JPH10208034A - Processor and method for image processing - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate the need for operation based upon operator's intuition by allocating a small number of gradations to a range of small distribution density while giving a large number of gradations to a range of large distribution density according to the quantity of extent of a calculated luminance distribution. SOLUTION: An image input device 10 outputs image data to the image processor 20 by picking up an image, etc. This image processor 20 extracts a distribution of luminance, detects the quantity of extent of the luminance distribution first on the basis of the detected luminance distribution, allocates a small number of gradations to the range of small distribution density while giving a large number of gradations to the range of large distribution density, and converts the image data on the basis of the allocated gradations of luminance. Namely, the distribution of luminance (y) is found as to pixels of the image data while the data are thinned out, and then standard deviation σcorresponding to the extent quantity of the luminance distribution is found; and parameters (γ1, γ2) for γ correction for generating correspondence relation of an S curve is calculated on the basis of the standard deviation σ.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an image processing apparatus and an image processing method, and more particularly, to an image processing apparatus and an image processing method for enhancing the contrast of dot matrix image data.

[0002]

2. Description of the Related Art Image data captured by a scanner or captured by a digital still camera is arranged in a dot matrix. Although it is sufficient if the captured image data is in a preferable state, sometimes it is desired to make a correction according to the taste by utilizing the characteristics of the digital image. In particular, when the contrast is low, it is difficult to correct a photo using a conventional color film, and such a photo must be retaken, but if it is a digital image, use image correction software or the like. It is also possible to enhance the contrast.

[0003] That is, conventionally, image correction software has been used to correct such digital image data, and an operator can perform various corrections by activating the software on a computer. For example, the enhancement of the contrast as described above can be performed by performing an operation of increasing the luminance. More specifically, by changing the correspondence between the luminance of the conversion source and the luminance of the conversion destination from a direct proportional state to a state in which the luminance is appropriately distorted into an S shape, the luminance having a certain width at the conversion source is converted. It is possible to correspond to a wider width earlier.

[0004]

In the above-mentioned conventional image processing apparatus, it is possible to enhance the contrast by an operator's operation, but the necessary parameter values must be given by intuition. For those unfamiliar with, there was a problem with operability, such as the inability to enhance the contrast as desired and the dusk of the picture as if it were midday.

[0005] The present invention has been made in view of the above problems, and has as its object to provide an image processing apparatus and an image processing method capable of eliminating the need for an operation relying on the intuition of an operator.

[0006]

In order to achieve the above object, according to the first aspect of the present invention, a luminance equivalent value of a predetermined gradation for each pixel in dot matrix image data is inputted. And performing a predetermined conversion process and outputting the result, thereby performing a conversion for enhancing the contrast in a relationship between the input and the output. The spread amount of the distribution is calculated, and based on the spread amount, a large number of tones are assigned to a range with a high distribution density and a small number of tones is assigned to a range with a small distribution density.

In the invention configured as described above, first, a brightness distribution in the image data is obtained for the dot matrix image data, and a spread amount of the brightness distribution is calculated. Since the amount of spread of the luminance distribution can be said to indirectly represent the width of the contrast, a small number of tones is assigned to a range with a small distribution density while giving a large number of tones to a range with a large distribution density based on the amount of spread. . For example,
If the luminance distribution is concentrated in a certain width, if the number of gradations is given so as to increase the width, the contrast will be enhanced and the contrast will be sharpened, but it will be assigned to the part where the luminance distribution is not concentrated. There is no problem even if the number of gradations is reduced.

Of course, when the luminance distribution is not concentrated but spreads evenly, it is not necessary to change the assignment of the number of gradations.

The operation of obtaining the spread amount of the luminance distribution is performed by relying on the intuition that the operator has performed in the conventional operation, and the range of the number of gradations according to the spread amount is determined. The operation of assigning a large number corresponds to setting the parameter given by the operator.

Here, various methods are available for obtaining the spread amount from the luminance distribution. A value corresponding to a known spread amount may be obtained,
The spread amount may be obtained by performing a measure for reducing the calculation amount. According to a second aspect of the present invention, in the image processing apparatus according to the first aspect, the spread amount of the luminance distribution uses a value corresponding to a standard deviation, and the known spread amount is used. When the standard deviation is large, the input / output ratio is reduced, and when the standard deviation is small, the input / output ratio is increased.

In the invention configured as described above, since the standard deviation indicates the amount of variation of each element with respect to the average value, if the variation is large, the conversion of the luminance is not required, and the input / output ratio is increased. Therefore, there is no need to perform a conversion that enlarges the change in the assignment of the number of gradations. That is, since the luminance distribution is not concentrated, there is no need to increase or decrease the assignment of the number of gradations. However, a small standard deviation means that there is little variation. In this case, since the luminance distribution is concentrated, the contrast is emphasized by giving a large number of gradations to a large distribution density range. Let it.

Of course, since the standard deviation is used to represent the variation, it is synonymous with the variance, and the calculation process does not necessarily require division by the number of samples.

Giving a larger or smaller number of gradations generally corresponds to an input / output ratio. If a larger number of gradations is given, the number of gradations is larger than that given on the input side. Also gives a large number of gradations and expands the gradation within that range, which is an enhancement of contrast. On the other hand, since the total number of gradations cannot be increased or decreased, if a large number of gradations is given in a certain range, it is impossible to give a large number of gradations in the remaining range as a reaction, and it is given on the input side. The number of gradations smaller than the number of gradations given is given, and the contrast is narrowed within the range.

