JPH0457273B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image signal processing device for performing appropriate image processing on an image information source containing a mixture of character images and halftone images. .

"Prior Art" An image processing apparatus that reads image information on a document and records or displays it performs an emphasis process on the read image to prevent deterioration in reproducibility of line drawing portions. As such image sharpening processing, a method using a spatial frequency filter is generally used.

FIG. 11 is an example of a spatial frequency filter.
This spatial frequency filter is applied in a cross shape to a group of pixels on the document 10 as shown in FIG. The pixel is a(i,
j-1), a(i-1, j), a(i+1, J), a
It can be expressed as (i, j+1). At this time,
The pixel of interest a(i,
Let the filter constant of j) be 1+4D (D is a constant),
If the filter constants for the four surrounding pixels are each −D, then the pixel of interest a ^* after sharpening processing
(i, j) can be calculated using the following formula.

a ^* (i, j) = (1 + 4D) a (i, j) - D (a (i, j - 1) + a (i - 1, j) + a (i
+1,j) +a(i,j+1)) ...(1) "Problem to be solved by the invention" However, since such a spatial frequency filter was originally set to sharpen character images, If there is a halftone image with smoothly changing density, such as a photograph or halftone dots, in the original,
The disadvantage is that the sharpening effect on these parts is small, and on the contrary, the noise components contained in these images are amplified. This is because, although the spatial frequencies that make up the halftone image are relatively low, the high frequency components are mainly emphasized for character images, so the noise components that exist in this high frequency region are meaningless. This was due to the amplification of the

In view of the above circumstances, an object of the present invention is to provide an image signal processing device that can appropriately sharpen not only the character image portion but also the halftone image portion.

"Means for Solving the Problem" In the present invention, as shown in principle in FIG. 1, when a pixel of interest belongs to a portion of a character image, the a first image enhancement means 11 that selectively enhances only the high spatial frequency region of the image by filter calculation using the image data D1 belonging to the area;
When this pixel of interest belongs to a portion of a halftone image, image data D2 of the peripheral pixels belong to a second region that is at least partially different from the first region.
an image signal processing device; be equipped. Here, the second region is set to a region whose outer periphery is outside the outer periphery of the first region, but these may partially overlap. The image signal processing device may include an image state determining means for determining whether the pixel of interest belongs to a character image portion or a halftone image portion, and may also include an image state setting means such as an external switch. hand,
It may be possible to set the determination result of the image state from the outside.

"Example" The present invention will be explained in detail with reference to Examples below.

FIG. 2 schematically shows the image signal processing device of this embodiment. Image sensor 2 of this device
1 scans the original 10 (see FIG. 12) in a plane;
An analog image signal 22 is output. The A/D converter 23 receives this analog image signal 22.
6-bit or 64-gradation digital image signal 2
Convert to 4. The digital image signal 24 is supplied pixel by pixel to an image data extraction section 25. The image data extraction section 25 extracts necessary image data groups 31 to 33 for the image state determination circuit 26 and the first and second image enhancement circuits 27 and 28, and supplies them thereto.

FIG. 3 shows the elements of this image data extraction section 25. The image data extraction unit 25 includes a series circuit consisting of four line memories 35 ₁ to 35 ₄ and latch circuits 36 ₁₁ to 36 ₁₅ , 36 ₂₁ to 36 ₂₄ , and 36 _{31 to} 4, respectively.
It has a series circuit consisting of 36 ₃₅ , 36 ₄₁ to 36 ₄₄ , and 36 ₅₁ to 36 ₅₅ . Here, each of the line memories ₃₅₁ to ₃₅₄ stores parallel image data of 6 bits per pixel.
It is a type of delay memory that outputs the output with a delay of the main scanning line, and each latch circuit 36 ₁₁ to 36 ₅₅ corresponds to 1 pixel 6.
This is an element that latches bit parallel image data one pixel at a time. The latch circuit ₃₆₁₁ located at the tip of the first series circuit receives the digital image signal 24.
is now being supplied. Latch circuits located at the ends of each of the second to fifth series circuits
36 ₂₁ , 36 ₃₁ , 36 ₄₁ , and 36 ₅₁ are digital image signals output from line memories 35 ₁ to 35 ₄ , respectively.
37 ₁ to 37 ₄ are now available.

