pixel and it is determined that one pixel is larger than the other pixel, one image before the subtraction is set as a peak pixel. And That is, in FIG. 3A, when it is determined whether or not the pixel (pixel value: 128) arranged in the pixel number 8 is the peak pixel, the pixel arranged in the pixel number 8 is subtracted by 10. (See FIG. 3 (2)),
Pixel arranged at this pixel number 8 (subtracted to pixel value: 118) Pixel arranged at non-subtracted pixel number 7 adjacent to both sides of the pixel (pixel value: 85) and arranged at pixel number 9 The pixel (pixel value 84) is compared with the pixel (pixel value: 118) arranged in the subtracted pixel number 8. in this case,
Since the pixel arranged in pixel number 8 is larger than the pixels arranged in pixel number 7 and pixel number 9, the pixel arranged in pixel number 8 is determined to be the peak pixel.

By the above means, the peak pixels from the pixel number 0 to the pixel number 20 are detected. In addition,
Pixels at both ends of window 1 (pixel numbers "0" and "2"
0 ") may always be a peak or the opposite (not a peak). Further, the pixel numbers 0 and 20 located at both ends of the window may be compared and compared with one pixel. In the present embodiment, it is assumed that the pixels located at both ends are not peaks, and the peak pixels are detected by the above means for all the pixels in the window 1 of FIG. 3 (1). The state is schematically illustrated as a binarized state by giving "0" to the pixels other than the peak pixel (step S4).

When the peak pixel in the window 1 is detected in this way, as shown in FIG. 4, the pixel value of the peak pixel detected by the copying means is copied to a pixel adjacent to the peak pixel and copied. The peak pixel is generated and the peak pixel is expanded. Note that FIG. 4 corresponds to FIG.
An example is shown based on what is schematically illustrated in the binarized state of (3) (step S5).

Next, the encoding processing means binarizes the window having the detected peak pixel and the copied peak pixel with a predetermined (dynamic) threshold value with which these pixels can be extracted. (Step S6: see FIG. 5 (1)).
The comparing means compares the binarized window 1b with the reference peak pattern 2 (sample window: see FIG. 5 (2)) consisting of sampled binarized data of the halftone dot image area with respect to each pixel number. Each logical product operation is performed (step S7), and the operation result (see FIG. 5C) and the sample window are compared and collated (step S8). The reference peak pattern 2 described above is a pattern based on binary data that matches the number of lines of the halftone dots to be detected.

As a result of comparison and collation, the binarized window 1b and the reference peak pattern 2 all match, and it is determined that the pixel of interest of pixel number 10 is a partial pixel of the halftone dot image area ( Step S9). If they do not match, it is determined that they are not part of the halftone dot image area. A window is formed by shifting the above steps by one pixel data so that all the image data of one line becomes the target pixel, and it is gradually determined whether or not it is a partial pixel of the halftone dot image area. Further, the entire area of the halftone dot image area is grasped by gradually judging for each line whether or not it is a part of the pixels of the halftone dot image area over the entire sub-scanning direction. Then, a high-quality halftone dot image is output by performing optimum filter processing on the grasped halftone dot image area.

The copying means of this embodiment copies one pixel on the left and right of the peak-detected pixel, but may be two pixels on the left and right, one pixel on one side, etc. depending on the characteristics of the input device. Then, FIG. 6 exemplifies a case where copying is performed to only one pixel on the right with respect to FIG.

Further, FIG. 7 exemplifies the flow until the comparison calculation with the reference peak pattern 2 is performed with respect to FIG. 6, and the detected peak pixel and the copied peak pixel copied to the right one pixel are included. The encoding processing unit binarizes the inside of the window 1c with a predetermined threshold value with which these pixels can be extracted (see FIG. 7 (1)), and obtains a logical product operation result in the same manner as above (FIG. 7). 7 (3)), this result is compared with the reference peak pattern to determine whether it is a halftone dot image area.

In this case, since the result of the logical product operation and the reference peak pattern 2 do not match at the pixel number 7, it is determined that the pixel of interest (pixel number 10) is not a part of the halftone dot image area. As described above, the result may vary depending on the characteristics of the input device, and therefore, in the logical product operation and comparison with the reference peak pattern 2, not only one pattern but a plurality of patterns should be used to perform the collation. Therefore, it is preferable to detect a halftone dot image area corresponding to a larger number of lines.

