JPH01168164A - Picture processing unit - Google Patents

Abstract

PURPOSE:To satisfy the resolution of a character part and the gradation of a photograph part by calculating a specific characteristic quantity to identify a character picture and a photograph picture from picture information, comparing it with a reference value so as to identify whether the picture information is a binary picture or a non-binary picture. CONSTITUTION:The unit is provided with a difference calculation means 43 detecting a maximum value and a minimum value of a change point of information relating to picture density of picture information to calculate a difference between the maximum and minimum values, a distance calculation means 54 calculating the distance of the change points on the picture causing the maximum and minimum value of the picture information change as an object of calculation by the means 43 and a slope calculation means 44, and the slope of the other value with respect to one value is calculated based on the difference and the distance. Then the slope is compared with the reference value to identify whether the picture information is a binary picture or a non-binary picture. Thus, the resolution of a character including a contrast character part with low contrast such as a blur character is preserved and the gradation of a photograph part is satisfied to apply picture processing.

Description

[Detailed Description of the Invention] [Objective of the Invention] (Industrial Application Field) The present invention is directed to a non-binarized image such as a binary image such as a character image mixed in a document image and a non-binarized image such as a photographic image. The present invention relates to an image processing device that processes images so as to simultaneously satisfy the resolution of characters and the gradation of photographs.

(Prior Art) An image processing device capable of image processing not only characters but also literary images in which gray scale images such as photographs etc. are mixed with characters can process characters, line diagrams, etc. read by a reading device such as a scanner. Image information with contrast is simply binarized using fixed equivalence, and image information with gradations such as photographs is binarized using a pseudo gradation means such as a dither method.

More specifically, in this image processing device, when the read image information is simply binarized using a fixed mental position, the resolution is preserved in the text and line image areas, so no image quality deterioration occurs, but in the photo image area, the image quality is not degraded. Because the gradation level is not preserved, the resulting image will have degraded image quality.Also, if the read image information is gradated using systematic dithering, etc., the gradation level will be preserved in the photographic image area, so the image quality will not deteriorate. However, in the area of characters and line images, the resolution decreases, so in order to prevent the image quality from deteriorating, two types of binarization are used: simple binarization using fixed equivalence and binarization using pseudo gradation. It uses a binarization method.

That is, by performing a single binarization process on read image information, it is impossible to simultaneously satisfy the image quality of each region having different characteristics, such as a character image and a photographic image. This also applies to various types of image processing; for example, if processing is not performed that matches the characteristics of the image, the image quality may deteriorate when enlarging or reducing a binarized image, or when encoding processing does not match the characteristics of the image. If processing is not performed using a compression method that is unique, the data will be compressed inefficiently. Therefore, it is essential to divide image information into regions according to the characteristics of the image and to perform adaptive processing on each region.

By the way, conventionally, as a method that simultaneously satisfies the resolution of the character part and the gradation of the photograph part, the image 8 is created in a local area within the image plane.
Determine the maximum density difference ΔlaX of 1 degree. This method compares this maximum density difference Δll1aX with a determination threshold Th, divides the image into text and line image areas and photographic image areas, and switches the binarization process according to the characteristics of each image area. has been proposed (for example, JP-A-58-3374) (Problems to be Solved by the Invention) In the conventional image processing method described above, low-contrast characters such as blurred character parts such as pencil wandering, etc. Since the maximum density difference in the area is small, it is determined to be a photographic area, and there is a problem in that the resolution of the text area is significantly impaired.

More specifically, as shown in FIG. 8, the document P has areas A of contrasting characters and line images, area B of photographic images with gentle density changes, and low-contrast character areas such as faded characters. In the case where the document P is composed of a region C, a typical signal level of image information in each region of the document P is as shown in FIG. Now, the dynamic range of image density is set to 8 bits (0 to 255.0 to 255 in hexadecimal).
ff) L, T, for each of the image areas A, B, and C, respectively, within a predetermined range within the image plane, for example, within a range of 16 pixels consisting of a 4 x 4 matrix, la to the maximum density difference.
When determining X, the maximum density difference ΔDmax in the region
= dd ~ "f (hex), Kel ΔD in 3 areas
max −10 to 40 (hex) and Δ[)wax −10 to 40 (MX) in the C region.

Therefore, the determination threshold Th for distinguishing between text and photographic areas is set to 8.
Assuming 0 (heX), the following judgment condition; ΔDmax
>Th...Text area △Dfflax≦Th...When each image area is identified by the photo area, area A is identified as the text area, B, C
The region is identified as a photographic section.

