JPH0348980A - Contour emphasizing process system - Google Patents

Abstract

PURPOSE:To prevent the emphasis of the noise component and to obtain a sharp picture by adding the picture data obtained by applying the contour emphasis to the difference between an original picture of the coordinates in a picture and a gradated picture with a function of a specific formula to the original picture. CONSTITUTION:An input picture is temporarily stored in a specific area of a picture memory 1a and sent to a space filter processing part 1b. Then a difference (z) between an original picture F and a gradated picture (f) is calculated via a space filter. The contents of the difference (z) are sent again to another specific area of the memory 1a and sent to a density conversion processing part 1c. The part 1c carries out the density conversion according to an equation I. For the conversion function, the density conversion is reduced for the picture having a low noise level and a small difference and then emphasized for the picture like a normal contour picture, etc., having a large difference. The conversion data is stored again in the memory 1a, and the resulting picture and the original picture are read out and sent to an arithmetic part 1d to be added together. These added pictures are outputted to an output device.

Description

[Detailed description of the invention] [Required] Regarding a contour enhancement processing method for emphasizing the contours in an image to obtain a sharp image, an object of the present invention is to perform contour enhancement processing that suppresses the influence of noise in the image. Then, for the difference between the original image and the blurred image, a function that reduces the edge enhancement of pixels with a small difference value, for example, an exponential function that is symmetrical about the origin and convex downward in the first quadrant. The configuration is such that the density conversion of pixel values for the cough difference is performed using the image data, and the result is added to the original image.

[Industrial Application Field] The present invention relates to a contour enhancement processing method for enhancing contours in an image to obtain a sharp image.

Images transmitted through image input devices such as television cameras and scanners, and image output devices such as cathode ray tube (CRT) displays and printers become somewhat blurred images due to the characteristics of each device.

Therefore, in the printing industry of newspapers, magazines, etc., a process called edge enhancement is used to correct the blurring and obtain a visually sharp image.

This conventional contour enhancement processing has a problem in that not only the contour in the image but also the noise in the image is emphasized, and an effective contour enhancement method has been awaited.

[Prior art and problems to be solved by the invention] Fig. 3 is a diagram explaining a conventional contour enhancement processing method, in which (a) shows an example of the configuration, and (b) shows the processing in the spatial filter processing section. It shows.

Conventionally, x, y are positions in an image, function k(x, y) is the density of the image at that position, and the original image is F(x.

y), an image obtained by blurring the original image is f(x, y), and an image obtained by performing edge enhancement processing is g(x, y), then g
(x, y) = F (x, y) + a x (F (x, y) -
f(x, y)) Here, a is a process in which the contour is emphasized by a constant.

This method is known as unsharp masking.

Hereinafter, a specific processing method of the contour enhancement processing will be explained with reference to FIG.

First, an image input from a television camera, scanner, etc. is first stored in the image memory 1a, and then sent to the spatial filter processing section 1b.

In the spatial filter processing unit 1b, the above-mentioned contour is filtered by passing through the spatial filter ■ which is a composite of spatial filters I, II, and I[[ shown in FIG. Emphasis processing can be performed.

Here, the spatial filter l is a logical filter to obtain the original image F (x, y), the spatial filter ■ is a logical filter to increase the density of the original image by a times, and the filter ■
is a logical filter for multiplying a "blurred image" (which is the same as an image passed through a smoothing filter) by a.

However, this conventional method has the problem that not only the contours in the image but also the noise in the image are emphasized, resulting in a grainy and confusing image.

In view of the above-mentioned conventional drawbacks, the present invention aims to provide a processing method that can perform contour enhancement processing for emphasizing the contours in an image to obtain a sharp image without being affected by noise in the image. This is the purpose.

[Means to solve the problem]

FIG. 1 is a detailed explanatory diagram of the present invention.

The above problem can be solved by an edge enhancement processing method configured as follows.

The original image F(x, y) at the coordinates (x, y) in the image.

Difference with blurred image f(x,y) F(x,y)-f(x,y)-z For, β h(z) −r lz/αl tz≧0β −rlz/αl, z<0 Here, β>1. α, T are constants The original image F(x.

y).