As shown in FIG. 8, as another mode of peak detection, the pixel value of one of the pixels adjacent to each other arranged in the 1 × 21 window is predetermined. Addition is performed with an arbitrary value (TH1 = 1) (FIG. 1, step S3B), and the one pixel thus added is compared with the other non-added pixel adjacent thereto, and one pixel is more than the other pixel. When it is determined that is also small, one pixel before addition may be set as the peak pixel.

[0026]

As described above, the present invention has the following advantageous effects. According to the present invention, the maximum density, the minimum density, the maximum density difference, the minimum density difference, and the number of pulses are calculated from the image data to be identified and compared using many parameters (threshold values), and based on the comparison result. Since the window from which the image data was acquired is formed with a specific pattern peculiar to the halftone dot image area gradually without performing the separation processing, it is necessary to perform the highly accurate halftone image area separation processing. Therefore, it is possible to provide an image processing apparatus capable of outputting an image with higher image quality.

[Brief description of drawings]

FIG. 1 is a flowchart showing a processing procedure of an image processing apparatus according to this embodiment.

FIG. 2 is an explanatory diagram that schematically illustrates the flow of acquiring image data and forming a window.

FIG. 3 is an explanatory diagram schematically illustrating a flow from a window in which image data is acquired to a detection of a peak pixel.

FIG. 4 is an explanatory diagram schematically illustrating a state in which a peak pixel detected subsequently to FIG. 3 is copied to an adjacent pixel.

FIG. 5 is an explanatory diagram schematically illustrating the flow from FIG. 4 to performing comparison calculation with a reference peak pattern from a binarized window.

FIG. 6 is an explanatory diagram schematically illustrating another mode in which the detected peak pixel is copied to an adjacent pixel.

FIG. 7 is an explanatory view schematically illustrating another aspect of the flow from the binarized window to the comparison calculation with the reference peak pattern.

FIG. 8 is an explanatory view schematically illustrating another aspect of the flow from detection of a peak pixel to binarization from a window in which image data is acquired.

[Explanation of symbols]

1 window 2 Reference peak pattern (sample window)

Claims (6)

[Claims] 1. One pixel data having a specific pixel number for storing a target pixel for determining whether or not it is a part of a halftone dot image area, and all input image data are all target pixels. Forming means for forming a window in which a predetermined number of image data are stored by shifting each minute, and gradually detecting whether or not the window formed by the forming means is formed in a specific pattern specific to a halftone dot image area. Pattern detection means, and a detection for gradually determining whether or not the pixel of interest is part of a halftone dot image area based on the detection result of the pattern detection means, and detecting the halftone dot image area based on the determination result An image processing apparatus comprising: 2. The pattern detecting means obtains by the peak detecting means, wherein each pixel in the window formed by the forming means compares the adjacent pixels to detect a peak pixel, and the peak detecting means. And a coding processing means for binarizing each pixel in the window with a predetermined threshold capable of extracting the peak pixel, and a window and halftone image area sample binarized by the coding processing means. The image processing apparatus according to claim 1, further comprising: a comparison unit configured to compare and collate with a sample window formed of binarized data. 3. The peak detecting means subtracts or adds a predetermined number of pixel values of one of the adjacent pixels, and adds or subtracts one of the added or subtracted pixels adjacent to the pixel. The image processing apparatus according to claim 2, wherein the peak pixel is detected by comparing the other non-existing pixel. 4. A copy unit for copying a pixel value of a peak pixel detected by the peak detection unit to a pixel adjacent to the peak pixel to generate a copied peak pixel, wherein the encoding processing unit includes: 3. Each pixel in the window including the peak pixel copied by the copying means is binarized with a predetermined threshold value capable of extracting the peak pixel and the copied peak pixel. Alternatively, the image processing device according to item 3. 5. The peak detection means compares adjacent pixels and determines that the pixel is a peak pixel when the pixel value is larger than the pixel value of the adjacent pixel or smaller than the pixel value of the adjacent pixel. The image processing device according to any one of claims 2 to 4. 6. The comparing means is arranged so that 2 in the window is displayed.
The binarized data and the sampled binarized data in the sample window are AND-operated with respect to the respective pixel numbers, and the arithmetic result and the sample window are compared to determine whether or not they are equivalent. The image processing apparatus according to any one of claims 2 to 5, wherein collation is performed.