Although it is possible to determine how to convert the number of gradations for each gradation,
As a simpler example, the invention according to claim 3 is characterized in that, in the image processing apparatus according to claim 2, γ correction is performed on the luminance distribution and the amount of change due to γ correction is reduced when the standard deviation is large. The value of γ is set, and the value of γ is set so that the amount of change in γ correction increases when the standard deviation is small.

In the invention configured as described above, γ
The correction method gives a corresponding relationship such as convex upward or convex downward. In the case of convex upward, a large number of gradations is assigned to the first half and a small number is assigned to the second half. The number of gradations will be assigned, and if it is convex downward, a small number of gradations will be assigned to the first half and a large number of gradations will be assigned to the second half. The magnitude of the assigned amount can be appropriately adjusted by the amount of change due to the γ correction, and is corresponded by the given γ value.

According to a fourth aspect of the present invention, in the image processing apparatus according to the third aspect, the approximate center position of the luminance distribution is determined, and high luminance is determined with reference to the approximate center position. By inverting the polarity of the luminance conversion between the side and the low luminance side, the luminance conversion of the approximate S-curve is performed in relation to the input and output.

In the invention configured as described above, γ
In order to invert the polarity of the luminance conversion by the correction, the γ correction of γ <1 is performed on the high luminance side, and the γ of γ> 1 is corrected on the low luminance side.
By performing the correction, the downwardly convex curve and the upwardly convex curve are continuous, and a so-called S-shaped curve input / output relationship is established. Here, since a large number of gradations are assigned to the central portion, if a continuous point of the two curves is brought to the approximate center of the luminance distribution, a large number of gradations are set in a range where the luminance distribution is to be expanded. Can be assigned.

In converting the luminance by various methods, the invention according to claim 5 is an image processing apparatus according to any one of claims 1 to 4, wherein the image processing apparatus according to claim 1 is configured such that the luminance of the conversion destination is within the range of the luminance of the conversion source. The configuration is such that the luminance is calculated and stored, and at the time of conversion, this correspondence is recalled and converted.

Although it is not impossible to calculate the luminance every time based on the conversion formula, the range of values that the luminance distribution can take is determined. For this reason, if the brightness of the conversion destination is calculated and stored in advance based on the brightness of the conversion source, the conversion can be performed only by invoking the correspondence at the time of conversion.

In converting the luminance, the image data may include the luminance data in some cases, or may include the luminance data only indirectly. Of course,
If it includes direct luminance data, it may be converted, or even if it is indirect luminance data, it may be converted into luminance data and then subjected to predetermined luminance conversion. However, the conversion of the luminance does not have to be extremely accurate, and it can be said that it is only necessary to roughly know.

Since strict accuracy is not required in that sense, the invention according to claim 6 is the image processing apparatus according to any one of claims 1 to 5, wherein the image data corresponds to the luminance. When represented by a plurality of component values, the configuration is such that the calculation of the luminance is obtained by linear addition of the same component values.

When image data is represented by so-called RGB (red, green, blue) gradation data, it can be said that each component value for red, green, and blue corresponds to luminance. For this reason, it can be said that linear addition of the same component value sufficiently represents luminance, and can be a very easy conversion method.

Assuming that the luminance for each pixel is determined, the luminance distribution as an image does not necessarily need to be determined for all the pixels of the image data. For example, the invention according to claim 7 is based on claims 1 to 6 above. In the image processing device described in (1), the luminance distribution is obtained by performing thinning-out corresponding to a predetermined extraction rate on the image data.

In order to obtain the distribution, it is possible to obtain a luminance distribution having a certain degree of accuracy according to the extraction rate even if thinning is performed at a predetermined extraction rate without obtaining the luminance for all the pixels. Can be.

Here, although there are various methods for thinning out, the invention according to claim 8 is the image processing apparatus according to claim 7, wherein the predetermined number of extractions on the short side in the vertical direction and the horizontal direction is reduced. Is ensured.

Since an image is planar, its image data is naturally distributed in the vertical and horizontal directions. To determine a certain extraction rate, at least a certain number of extractions must be secured on the short side. The certainty corresponding to the extraction rate is maintained.

Further, according to a ninth aspect of the present invention, in the image processing apparatus according to any one of the first to eighth aspects, a limit is set on a degree of enhancement of contrast.

In some cases, the contrast is narrow. For example, it is natural that the width of the luminance distribution is narrow in the evening scene, and if it is expanded beyond necessity, it becomes a day scene. A similar example is possible in other cases, and such a phenomenon is avoided by setting a limit on the enlarged range of the luminance distribution.

Further, according to a tenth aspect of the present invention, in the image processing apparatus according to the first to ninth aspects, the binary image data is determined based on the luminance distribution.
In the case of binary image data, contrast enhancement is not performed.

Since it can be said that there is no substantial luminance distribution in the binary image, the contrast is not enhanced when the binary image data is determined from the luminance distribution.

Since the binary image data can have a certain color, it can have two luminances corresponding to the presence or absence of the color. It is possible to determine whether or not it is the luminance of the color, but in the case where there is no information suggesting that,
According to an eleventh aspect of the present invention, in the image processing apparatus according to the tenth aspect, when the luminance distribution is concentrated at both ends within the reproducible range, it is determined that the image data is monochrome binary image data. There is a configuration.