[Effect]

That is, according to the present invention, in the contour enhancement processing method for emphasizing the contours in an image to obtain a sharp image, the original image and
Focusing on the fact that the value of the noise part of the difference in the "blurred image" is not a very large value, the edge enhancement is performed by making the value of the small difference smaller and adding it to the original image.

When the difference between the original image and the "blurred image" is small, the coordinate (x
, y) and the blurred image f
Difference from (x, y) F(x, y) - f(x, y) =
For z, β h(z) = T lz/αl , z≧0β - γlz/αl , zo.

Here, β>1. α, T are constants Use the function

As shown in the principle explanatory diagram of Fig. 1, this function becomes smaller when the difference 2 is small (indicated by Z in the figure), but like the contour line in the image, Pixels with a certain degree of difference are converted to have a larger difference, so noise components are suppressed from being emphasized, and an image with only the outline emphasized can be obtained, resulting in a sharp image that suppresses the roughness caused by noise. This has the effect of allowing you to obtain a clear image.

〔Example〕

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

The above-mentioned FIG. 1 is a diagram explaining the present invention in detail, and FIG.
The figure shows an embodiment of the zero invention, in which the original image F(x) obtained by the spatial filter processing section 1b.

y) and the "blurred image" f(x, y), the density conversion processing unit 1c performs density conversion h(z), which is necessary to carry out the present invention. It is. Note that the same reference numerals indicate the same objects throughout the figures.

Hereinafter, the outline enhancement processing method of the present invention will be explained with reference to FIG. 2 while referring to FIG.

First, in the same way as in the conventional method (see Figure 3), an image input from a scanner or the like is temporarily stored in a specific area of the image memory 1a, sent to the spatial filter processing section 1b, and processed by the above-mentioned spatial filter ( See Figure 3(b), but a.
1), the original image F(x, y) and the blurred image f(
x, y) is calculated.

F(x, y) - f(x, y) = zThe contents of the cough difference Z are stored again in another specific area of the 5-image memory a, and from there are sent to the density conversion processing unit 1c of the present invention. Sent.

In the density conversion processing section 1c, β h(z) = r lz/lx l, z≧0β - γIz/αJ, z<0 Here, β>1. α, T are constants Concentration conversion is performed according to the following.

This conversion function h (z) is the conversion function shown in Figure 1, so images with small differences such as noise are converted to smaller images, and images with large differences, such as normal contour images, are enhanced. and converted.

This conversion data is stored again in another specific contour of the image memory -a, and the resulting image and the original image are read out and sent to the arithmetic operation section 1d, where they are added (i.e.,
The converted image is added to the original image), and the resulting image is stored in the image memory -a and output to an output device such as a printer (not shown).

As described above, in the contour enhancement processing method of the present invention, the difference between the original image and the blurred image is transformed to become smaller for images with small differences such as noise, and the noise is reduced. The feature is that it is able to emphasize contours with less influence.

There are certain effects that can be achieved.

[Brief explanation of drawings]

FIG. 1 is a diagram explaining the principle of the present invention. [Effects of the Invention] As explained above in detail, the contour enhancement processing method of the present invention emphasizes the contours in an image to obtain a sharp image.
For the difference between the original image and the 2-blurred image, use a function that reduces the edge enhancement of pixels with a small difference value, for example, an exponential function that is symmetrical about the origin and convex downward in the first quadrant. Since the density of the pixel value for the difference Z is converted and added to the original image, when the difference Z is small, such as noise in the image, the cough difference becomes smaller; Pixels with a certain degree of difference, such as contour lines, are density-converted to have a larger difference, suppressing noise components from being emphasized, and producing an image in which only the contours are emphasized. Obtaining a sharp image with reduced roughness caused by noise In the drawing, 1a is an image memory, and lb is a spatial filter processing unit. 1c is a density conversion processing unit, and ld is an arithmetic operation unit 2 which is a difference between the original image and the blurred image. h (z) is the concentration conversion function. are shown respectively. Figure 1 Figure

Claims (1)

[Claims] Difference F (x, y) - f (x, y) between the original image F (x, y) and the blurred image f (x, y) at coordinates (x, y) in the image = z, h(z)=γ|z/α|＾β, z≧0 −γ｜z/α｜＾β, z<0 where β>1, α, γ are constant functions A contour enhancement processing method characterized by adding image data whose contours have been enhanced in step 1 to an original image F(x, y).