As a result, simple binarization processing using a fixed threshold is performed in the A area, and dither processing is performed in the B and C areas. In the area C, character resolution is significantly impaired because the characters are erroneously identified and processing is performed to preserve the gradation level. In addition, in order to accurately identify the CwA region,
If the determination threshold value is set to a small value such as 30 (hex) or 10 (heX), 8 areas will be identified as text parts, resulting in a photographic image in which the gradation level will not be preserved. In other words, in the conventional method, it is not possible to divide the A area of text and line images, the 8 areas of the photo image, and the C area of the low contrast text area into accurate image areas, so the resolution of the text area and the gradation level of the photo area cannot be divided into accurate image areas. There is a problem in that it is not possible to adaptively perform image processing according to the image characteristics for each region so as to simultaneously satisfy the following.

The present invention has been made in view of the above, and its main purpose is to provide image information in which a text portion and a photo portion are mixed so as to simultaneously satisfy the resolution of the text portion and the gradation level of the photo portion. Another object of the present invention is to provide an image processing device that can adaptively perform image processing according to image characteristics for each image region.

[Structure of the Invention] (Means for Solving the Problems) In order to solve the above problems, the image processing device of the present invention includes the following:
Difference calculating means for detecting local maximum values and local minimum values of changing points of information related to image density of image information and calculating a difference between the local maximum values and local minimum values, and the image information to be calculated by this difference calculating means. a distance calculation means for calculating a distance on an image between change points that generate a maximum value and a minimum value of change; and a slope value calculation means for calculating a slope value of one value with respect to the other value in the difference and the distance. , a comparison identification means for comparing the slope value calculated by the slope value calculation means with a predetermined reference value and identifying whether the image information is a binary image or a non-binary image. shall be.

(Function) The image processing device of the present invention calculates the difference between the maximum value and minimum value of the change point of information related to image density and the distance between the change points that generate the maximum value and the minimum value, and calculates the difference between the change points that generate the maximum value and the minimum value. The slope value of one value with respect to the other value is calculated in the distance and distance, and this slope value is compared with a reference value to identify whether the image information is a binary image or a non-binary image.

(Example) Hereinafter, the present invention will be explained in detail using the drawings.

FIG. 1 is a block diagram showing the overall configuration of an image processing apparatus according to an embodiment of the present invention. The image processing device shown in the same figure is a scanner that is a reading means (not shown).
An image signal 11 of 8 bits per pixel is temporarily stored in a line buffer 1, and the image signal stored in this line buffer 1 is inputted to an identification circuit 2 in synchronization with a clock signal (not shown), and is composed of a delay memory and the like. Delay circuit 10
The signal is supplied to the comparator circuit 9 via. The identification circuit 2 performs pixel-by-pixel identification on the image signal and supplies an identification signal 12, which is the identification result, to the selection circuit 8 as a control signal.

The selection circuit 8 receives a first threshold signal 15a and a second threshold signal 15b from the first threshold generation circuit 5 and the second threshold generation circuit 6, respectively, and selects the first threshold value based on the identification signal 12. The optimum threshold signal is selected from the second threshold signals 15a and 15b and outputted to the comparison circuit 9 as the selected threshold signal 16. The comparison circuit 9 compares this selection threshold signal 16 with the image signal 17 of the pixel through the delay circuit 10, and outputs an output image signal 18 of "1" when the image signal level of the pixel is higher than the selection threshold signal 16. is output, and “0” when the image signal level of the pixel is less than the selection threshold value @16.
An output image signal 18 is output.

Next, before explaining the identification circuit 2 in detail, the window area referred to in the identification circuit 2 will first be explained with reference to FIG. In FIG. 5, the 4×4 area shown by broken lines is referred to to identify the pixels shown by diagonal lines. Further, in order to smooth the 4×4 area indicated by the broken line and emphasize edges, reference is made to the 8× pixels shown in the figure.