That is, for a black-and-white image, it can be said that the luminance distribution is concentrated at both ends within the reproducible range, and the determination can be made.

Further, according to a twelfth aspect of the present invention, in the image processing apparatus according to the first to eleventh aspects,
The frame portion of the image data is determined based on the prominent luminance distribution, and if there is a frame portion, the data of the frame portion is not used for enhancing the contrast.

When an image is processed, it often happens that the image has a frame. If the image exists as a single-color frame, only the luminance distribution corresponding to the color is naturally prominent. Therefore, if such a prominent luminance distribution is used as a standard for judging the enhancement, it may not be possible to make an effective judgment. Therefore, the frame is judged to be a frame portion and is not used for enhancing the contrast.

Further, as an example, in the invention according to claim 13, in the image processing apparatus according to claim 12, the luminance distribution concentrated on the edge within the reproducible range is the frame portion. Is determined.

A white frame or a black frame is frequently used and employed, and may also be generated as a result of trimming.
It corresponds to the end within the reproducible range. Therefore, the luminance distribution concentrated at this end is determined as the frame.

According to a fourteenth aspect of the present invention, in the image processing apparatus of the first to thirteenth aspects, the contrast is not enhanced when the image data is not a natural image.

It is natural images such as photographs that narrow contrast is likely to cause a problem, and it can be said that such images are hardly necessary in business graphs. Conversely, tweaking things like business graphs can result in a different image from the creator. Therefore, the luminance distribution is expanded only in the case of such a natural image.

As an example of the determination as to whether or not the image data is a natural image, in the image processing apparatus according to the present invention, when the luminance distribution exists in a spectral form, the image data It is configured to include a natural image determination unit that determines that the image is not the same.

As a feature of a natural image, it can be said that the luminance distribution has a smooth width. Therefore, if the luminance distribution appears in the form of a line spectrum, it is generally safe to judge that the image is not a natural image. In the invention according to claim 15 configured as described above, the natural image determining means determines the state of the luminance distribution,
If the image data exists in the form of a line spectrum, it is determined that the image data is not a natural image, so that the luminance distribution is not enlarged.

Further, according to the present invention, for the luminance equivalent value for each pixel in the dot matrix image data, the same luminance equivalent value is input, subjected to a predetermined conversion process, and output, whereby the input is achieved. In performing the conversion for enhancing the contrast based on the relationship between the image data and the output, the luminance distribution in the image data is obtained, and the spread amount of the luminance distribution is calculated. The configuration is such that a small number of gradations is assigned to a range with a small distribution density while giving a number.

That is, there is no difference that the present invention is not necessarily limited to a substantial device but is also effective as a method.

By the way, such an image processing apparatus may exist alone or may be used in a state of being incorporated in a certain device. The idea of the invention includes various modes. In addition, it can be changed as appropriate, such as software or hardware.

As one example, a printer driver that converts input dot matrix image data into image data corresponding to printing ink and prints the image data on a predetermined color printer is also used for each pixel in the image data. The spread amount of the luminance distribution is calculated while obtaining the luminance distribution based on the equivalent value, and based on the spread amount, a large number of gradations are given to a range where the distribution density is large and a small number of gradations are given to a range where the distribution density is small. Can be configured so as to enhance the contrast by performing a conversion process between the input and the output so that the correspondence is assigned.

That is, the printer driver converts the input image data in accordance with the printing ink. At this time, the printer driver obtains a luminance distribution based on the luminance equivalent value of each pixel in the image data, and expands the luminance distribution. Between the input and the output so that there is a correspondence relationship that calculates the amount and assigns a large number of gradations to a range with a high distribution density while assigning a small number of gradations to a range with a small distribution density based on the spread amount. The input image is converted so as to enhance the contrast by performing a conversion process, and the input image is printed.

When the software of the image processing apparatus is realized as an example of realizing the idea of the present invention, the software naturally exists on a recording medium on which such software is recorded, and it cannot be said that the software is used. Of course, the recording medium may be a magnetic recording medium or a magneto-optical recording medium, and any recording medium to be developed in the future can be considered in the same manner. Also, the duplication stages of the primary duplicated product, the secondary duplicated product, and the like are equivalent without any question. In addition, the present invention is not limited to the case where the present invention is used even when the supply is performed using a communication line.

Further, even when a part is implemented by software and a part is implemented by hardware, there is no difference in the concept of the invention, and it is necessary to store a part on a recording medium. It may be in a form that is appropriately read in accordance with it. Further, it is needless to say that the present invention can be applied to an image processing apparatus such as a color facsimile machine and a color copier.

[0048]

As described above, according to the present invention, the degree of contrast enhancement is automatically determined from the spread amount of the luminance distribution, so that even an unskilled person can easily enhance the contrast by an appropriate amount. A processing device can be provided.

According to the second aspect of the present invention, the input / output ratio is simply determined based on the well-known standard deviation, and the configuration is simplified.

Further, according to the third aspect of the present invention,
Since the conversion is performed by the γ correction, the configuration is simplified.

Further, according to the invention of claim 4,
By performing luminance conversion based on the correspondence relationship of the approximate S-shaped curve with reference to the approximate center position of the luminance distribution, it is possible to perform efficient distributed conversion when it is desired to enhance contrast.

Further, according to the invention of claim 5,
Conversion can be facilitated.

Further, according to the invention according to claim 6,
The luminance can be easily obtained with a necessary and sufficient degree of accuracy.

Further, according to the seventh aspect of the present invention,
The processing amount can be reduced.