Now, it is assumed that the latch outputs of the respective latch circuits 36 ₁₁ to 36 ₅₅ are represented by image data 38 ₁₁ to 38 ₅₅ , and among these, the latch output corresponding to the pixel of interest is image data 38 ₃₃
shall be. In this case, the image data group 31 supplied to the image state determination circuit 26 is composed of nine pieces of image data ₃₈₂₂ to ₃₈₄₄ shown in FIG. The image state determination circuit 26 detects the maximum value D _MAX and minimum value D _MIN of the image density in the image data group 31 in these 3×3 image areas, and compares the difference between them with a predetermined threshold l _TH . If it is larger than the threshold _lTH , the image data 38 _{to 33} of the pixel of interest is determined to be part of a character image; otherwise, it is determined to be part of a halftone image. These can be expressed as the following formula.

D _MAX −D _MIN >l _TH . . . Character image D _MAX − D _MIN ≦l _TH . . . Halftone image Here, the threshold l _TH is usually a value of about 10 to 15 for 6-bit image data.

The discrimination principle of the image state discrimination circuit 26 as described above will be explained with reference to FIG. This figure shows the signal level (density) for one line in a certain part of the document. In this figure, a signal level of "1" means that that part of the document is ideally white, and a signal level of "64" means that that part is ideally black. The signal level changes smoothly as a whole in the first half of the scan.
In these parts, the signal level changes in multiple stages, and it can be seen that these parts are halftone image areas. In the halftone image area, there is no significant difference in density between two adjacent points. Therefore, if the difference D _MAX − D _MIN in the 3×3 image data group 31 to which the pixel of interest belongs is equal to or smaller than the threshold l _TH , the image data of the pixel of interest 38 ₃₃
has a strong tendency to belong to halftone images.

On the other hand, in the latter half of the scan in FIG. 5, there is a sharply cut waveform. This portion is an edge portion of a line image such as a character, and is a typical character image area. In such a character image area, there is a high possibility that a large difference in density will occur even at two points that are close to each other. Therefore, there are differences in the 3×3 image data group 31.
If D _MAX - D _MIN is larger than the threshold l _TH , the image data 38 ₃₃ of the pixel of interest has a strong tendency to belong to a character image. The determination result of the image state determination circuit 26 is output as a determination result signal 39 and becomes a control signal for the multiplexer 41.

On the other hand, nine image data 38 ₂₂ to 38 ₄₄ similar to those shown in FIG. 4 are supplied to the first image enhancement circuit 27 as the image data group 32. The first image enhancement circuit 27 is equipped with a spatial frequency filter shown as an example in FIG. 6 corresponding to _FIG . Multiply the remaining 8 image data 38 ₂₂ ~
38 ₃₂ , 38 ₃₄ to 38 ₄₄ are multiplied by a coefficient "-0.5" to calculate their sum. Based on this calculation result, the image data 38 _{to 33} of the pixel of interest is emphasized. The output characteristics of the spatial frequency filter in this case are as shown in FIG. In other words, since the image data of pixels adjacent to the pixel of interest is used for filter calculation,
The high frequency components that make up the character image are emphasized, making these parts sharp.

On the other hand, the second image enhancement circuit 28 has 5
Nine image data 38 ₁₁ to 38 ₅₅ shown in FIG. 8 are supplied as a ×5 image data group 33. These are data groups located further outside the image data shown in FIG. The second image enhancement circuit 28
It is equipped with a spatial frequency filter having the contents shown as an example in FIG. 9, which is shown in FIG. The output of this spatial frequency filter is transmitted to the first image enhancement circuit 2.
7, the total value is obtained by multiplying each image data 38 ₁₁ to 38 ₅₅ by a coefficient. Since this second image enhancement circuit 28 uses pixels further away from the pixel of interest for filter calculation, the area of the halftone image or photographic image forming the halftone image is selectively enhanced. The output characteristics of the spatial frequency filter in this case are as shown in FIG. Since the gain in the high frequency portion is relatively reduced in this way, it is possible to compensate for the deterioration in image quality of the halftone image without amplifying the noise component.