FIG. 2 is a detailed circuit block diagram of the image processing apparatus shown in FIG. 1. In the figure, an identification circuit 2 is a circuit that identifies whether an image signal is a text portion or a non-text portion, that is, a photo portion, and the image signal from the line buffer 1 is supplied to a smoothing circuit 40. Noise components of the image are removed. The smoothing circuit 40 removes noise components and performs the smoothing circuit 40 as shown in FIG.
As shown in a), the smoothed image signal is supplied to an edge detection circuit 51 and an edge emphasis circuit 41. The edge detection circuit 51 uses a Laplacian filter, extracts the edge portion of the smoothed image signal supplied from the smoothing circuit 40, and outputs an edge image signal as shown in FIG. 6(b). This edge image signal is supplied to an edge position detection circuit 52 and an edge emphasis circuit 41. The edge position detection circuit 52 detects the edge position from the edge image signal and supplies the edge position signal to the inter-edge distance calculation circuit 54 and the subtraction circuit 43. This inter-edge distance calculation circuit 5
4 outputs an inter-edge distance signal S (L) based on the edge position signal from the edge position detection circuit 52. This inter-edge distance signal S(·L) corresponds to the inter-edge distance signal shown in FIG. 6(b), and is supplied to the division circuit 44.

On the other hand, the edge emphasis circuit 41 outputs an edge emphasis image signal based on the smoothed image signal supplied from the smoothing circuit 40 and the edge image signal supplied from the edge detection circuit 51, and supplies it to the subtraction circuit 43. The subtraction circuit 43 calculates the density difference ΔD of image information at the edge position from the edge emphasis image signal of the edge emphasis circuit 41 based on the edge position signal supplied from the edge position detection circuit 52, and divides this density difference ΔD into the division circuit 44. supply to.

The barrier removal circuit 44 uses the edge distance signal @5(L) from the edge distance calculation circuit 54 as shown in the following equation to subtract circuit 43.
A normalized image feature signal F1, that is, a slope value, is calculated by dividing the density difference ΔD from .

F=△D/L (1) The image feature signal F from the division circuit 44 is supplied to a comparison circuit 45, and is compared with a determination threshold Th stored in advance in a register, etc., not shown. be done.

As a result of this comparison, the comparison circuit 45 determines that the image signal is a character pixel and outputs a "1" level identification signal 12 when the image feature equality signal F is larger than the determination equality value Th as shown below. Further, when the image characteristic signal F is less than the determination threshold Th, it is determined that the pixel is a non-character pixel, that is, a photographic pixel, and an identification signal 12 of the rOJ level is output.

F>Th Character pixel F≦Th Non-character pixel Next, the binarization processing of each pixel identified as a character pixel or a photographic pixel by the identification circuit 2 as described above will be described.

The image signal supplied from the line buffer 1 is supplied to the maximum value/minimum value detection circuit 42, and the maximum/minimum value detection circuit 42 selects the maximum image density from among the image densities within the window 4X4=16 pixels. Detect DIlla× and minimum image density Q win, and set this maximum image density D ma
x and minimum image density [ ] 1n is dynamic threshold detection circuit 3
3. The dynamic threshold calculation circuit 33 calculates the binarization threshold Bh from the maximum pixel density DIIa× in the window detected by the maximum value minimum value detection circuit 42 and the minimum image density D min according to the following equation.

Bh = (Diax + Dwin)/2- (2)
This binary threshold Bh is supplied to the first threshold generation circuit 5, and the first threshold generation circuit 5 outputs the first threshold signal 15a to the selection circuit 8 based on this binary threshold Bh. The second threshold generation circuit 6 generates a binarization threshold for a non-text portion, that is, a photographic portion, and outputs a second threshold signal 15b to the selection circuit 8 based on the dither threshold shown in FIG. do.

The selection circuit 8 uses the identification signal 12 as a control signal to select the first
A selection threshold signal 16 is determined from the second threshold signals 15a and 15b under the following conditions.

11 1 and 14 (i116 1, identification signal 12="1" When first threshold signal 15 (
a (binarization threshold ah) 2, identification signal 12 = "On", second threshold signal 15 (when distinguishing from a photograph) b (threshold for photograph) selection circuit 8 is configured as described above. The determined selection threshold signal 16 is compared with the image signal 17 of the pixel supplied via the delay circuit 10, and an output image signal 18 is output. It is now possible to perform image processing that satisfies the gradation level of the photographic area while preserving the resolution of the image.

Next, a more detailed circuit of the identification circuit 2 will be explained.

FIG. 3 is a detailed circuit diagram of the identification circuit 2, and FIG. 4 is its timing diagram. In FIG. 3, the smoothing circuit 201
206 constitute the smoothing circuit 40, and edge detection circuits 211-214 constitute the edge detection circuit 51. In FIG. 4, CLK is a reading block and D
IN indicates the timing at which image data is input from the line buffer 1 to the smoothing circuits 201 to 206. Further, the smoothing performed here is performed with reference to pixels in a 3×3 block centered on the reference pixel.