Further, according to the invention of claim 8,
The bias of the extraction points of the image is eliminated, and the luminance distribution is likely to be accurate.

Further, according to the ninth aspect of the present invention,
It is possible to prevent the contrast of the image from being excessively enhanced and changing the atmosphere of the image.

Further, according to the tenth aspect of the present invention, it is possible to easily determine an unnecessary condition for contrast enhancement so as not to perform the enhancement. Further, according to the eleventh aspect of the present invention, In addition, a frequently-used black and white image can be efficiently determined.

Further, according to the twelfth aspect of the present invention, it is possible to prevent the processing from becoming inaccurate due to the brightness of the frame portion which tends to appear in the image. The frequently used black and white frame portion can be easily determined.

Further, according to the fourteenth aspect of the present invention, it is possible to perform the processing only in the case of a natural image for which contrast enhancement is necessary. Can be easily determined.

Further, according to the invention of claim 16, since the degree of contrast enhancement is automatically determined from the spread amount of the luminance distribution, even an unskilled person can easily emphasize the contrast by an appropriate amount. An image processing method can be provided.

[0061]

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

FIG. 1 is a block diagram showing an image processing system according to an embodiment of the present invention, and FIG. 2 is a block diagram showing an example of a specific hardware configuration.

In the figure, an image input device 10 outputs image data to an image processing device 20 by taking an image or the like, and the image processing device 20 performs image processing such as predetermined contrast enhancement and performs image processing. 30 and the image output device 30 displays an image with enhanced contrast.

Here, a specific example of the image input device 10 corresponds to the scanner 11, the digital still camera 12, or the video camera 14, and a specific example of the image processing device 20 corresponds to a computer system including the computer 21 and the hard disk 22. A specific example of the image output device 30 corresponds to the printer 31, the display 32, or the like. Of course, other than these, the present invention can be applied to a color copying machine, a color facsimile machine, and the like.

In the present image processing system, an image having low contrast is provided with an optimum contrast. Therefore, image data obtained by taking a photograph with the scanner 11 as the image input device 10 may be used.
For example, image data having low contrast captured by the digital still camera 12 or a moving image captured by the video camera 14 is to be processed and input to the computer system as the image processing device 20. The calculation speed of the input image of the video camera 14 may not be enough. In such a case, it is possible to deal with this by setting the first condition that requires a calculation time for each shooting scene and performing only image conversion of each frame under the same condition setting during shooting.

The image processing apparatus 20 includes at least a luminance distribution detecting means for extracting a luminance distribution, a luminance distribution spread amount detecting means for first detecting a spread amount of the luminance distribution based on the detected luminance distribution, A tone number assigning means for assigning a large number of tones to a range with a high distribution density and assigning a small number of tones to a range with a small distribution density based on the spread amount; And image data conversion means for converting data. Needless to say, the image processing apparatus 20 may be configured as a color conversion unit that corrects a difference in color between models or a resolution conversion unit that converts a resolution corresponding to each model. . In this example, the computer 21 has a RAM
While using the internal ROM and hard disk 2
2 is executed.

The execution result of the image processing program is obtained as contrast-enhanced image data as will be described later, and the image output device 3 is executed based on the obtained image data.
The image is printed by the printer 31 which is 0 or displayed on the display 32 which is the same image output device 30. This image data is more specifically RGB (green, blue, red) gradation data, and the image is displayed in the vertical direction (heig).
ht) and dot matrix data arranged in a grid in the horizontal direction (width).

In the present embodiment, a computer system is incorporated between image input / output devices to perform image processing. However, such a computer system is not always required, and as shown in FIG. An image processing device for enhancing contrast is incorporated in the digital still camera 12a, and the digital still camera 12a displays the converted image data on the display 32a or the printer 31a.
May be a system that prints the data. As shown in FIG. 4, in a printer 31b that inputs and prints image data without using a computer system, image data input via a scanner 11b, a digital still camera 12b, a modem 13b, or the like is automatically converted. It is also possible to configure so that contrast is enhanced.

FIG. 5 shows a luminance distribution detecting process corresponding to the luminance distribution detecting means and the luminance distribution spread amount detecting means among the image processing executed by the computer 21. FIG. 6 shows a luminance conversion process corresponding to the conversion means.

FIG. 5 mainly corresponds to the luminance distribution detection processing. First, the luminance distribution detection processing will be described.

Before explaining how to represent luminance, pixels to be distributed will be described. As shown in step S102 of FIG. 5, a thinning process for thinning out target pixels is executed. As shown in FIG. 7, in the case of a bitmap image, a two-dimensional dot matrix including predetermined dots in the vertical direction and predetermined dots in the horizontal direction is established. It is necessary to check the brightness for. However, this distribution extraction processing is aimed at obtaining the spread amount of the luminance distribution, and does not necessarily need to be accurate. Therefore, it is possible to perform the thinning to the extent that it falls within a certain error range. According to the statistical error, the error with respect to the number N of samples can be approximately expressed as 1 / (N ** (1/2)). Here, ** indicates a power. Therefore, in order to perform processing with an error of about 1%, N = 10000.

Here, the bitmap screen shown in FIG. 7 has the number of pixels of (width) × (height), and the sampling period ratio is ratio = min (width, height) / A + 1 (1). This min (width, height) is w
the smaller of height and height, A
Is a constant. In addition, the sampling period ra
tio indicates how many pixels are sampled. Pixels marked with a circle in FIG.
2 is shown. In other words, one pixel is sampled every two pixels in the vertical direction and the horizontal direction, and sampling is performed every other pixel. When A = 200, the number of sampling pixels in one line is as shown in FIG.

