JP2843232B2 - Apparatus and method for contrast enhancement - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates improves the contrast of the input image, co for obtaining a sharp image output contrast
It relates to enhancement apparatus and method, in particular, a high concentration distribution frequency of occurrence relates to emphasize to Turkey contrast enhancement apparatus and method <br/> contrast by separating as much as possible different density values.

[0002]

2. Description of the Related Art Conventionally, Japanese Patent Application Laid-Open No. 61-45676 discloses a method for enhancing the contrast by converting the density value histogram of input pixels and the frequency value into a density distribution so as to have a uniform density distribution over all density regions. It has been disclosed.

This will be described with reference to pixel data having information of coordinates (x, y) (position) and density as shown in FIG. In the example shown in FIG. 1, the density value is 2 of 6 and 7.
Value only.

First, the cumulative frequency of pixels at each density value on one screen is calculated, and the density value is plotted on the horizontal axis, and the frequency of occurrence (the number of pixels) for each density value is plotted on the vertical axis, as shown in FIG. Create The histogram is useful for easily grasping the general properties of an image such as overall bright, dark, and low contrast.

Next, as shown in FIG. 3, the frequency of occurrence of each density value is equalized, and the histogram is flattened. By this histogram flattening, the contrast is enhanced as compared with the original image.

In FIG. 3, a pixel has coordinates (x,
According to the lexicographic order of y), they are assigned in ascending order of density.

In general, it is said that it is a good method to perform processing by ordering coordinates according to the density around a pixel. In FIG. 1, coordinates (2,1), (2,2), (3,1),
(3, 2), (4, 1), and (4, 2) have a density of 7, which is higher than that of the surroundings. FIG.

[0008]

THE INVENTION Problems to be Solved] However, the conventional
Also referred to as the contrast enhancement processing (high-contrast processing
) , There is a problem that the result greatly changes depending on which pixel of the same density before the flattening process is assigned to which density after the flattening process.

As described above, in order to perform the processing of ordering the coordinates according to the density around the pixel as shown in FIG. 4, the algorithm is complicated and it is difficult to realize the apparatus. However, if such processing is not performed at random and there is a problem, the accuracy of the image itself is considerably reduced.

The present invention has been made to solve the above-mentioned conventional problems. In the flattening process, only density value information is converted without requiring both pixel coordinate information and density value information. it allows the high density distribution of occurrence frequency, an object of <br/> strong tone to Rukoto contrast to separate as much as possible different density values.

[0011]

Means for Solving the Problems The present invention, the input image, the contrast enhancement unit that performs processing to enhance the contrast, a histogram showing <br/> the degree number of each density value of each pixel of the input image means for creating, the histogram
All the input pixels in descending order of density value.
Means for flattening the frequency relative density level, the density value of input Chikaraga containing the same in the obtained density distribution in the flattened
By converting the density values <br/> of approximate mean took values of the distribution of the original input image density value, by Ru and means for emphasizing the contrast is obtained by achieving the above object.
The present invention also the input image, the contrast enhancement method for performing processing for emphasizing contrast, a histogram that indicates the degree number of each density value of each pixel of the input image, using the histogram, the input pixel Low density value
The frequencies are flattened for all the density levels in order from the density, and the density value of the input pixel is changed to the density distribution obtained by the flattening.
The average of the distribution of the same original input image density values in
The above object is also achieved by converting the density value into a density value .

[0012]

According to the present invention, the density value of each pixel of the input image and the histogram of the frequency values are used to determine the low density value of the input pixel.
The frequencies are flattened for all the density levels in order from the one , and the density value of the input pixel is converted to the density distribution obtained by the flattening.
The average of the distribution of the same original input image density values in
Is converted to the density value of
By separating the density values as different as possible, the contrast is enhanced.

[0013]

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

FIG. 5 is a block diagram showing the basic configuration of the present invention.

In FIG. 5, 10 is a camera, 12 is an analog / digital converter, 14 is a frame memory, 16 is a histogram generator, 18 is a density value separation circuit,
0 is a digital / analog converter and 22 is a display.

FIG. 6 is a block diagram showing an outline of the configuration of this embodiment. In FIG. 6, the density value separating circuit 18 of FIG. 5 is replaced by a density converting circuit 26 comprising a CPU 24 and a RAM. It has been replaced.

The operation of the embodiment will be described below.

An image captured by the camera 10 is converted into an analog /
The digital signal is converted into a digital signal by the digital converter 12 and input to the frame memory 14. From this, a histogram representing the frequency for each density value is created by the histogram generator 16. Using the generated histogram and the data in the frame memory, the CPU 24 calculates a converted value of the density value.

This will be described using a specific example.

FIG. 7 shows a histogram obtained from the pixel data as shown in FIG. Unlike FIG. 2, this does not require the coordinates (x, y) of each pixel. Simply use frequencies.

FIG. 8 shows a flattened structure. FIG.
The density 6 is frequency 18 and the density 7 is frequency 6, so the total frequency is 2
4, when this is flattened, the number of density values (the number of gradations) becomes 8
Therefore, 24/8 = 3, and the frequency is 3 for all the densities.