First, the first clock LJ-3) is shown in FIG.
This is the timing at which eight pixels from row (I-3) to row (1+4) of column ) are supplied to the smoothing circuits 201 to 206. Three pixels from rows (1-3) to (1-1) are input to the smoothing circuit 201. The smoothing circuit 202 has lines from (1-2) to (1
) input the 3 pixels of the row. Similarly, three pixels are input for each of the other smoothing circuits. Subsequently, when the 8 pixels in columns (J-2) and Ll-1) shown in FIG.
+3) Smoothing is performed on the rows.

That is, at the third clock, the smoothing circuit 201 outputs Ll-2.
The pixels in the (1-2) row of the column (1-2) are smoothed, and the smoothing circuit 202 smoothes the pixels in the (I-1) row of the (J-2) column. The same applies to the smoothing circuits 203 to 206. Similarly, when column (J) is input at the fourth clock, smoothing is performed for rows (1-2) to (I+3) of column (J-1).

SLIM is the timing at which the image data smoothed by the smoothing circuits 201 to 206 is supplied to the edge detection circuits 211 to 214. Here, edge detection is performed with reference to pixels in a 3×3 block centered on a reference pixel. (J-2) of the fourth clock is L shown in Figure 5.
The smoothed 6-pixel image data from rows (1-2) to rows (1+3) of column J-2) is sent to edge detection circuits 211 to 21.
This is the timing when the signal is supplied to 4. Edge detection circuit 21
1, the smoothed three-pixel image data from row (I-2) to row (1) of column (J-2) is input. The edge detection circuit 212 has Ll-2) column (1-1> row to (I+1)
) provides image data for three pixels in a row. Similarly, three pixels are input to each of the other edge detection circuits. Subsequently, when the 6 pixels in columns (J-1) and (J) shown in FIG. 5 are input at the fifth and first clocks, ('J-1) column ('
Edge detection is performed for the (I+2) rows from the 1-1> row. That is, the edge detection circuit 211 performs edge detection for the pixels in the (T-1) row, for example (J-1). The edge detection circuit 212 performs edge detection for pixels in column LJ-1) and row (I). The same applies to other edge detection circuits. Similarly, at the 7th clock (J+
When 1> column is input, from row (1-1) of column LJ) to (1
+2) Detect row edges.

EDGl is the timing at which image data whose edges have been detected by the edge detection circuit 51 is supplied to the edge position detection circuit 52 and the edge emphasis circuit 41. L of the 7th clock
1-1> is the image data of the four edge-detected pixels in column (J-1) and rows (1-1) to (1+2) shown in FIG. This is the timing of the supply. Here, edge emphasis circuit 4
1 determines the position of the edge portion of the image from the image data whose edges have been detected by the edge detection circuit 51 for the four pixels in column (J-1) shown in FIG. Then at the 8th clock (J
) column, when the data of 4 pixels from row (1-1> to row (I+2)) is input, the position of the edge part is detected for the 4 pixels of column (J).The same goes for the 9th clock onward.On the other hand,
In the edge enhancement circuit 41, at the seventh clock (J-1> column, edge enhancement of four pixels from row (1-1) to row (1+2) is performed by subtracting the edge detection image from the original image.Next, , when 4 pixel data from column (J) and row (r-1) to (1+2> rows) is input at the 8th clock, LJ)
Edge enhancement is performed for four pixels in a column. Edge enhancement is performed in the same manner from the 9th clock onwards.

EDG2 is the timing at which the edge position signal detected by the edge position detection circuit 52 and the image data edge emphasized by the edge emphasis circuit 41 are supplied to the inter-edge distance calculation circuit 54 and the subtraction circuit 43. At (J) of the 11th clock, it is within the 4×4 window from the (1-1) row of the (J-1) column to the (1+2> row of the (J+2) column shown in FIG. 5, that is, the broken line in FIG. 5. The edge position signal and edge-enhanced image data within are supplied to a subtraction circuit 43.

At the 12th clock, from (I-1) row of (J) column to (J+
3) The edge position signal and edge-enhanced image data within the 4×4 window of column (T+2>row) are supplied to the subtraction circuit 43. The same applies to the 13th clock and subsequent clocks.