As can be seen from the figure, except for the case where the sampling period ratio = 1 where sampling is not performed, when the width is 200 pixels or more, the number of samples is at least 100 pixels or more. Therefore, when the number of pixels is 200 or more in the vertical and horizontal directions, (100 pixels) × (100 pixels) = (10000 pixels)
Is ensured, and the error can be reduced to 1% or less.

Here, min (width, hei)
(ght) is based on the following reason.
For example, assuming width >> height as in the bitmap image shown in FIG. 10A, if the sampling cycle ratio is determined by the longer width, as shown in FIG. In addition, it may happen that only two lines, the upper and lower lines, are extracted in the vertical direction. However, min (wi
(dth, height), if the sampling period ratio is determined based on the smaller one, it is possible to perform the thinning including the intermediate part even in the smaller vertical direction as shown in FIG. Become.

In this example, thinning is performed at an accurate sampling cycle for pixels in the vertical and horizontal directions. This is suitable when processing is performed while thinning out sequentially input pixels. However, when all pixels have been input, pixels may be selected by randomly designating coordinates in the vertical direction and the horizontal direction.
In this way, when the minimum required number of pixels, such as 10,000 pixels, is determined, the process of randomly extracting until 10,000 pixels is repeated is repeated.
It suffices to stop the extraction at the time of the pixel.

If the pixel data of the selected pixel has a luminance as a component element, the distribution can be obtained by using the luminance value. However, even in the case of image data whose luminance value is not a direct component value, the image data has a component value representing luminance indirectly.
Therefore, a luminance value can be obtained by converting from a color space in which the luminance value is not a direct component value to a color space in which the luminance value is a direct component value.

The color conversion between these different color spaces is not uniquely determined by the conversion formula, but a correspondence is obtained between the color spaces having the respective component values as coordinates. It is necessary to sequentially perform conversion by referring to a color conversion table that stores the relationship. In relation to a table, the component values are represented as gradation values. In the case of 256 gradations having three-dimensional coordinate axes, about 16.7 million (2
It must have a color conversion table of 56 × 256 × 256) elements. As a result of considering efficient use of storage resources, instead of preparing correspondences for all coordinate values, usually prepare correspondences for appropriate discrete grid points and use interpolation together I am trying to do it. Since this interpolation calculation can be performed through several multiplications and additions, the amount of calculation processing becomes enormous.

That is, if a full-size color conversion table is used, the amount of processing is reduced, but the table size becomes an unrealistic problem. Often becomes the target.

In view of such a situation, the present embodiment employs the following conversion formula for obtaining luminance from the three primary colors of RGB as used in a television or the like. That is, the luminance yp at the point P is R
From the GB component values (Rp, Gp, Bp), yp = 0.30Rp + 0.59Gp + 0.11Bp (2) In this way, a luminance value can be obtained only by three multiplications and two additions. And
A frequency distribution is obtained using a variable area of an array corresponding to all gradations.

In the present embodiment, as a result of targeting the RGB color space, such a conversion formula is employed. However, since each component value indicates the brightness of the color in the background, There is a property that when each component value is viewed independently, it linearly corresponds to luminance. Therefore, more simply, it is not possible to simply simplify to yp = (Rp + Gp + Bp) / 3 (3) without considering each addition ratio, and furthermore, yp = Gp (4) As described above, it is also possible to use the green component value having the largest ratio in the equation (3) as the luminance value.

In the thinning-out process, R
The luminance is simultaneously obtained from the GB image data to obtain a distribution.
Finally, in step S114, a standard deviation corresponding to the spread amount is obtained based on this distribution.
Before that, there is something to consider.

The first is a case where the image is a binary image such as a black and white image. For a binary image including a black-and-white image, the concept of enhancing contrast is inappropriate. FIG.
Assuming that there is a black-and-white image as shown in FIG. 12, the luminance distribution for this image is concentrated at both ends within the range of assignment of the number of gradations as shown in FIG. Basically, the gradation is "0"
And the gradation “255”.

Therefore, when performing a black-and-white check in step S104, it can be determined whether or not the sum of the number of pixels of the gradation "0" and the gradation "255" matches the number of pixels selected by thinning. Then, in the case of a monochrome image, step S1 is executed to interrupt the processing without executing the following processing.
At 06, non-enlargement processing is executed. In the present embodiment, since the distribution extraction processing and the luminance conversion processing are largely divided, in the non-enlargement processing, a flag is set so that the subsequent luminance conversion processing is not executed, and the distribution extraction processing ends.

The binary data is not limited to black and white, but may be colored binary data. In such a case as well, it is not necessary to perform processing for enhancing the contrast. If the distribution state is examined and the distribution is concentrated only on two values (one is approximately “0”), the processing is interrupted as binary data. What should I do?

The second is to consider whether the image is like a business graph or a natural image such as a photograph. Although a process of enhancing contrast may be required in a natural image, it is often preferred not to enhance contrast in a business graph or a painting. Therefore, in step S108, it is checked whether the image is a natural image.