In FIG. 7, since the frequency h (n) = 0 for the density values n = 0 to 5, the conversion value C of the density value is set to 0 for these. For the density value n = 6, FIG.
In what was degree h (n) = 18 is, by flattening, to the density value n = 0 in FIG. 8 to 5, so moved by power h (n) = 3, immediately to the rear out (1) The following equation

(Equation 1) Therefore, the conversion value C (6) is set to 2 or 3 (2 in the figure).
Also Similarly for a concentration value n = 7, the concentration value n = 7 in FIG. 7
Has a frequency h (n) = 6, but in FIG. 8, since the frequency h (n) = 3 has been shifted to density values n = 6 and 7, respectively, the following equation (1) is used.
The following equation according to

(Equation 2) Therefore, the conversion value C (7) is set to 6 or 7 (6 in the figure).

These conversion values indicate the density value to which the frequency before flattening is shifted. In the above example, 18 frequencies having the density value 6 originally correspond to the density value 2 and the density value 7 corresponds to the density value. Six frequencies shift to the density value 6.

FIG. 9 shows this as a histogram.
FIG. 10 shows the pixel data.

As described above, what originally had a density difference of only 1 at densities of 6 and 7 shifts to a density difference of 4 at densities 2 and 6, thereby enhancing the contrast.

The density conversion circuit 26 shown in FIG. 6 converts the density into these converted values, passes through the frame memory 14, converts it again into an analog signal with the digital / analog converter 20, and displays it on the display 22.

The above-mentioned calculation for obtaining the conversion value is represented by the following general formula. When the density value (integer) of the histogram is n and the histogram frequency (integer) at each density value n is h (n), when n takes a value from 0 to the density MAX, T (n) = T (n-1) ) + H (n) where T (-1) = 0 B (n) = [ n ] where [n] is the maximum integer A not exceeding n = the total frequency / the gradation value , and the density conversion value ( Integer) C (n) is calculated by the following equation (1).

[0028]

(Equation 3)

However, in the equation (1), B (i ÷ A) is always an integer, and when h (n) = 0 , the density conversion is performed.
And C = 0.

A density value separation circuit in the basic configuration extracts data from the histogram generator and the frame memory, and calculates a conversion value C by performing density conversion according to the above equation.

As described above, according to the present invention, the contrast is enhanced by converting only the density value information without requiring the coordinate information of the pixel.

For example, according to the conventional technique disclosed in Japanese Patent Application Laid-Open No. 61-45676, all the density levels are reduced to one as shown in FIG. In the present invention, contrast is further enhanced by separating a density distribution having a high appearance frequency into density values as different as possible, for example, as shown in FIG. Is what you do.

[0033]

As described above, according to the present invention, both coordinate information and density information of a pixel are not required.
Using only the density value information, the density conversion value can be obtained by a simple calculation, and the contrast can be further enhanced.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram showing an example of input pixel data.

FIG. 2 is a histogram showing the relationship between the density and frequency of the pixel data in FIG.

FIG. 3 is a graph obtained by flattening the histogram of FIG. 2;

FIG. 4 is an explanatory diagram showing pixel data converted by flattening in FIG. 3;

FIG. 5 is a block diagram showing a basic configuration of the present invention.

FIG. 6 is a block diagram showing a schematic configuration of the present embodiment.

FIG. 7 is a histogram for explaining the embodiment.

FIG. 8 is a graph obtained by flattening the histogram of FIG. 7;

FIG. 9 is a histogram showing density value separation according to the embodiment.

FIG. 10 is an explanatory diagram showing the histogram of FIG. 9 as pixel data.

FIG. 11 is a histogram as an example for comparing the conventional technique with the present invention.

FIG. 12 is a histogram showing conventional flattening.

FIG. 13 is a histogram showing density value separation according to the present invention.

[Explanation of symbols]

 Reference Signs List 10 camera 12 A / D 14 frame memory 16 histogram generator 20 D / A 22 display 24 CPU 26 density conversion circuit

Claims (2)

(57) [Claims] With respect to [claim 1] input image, we emphasize the contrast
In processing the contrast enhancement unit that performs, means for creating a histogram showing the degree number of each density value of each pixel of the input image, using the histogram, having a low density value of the input pixel
Flatten frequencies for all density levels in order from
Means and the density value of the input pixel are added to the density distribution obtained by the flattening.
Of the same original input image density value distribution
To enhance the contrast by converting to the density value of
Means for contrast enhancement. To 2. An input image, the contrast enhancement method for performing processing for emphasizing contrast, a histogram that indicates the degree number of each density value of each pixel of the input image, by using the histogram of the input pixel Those with low concentration values
In order, the frequencies are flattened for all the density levels , and the density values of the input pixels are added to the density distribution obtained by the flattening.
Of the same original input image density value distribution
To enhance the contrast by converting to the density value of
Contrast enhancing method.