Now, the multiplexer 41 shown in FIG.
Two types of image signals 43 and 44 output from the second image enhancement circuits 27 and 28 are input in parallel, and one of them is selected according to the content of the discrimination result signal 39. This is then sent as an image signal 45 to a subsequent recording section or display section (not shown). Here, the image signal 43 by the first image enhancement circuit 27 is selected when the image state discrimination circuit 26 has discriminated the pixel of interest as a character image, and the image signal 44 by the second image enhancement circuit 28 is selected. What is selected is the state in which the image state determination circuit 26 has determined that this pixel of interest is a halftone image. The multiplexer 41 selects the image signals 43 and 44 for each pixel.
This allows fine control of image enhancement.

In the embodiment described above, the image state determination circuit 26
However, for example, a switch for indicating the image condition is arranged in the image signal processing circuit, and the image enhancement for character images and that for halftone images are determined by the operator's judgment. You may give instructions each time. Such a switch is not limited to simply instructing a switching point on a document-by-document basis or in the sub-scanning direction, but may also be a complex instruction means that causes a specific area of the document to be processed by a specific filter.

In addition, in the embodiment, the image data of the first area and the image data of the second area are
Although the image data of the regions are made independent from each other except for the image data of the pixel of interest, it is of course possible to use image data in which these partially overlap. Furthermore, in the embodiment, image sharpening is controlled by determining the image state of each pixel, but it is also possible to extend and apply the determination result for one pixel of interest to a predetermined image area including this pixel of interest. This simplifies image processing.

"Effects of the Invention" As explained above, according to the present invention, the areas used for emphasis processing are changed between the character image part and the halftone image part, and the character image part is , selectively emphasizes only the high spatial frequency range, and for the halftone image part, selectively emphasizes only the spatial frequency range lower than the spatial frequency range for the text image part. By performing processing, it is possible to set the spatial frequency range to be emphasized to the optimum value, and it is possible to appropriately sharpen the character image part, and also to sharpen the halftone image part without emphasizing noise components. can also be sharpened appropriately. Therefore, it is possible to create a three-dimensional effect in the image, especially in the halftone image portion, and it is possible to reduce noise and improve image quality.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the basic configuration of the present invention, and FIGS. 2 to 10 are for explaining an embodiment of the present invention, of which FIG. 2 schematically shows an image signal processing device. 3 is a block diagram embodying the main parts of the image data extraction section, FIG. 4 is an explanatory diagram showing the arrangement structure of the image data group 31, and FIG.
The figure is a waveform diagram showing the change in signal level in a certain main scanning line when reading a document.
7 is a characteristic diagram of this spatial frequency filter, FIG. 8 is a block diagram showing the arrangement structure of the image data group 33, and FIG. 9 is a block diagram showing the arrangement structure of the image data group 33. FIG. 10 is a block diagram of the spatial frequency filter used in the image enhancement circuit 28, and FIG. 10 is a characteristic diagram of the latter spatial frequency filter.
FIG. 1 is a block diagram of a conventional spatial frequency filter that is uniformly used, and FIG. 12 is an explanatory diagram showing the relationship between the spatial frequency filter shown in FIG. 11 and a document. 11...First image enhancement means, 12...Second image enhancement means, 25...Image data extraction unit, 26
...image state discrimination circuit, 27...first image enhancement circuit, 28...second image enhancement circuit.

Claims (1)

[Claims] 1. When a pixel of interest as a pixel to be processed belongs to a portion of a character image composed of characters or line drawings, the first pixel of peripheral pixels located around this pixel of interest a first image enhancement means that performs a process of selectively emphasizing only a high spatial frequency region with respect to image data of a pixel of interest through a filter operation using image data belonging to the region; When the pixel belongs to a part of the halftone image other than the part, the peripheral pixels belonging to the second area whose outer periphery is outside the outer periphery of the first area are filtered using image data of the surrounding pixels. For pixel image data,
an image signal processing device comprising: second image enhancement means that selectively enhances only a spatial frequency range lower than the spatial frequency range emphasized by the first image enhancement means; . 2. The image signal processing device according to claim 1, wherein the second area is an area that does not overlap with the first area. 3. Image signal processing according to claim 1 or 2, characterized in that the image signal processing comprises image state determining means for determining whether the pixel of interest belongs to a character image portion or a halftone image portion. Device. 4. Claims 1 or 2, characterized by comprising an image state setting means that can externally set whether the pixel of interest belongs to a character image part or a halftone image part. The image signal processing device described.