[0086] A natural image has a very large number of colors including shadows, but a certain number of paintings such as business graphs and draw systems often have a limited number of colors. Therefore, if the number of colors is small, it can be determined that the image is not a natural image. To accurately determine the number of colors, it is necessary to determine how many of the 16.7 million colors are used as described above, but this is not practical. On the other hand, when the number of colors is extremely small, as in a business graph, the probability of different colors and the same luminance is low. That is, the approximate number of colors can be determined based on the luminance. If the number of colors is small, the distribution of luminance is sparse, and in a business graph, it appears in a linear spectrum. Therefore, in step S108, 2
The number of luminance values whose distribution number is not “0” among the 56 gradation luminances is counted. Then, if it is equal to or less than the "64" color (gradation) which is approximately 1/4, it is determined that the image is not a natural image, and the non-enlargement process is executed in step S106, as in the case of binary data. Of course, the threshold "6
Whether or not the color is equal to or less than the "4" color (gradation) can be appropriately changed.

It is also possible to determine whether or not the distribution is in the form of a linear spectrum based on the adjacency ratio of luminance values whose distribution number is not “0”. That is, it is determined whether or not the number of distributions is a luminance value other than “0” and the number of distributions is adjacent luminance values.
If at least one of the two adjacent luminance values is adjacent, nothing is performed, and if both are not adjacent, counting is performed, and as a result, the ratio between the number of non-zero luminance values and the count value Judge it. For example, if the number of luminance values other than “0” is “64” and the number of non-adjacent luminance values is “64”, it is understood that the luminance values are distributed in a linear spectrum.

Further, when the image processing program is executed via the operating system, the determination can be made by the extension of the image file. Of the bitmap files, especially for photographic images and the like, file compression is performed, and an implied extension is often used to indicate the compression method. For example, "JP
If the extension is "G", it is understood that the file is compressed in the JPEG format. Since the operating system manages the file name, if the printer driver etc. sends an inquiry to the operating system, the extension of the file will be answered. It may be determined that there is, and the contrast may be enhanced. Also, "X
If the extension is unique to the business graph such as “LS”, it can be determined that contrast enhancement is not performed.

The third consideration is whether there is a frame around the image as shown in FIG. If such a frame portion is white or black, the luminance distribution is as shown in FIG.
As shown in (1), a linear luminance profile appears at both ends within the range of the number of gradations, and a smooth luminance distribution also appears inside other than both ends corresponding to the internal natural image.

Needless to say, since it is more appropriate not to take the frame portion into consideration of the luminance distribution, the sum of the number of pixels of the gradation “0” and the gradation “255” is sufficiently large in the check of the frame portion in step S108. Then, it is determined whether the number of pixels does not match the number of pixels selected by thinning out, and if affirmative, it is determined that there is a frame portion, and the frame portion process is performed in step S112. In the frame processing, the number of pixels of the gradation “0” and the gradation “255” in the luminance distribution is set to “0” in order to ignore the frame. As a result, in the following processing, it can be handled in the same manner as the case without the frame portion.

In this example, a white or black frame portion is targeted, but a frame of a specific color may be present. In such a case, a line spectrum shape protruding from the original smooth curve drawn by the luminance distribution appears. Therefore,
A line spectrum having a large difference between adjacent luminance values may be considered as a frame part and not subjected to luminance distribution. In this case, since the color may be used in a portion other than the frame portion, the average of the brightness values on both sides may be assigned.

In consideration of the above considerations, when enhancing the contrast, the standard deviation of the luminance distribution is obtained in step S114, and the median yMe is obtained for the luminance conversion processing in the subsequent stage. Although there are two ways of thinking about the standard deviation, in the present embodiment, it is calculated based on the following equation.

[0093]

(Equation 1)

Although the standard deviation corresponds to the spread of the luminance distribution, variance may be used to represent the spread.

The above processing corresponds to the distribution detection processing. Next, the luminance conversion processing for converting the image data based on the standard deviation σ which is the spread amount of the luminance distribution obtained in this manner will be described. When the non-enlargement process is performed in step S106 as described above, the process refers to a predetermined flag in step S202, detects the flag, and ends the image processing without performing the following process.

In the luminance conversion process, a small number of gradations is assigned to a range with a small distribution density while giving a large number of gradations to a range with a large distribution density based on the spread amount of the luminance distribution.
Here, a description will be given of a pattern in which a large number of gradations are assigned to a range with a high distribution density and a small number of gradations are assigned to a range with a small distribution density. When the luminance y (input) before the conversion and the luminance Y (output) at the conversion destination are in a directly proportional relationship as Y = y (6), as shown in FIG. , The gradation range r0 assigned before the conversion and the gradation range R0 assigned after the conversion match. However, as shown in FIG. 16, when the input / output correspondence becomes a so-called S-shaped curve, the gradation ranges R1 and R2 assigned after the conversion become larger than the gradation range r0 assigned before the conversion. That is, the number of gray levels increased. On the other hand, regarding the range outside the gradation range r0 on the low-luminance side and the high-luminance side in the input, the gradation range assigned after conversion is reduced.

That is, such a correspondence means that a small number of gradations is assigned to a range where the distribution density is small while a large number of gradations is given to a range where the distribution density is large. Here, various specific assignment methods for realizing this correspondence relationship are possible. FIG. 17 shows the result of performing γ correction of γ <1 for the region from the center position ymid of the gradation range to the upper end of the gradation range on the high luminance side, and the upper quadrant from the center position ymid before conversion. The converted gradation range R for the gradation range r up to yq3 is expanded. Similarly, FIG. 18 shows a region from the center position ymid of the gradation range to the lower end of the gradation range on the low luminance side where γ correction of γ> 1 is performed. The converted gradation range R for the gradation range r up to the dividing point yq1 is expanded.