DIV is the timing at which the inter-edge distance calculated by the inter-edge distance calculation circuit 54 and the density difference ΔD detected by the subtraction circuit 43 are supplied to the division circuit 44. At (J) of the 12th clock, the image feature amount signal F of column (1) of column LJ shown in FIG. 5, that is, the shaded area in the figure is calculated. Further, in the FTR, the division result from the division circuit 44 is sent to the comparison circuit 45.
This is the timing at which the signal is supplied.

In the above embodiment, the reference image range is described as a 4×4 predetermined image area, but the present invention is not limited to this, and it is possible to take an image area of any arbitrary range. can. Further, in this embodiment, an example of identification at a rate of one pixel has been shown, but identification may be performed in units of (NXN) blocks (provided that N≧2 is a positive integer). Further, the binarization threshold Bh in the dynamic threshold processing is not limited to the above-mentioned threshold, but any arbitrary value, for example, the average image density [)a within a predetermined range, can also be used. Furthermore, a positive or negative tolerance value α may be added to each value, for example, the binarization threshold ah may be set to ah+α. The dither threshold value may also be freely set to an equivalent arrangement other than the dot-dispersed type, for example, a dot-concentrated threshold value arrangement.

Furthermore, in this embodiment, the image signal read by the reading means, that is, the image information, is used as a special quantity value and a determination threshold for identifying character images and photographic images, that is, binary images and non-binary images, from image information. However, it is also possible to identify the amount based on the image density, that is, the logarithm of the reciprocal of the reflectance, or a converted signal that takes into account human visual characteristics. good.

[Effects of the Invention] As explained above, according to the present invention, the difference between the maximum value and the minimum value of the change point of information related to image density and the distance between the change points that generate the maximum value and the minimum value , calculate the slope value of one value with respect to the other value in this difference and distance, and compare this slope value with a reference value to determine whether the image information is a binary image or a non-binary image. Because this technology identifies low-contrast characters, such as blurred pencil writing, which would otherwise be judged as a photographic area and have significantly lost resolution, it is now possible to perform appropriate image processing while preserving the resolution. It is possible to improve image quality and processing efficiency in various image processing.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an image processing apparatus according to an embodiment of the present invention, FIG. 2 is a detailed block diagram of the image processing apparatus of FIG. 1, and FIG. 3 is a block diagram of an image processing apparatus used in the image processing apparatus of FIG. A detailed block diagram of the identification circuit; FIG. 4 is a timing diagram of the circuit in FIG. 3, FIG. 5 is a diagram showing a window area referred to by the identification circuit, and FIG. 6 is a diagram regarding an edge position detection image for an original image. , FIG. 7 is a diagram showing an example of the dither threshold value, FIG. 8 is a diagram showing each image area, and FIG. 9 is a schematic diagram showing the image signal level in each image area. 2...Identification circuit 5...First threshold generation circuit 6...Second equivalent value generation circuit 8...Selection circuit 9...Comparison circuit 40...Smoothing circuit 41...Edge Emphasis circuit 43... Subtraction circuit 44... Division circuit 45... Comparison circuit 51... Edge detection circuit 52... Edge position detection circuit 54... Edge distance calculation circuit

Claims (3)

[Claims] (1) A difference calculation means that detects the maximum value and minimum value of the change point of information related to the image density of image information and calculates the difference between the maximum value and the minimum value; distance calculating means for calculating the distance on the image between the change points that generate the local maximum value and the local minimum value of the image information change;
a slope value calculation means for calculating a slope value of one value with respect to the other value in the difference and the distance; and comparing the slope value calculated by the slope value calculation means with a predetermined reference value;
An image processing apparatus comprising: a comparison identification means for identifying whether the image information is a binary image or a non-binary image. (2) The difference calculating means includes a smoothing means for smoothing image information within a predetermined range, an edge emphasizing means for emphasizing edges of the smoothed image signal from the smoothing means, and a predetermined range in the edge emphasizing means. and means for determining the maximum density difference of edge-enhanced image signals corresponding to pixels whose Laplacian value, which is an image signal of an edge portion extracted in the image signal, takes a maximum value and a minimum value. The image processing device according to item 1. (3) The distance calculating means includes a smoothing means for smoothing image information within a predetermined range, an edge emphasizing means for emphasizing edges of the smoothed image signal from the smoothing means, and a predetermined range in the edge emphasizing means. 2. The image processing apparatus according to claim 1, further comprising means for determining a distance between pixels at which a Laplacian value, which is an image signal of an edge portion extracted in the image signal, takes a maximum value and a minimum value.