On the other hand, in these cases, the ratio of the converted gradation range R to the pre-conversion gradation range r changes depending on how γ is given. In the present embodiment, the ratio is controlled based on the standard deviation σ, which is the spread amount of the luminance distribution. That is, assuming that the center position ymid of the gradation range is “128”, γ1 is given below this center position ymid, and γ1 is given above the center position ymid.
When y ≦ 128, γ1 = (σstd_limit / σ) ** a (7) When y> 128, γ2 = (σ / σstd_limit) ** a (8), and in step S204 To perform these parameter calculations. Note that, as described above, this parameter calculation constitutes the number-of-gradations assigning means. Where σstd_li
mit and a are parameters obtained experimentally in consideration of the conversion result. In the present embodiment, σstd_limit is set to “128” and a is set to “0.1”.
Since the standard deviation σ is generally smaller than “128”, if the standard deviation σ is large in these relational expressions, γ2 and γ
1 each approaches “1”, and the slope of the S-shaped curve becomes gentle. This means that when the spread amount is large, the gradation range R of the conversion destination does not become so large with respect to the gradation range r centered on the center position ymid, and more specifically, the luminance of the image data. Is widely distributed, it means that conversion for expanding the luminance range is not performed. On the other hand, if the standard deviation σ is small,
γ2 and γ1 are respectively apart from “1”, and S
The slope of the curve becomes steep. This means that, when the spread amount is small, the gradation range R of the conversion destination is expanded widely with respect to the gradation range r centered on the center position ymid. When the luminance is distributed only in a narrow range, it means that conversion for expanding the luminance range is performed.

As shown in this example, when the gradation range is divided into a low-luminance side and a high-luminance side, and γ correction is applied to each of them, a relationship in which γ is the inverse of each other is given. The connection is made smoothly at the connection point between the side and the high luminance side, and a good S-shaped curve can be given. Of course, changing the polarity of the γ correction between the low luminance side and the high luminance side itself forms the correspondence of the S-shaped curve, and can greatly change the luminance Y after conversion with respect to the luminance y before conversion. Become like

In the present embodiment, the correspondence between the S-shaped curves is established by γ correction, but FIG. 19 shows that Y = f · y + g before and after the center position ymid of the gradation range.
An example is shown in which a linear correspondence relationship is realized. In this example, a linear correspondence is formed again in the region below the lower quadrant yq1 and above the upper quadrant yq3. Also in this example, when the standard deviation σ is large, the slope f may be made closer to “1”, and when the standard deviation σ is small, the slope f may be made larger than “1”. Of course, in this case, in order to prevent a sudden change in the correspondence, FIG.
As shown in (2), two corresponding straight lines may be connected smoothly.

On the other hand, when the luminance distribution falls within the center of the gradation range, good luminance conversion can be performed only by setting γ1 and γ2 as described above. However, as shown in FIG. The median yMe is slightly closer to the low brightness side,
As shown in FIG. 22, the median yMe of the luminance distribution may be slightly shifted to the higher luminance side.

On the other hand, in the example shown in FIG. 23, the entire range is subjected to γ correction of γ <1, and in this case, the gradation range r on the low luminance side before the conversion and the gradation range after the conversion R is enlarged, and in the example shown in FIG.
> 1 is applied, and in this case, the gradation range R after conversion with respect to the gradation range r on the high luminance side before conversion is expanded.

Therefore, the magnitude relation between the median yMe obtained in step S114 and “128”, which is the center position of the gradation range, is compared, and as shown in FIG. 21, the median yMe is biased toward the region on the low luminance side. If so, by performing γ correction of γ <1 on the entire range as shown in FIG. 23, the gradation range r on the low luminance side where the distribution density was high before conversion is obtained.
Will expand. Further, as shown in FIG. 22, if the median yMe is biased toward the high-luminance area,
By performing the γ correction of γ> 1 on the entire range as shown in FIG. 24, the gradation range r on the high luminance side where the distribution density was high before the conversion is expanded. In this way, it is possible to expand or reduce the assignment of the number of gradations without necessarily having the correspondence relationship of the S-shaped curve. Of course, in these cases, γ may be determined based on the equations (7) and (8) depending on whether γ <1 or γ> 1.

Further, in the example shown in FIG. 25, five points of gradation "0", lower quadrant yq1, center position ymid, upper quadrant yq3, and gradation "255" are used as reference points. The conversion points at the lower quadrant yq1 and the upper quadrant yq3 are determined based on the standard deviation while setting Y = y for the gradation "0", the center position ymid, and the gradation "255". .
Then, the correspondence between these five points may be obtained by spline interpolation calculation or Newton interpolation. Of course, three points on the lower side and three points on the upper side from the center position ymid may be obtained by spline interpolation calculation or Newton interpolation, respectively.

By the way, if an excessively large gradation range is assigned to the gradation range before the conversion, it may be rather undesirable. In dusk conditions such as the evening, the contrast range from the brightest to the darkest is naturally narrow, but as a result of trying to greatly increase the contrast of this image, it is converted like a daytime image. Maybe. Since such a conversion is not desired, a limit is set for the enlargement ratio, and γ1,
Both γ2 are restricted. For example, even if γ2 <0.7, γ2 = 0.7, and γ1> 1.3, γ1
= 1.3.

As described above, in the present embodiment, the parameters γ1 and γ2 can be obtained, and step S204 ends. In addition, it can be said that such a change in the assignment of the number of gradations changes the assignment of the number of quantization bits, in other words.

By the way, it is irrational to execute the calculation by the γ correction every time the luminance is converted. That is, since the range of the luminance y can be only "0" to "255", the luminance Y after conversion can be obtained in advance for all possible values of the luminance y. That's why. Therefore, this correspondence is obtained in step S206 and stored as a table as shown in FIG.

Here, a specific calculation of the correspondence is as follows.

When y ≦ 128, Y = 128 * (y / 128) ** γ1 (9) When y> 128, Y = 128 * ｛(y−128) / 128｝ ** γ2 + 128 (10) , The conversion point of the γ correction can be changed based on the median yMe. That is, when y ≦ yMe, Y = yMe * (y / yMe) ** γ1 (11) When y> yMe, Y = yMe * ｛(y−yMe) / yMe｝ ** γ2 + yMe (12) For example, it becomes an S-shaped curve in which the polarity of the luminance conversion is inverted between the high luminance side and the low luminance side around the median yMe, and a large number of gradations can be assigned to both sides around the high distribution density, There is an effect that conversion can be performed without significantly affecting brightness.

When the calculated conversion table is formed, the image data can be changed.

Finally, the image data is converted in step S208. Up to this point, the correspondence relationship for converting the luminance has been obtained, and for example, the conversion relationship is not the component value (Rp, Gp, Bp) on the RGB coordinate axis. However, the conversion equation of equation (2) can also be applied in the correspondence relationship with the RGB component values (Rp, Gp, Bp). Further, corresponding to the luminances y and Y of the gradation “0” to the gradation “255”, the RGB component values (r, g, b) and (R, G, B) are in the same range. Therefore, it can be said that the conversion table of the luminances y and Y described above may be used as it is.

Accordingly, in step S208, the process of obtaining the converted image data (R, G, B) is repeated by referring to the conversion table shown in FIG. 26 for the image data (r, g, b) of all pixels. become.

Next, the operation of this embodiment having the above configuration will be described step by step.

Assuming that a photograph is taken by the scanner 11 or the like, image data representing the photograph in the form of RGB gradation data is taken into the computer 21, and the CPU operates as shown in FIGS.
Is executed to enhance the contrast of the image data.

First, in step S102, the image data is thinned out within a predetermined error range, and the luminance y of the selected pixel is obtained to obtain a distribution. The distribution cannot be used as it is. First, it is determined in step S104 whether the image is a binary image such as black and white, and in step S108, it is determined whether the image is a natural image. Unless the image is a binary image or a non-natural image, step S1
In step 10, it is determined whether or not there is a frame in the image data. If there is a frame, the standard deviation σ is obtained in step S114 for the luminance distribution obtained by removing the frame. In the present embodiment, the standard deviation σ may be simply obtained, but when the polarity conversion point of the γ correction is changed in accordance with the luminance distribution, the median yMe is obtained.

When the standard deviation σ of the luminance distribution is obtained, y ≦
For 128, γ1 = (σstd_limit / σ) ** a (7) Y = 128 * (y / 128) ** γ1 (9) For y> 128, γ2 (σ / σstd_limit) ** a (8) ) Y = 128 * {(y−128) / 128} ** γ2 + 128 (10) From the relational expression, the parameter γ is determined in step S204.
At step S206, the conversion relationship from the luminance y to the luminance Y is stored in a table. Then, image data for all pixels is converted with reference to the conversion table completed in step S208.

Of course, as described above, when the image is not a binary image or a natural image, such image processing is not performed.
When the image processing of the present invention is performed, a bright and dark image is obtained by correcting the brightness so as to be widened even though the contrast is very low in the state of the photograph. Will be able to

In the above embodiment, γ1
Although the parameter σstd_limit, a for obtaining γ2 and γ2 is fixed, the user may select the parameter on the computer 21 via a predetermined GUI. Also,
It is also possible for the user to designate a part of the image data and execute such contrast enhancement processing only within the range.

In this way, after obtaining the distribution of the luminance y for the pixels of the image data while thinning out in step S102, the standard deviation σ corresponding to the spread amount of the luminance distribution is obtained in step S114, and the standard deviation σ is obtained. Since the parameters (γ1, γ2) of γ correction for forming the correspondence of the S-curve are calculated in step S204 based on the deviation σ, the image data is converted in step S208 and Automating the task of assigning a large number of tones to a region with a high distribution density at luminance y and assigning a small number of tones to a region with a low distribution density makes it possible for even unskilled persons to easily emphasize contrast. become able to.

[Brief description of the drawings]

FIG. 1 is a block diagram of an image processing system to which an image processing apparatus according to an embodiment of the present invention is applied.

FIG. 2 is a block diagram illustrating a specific hardware configuration example of the image processing apparatus.

FIG. 3 is a schematic block diagram illustrating another application example of the image processing apparatus of the present invention.

FIG. 4 is a schematic block diagram illustrating another application example of the image processing apparatus of the present invention.

FIG. 5 is a flowchart showing a luminance distribution detection processing portion in the image processing apparatus of the present invention.

FIG. 6 is a flowchart showing a luminance conversion processing part in the image processing apparatus of the present invention.

FIG. 7 is a diagram illustrating coordinates in a conversion source image.

FIG. 8 is a diagram showing a sampling cycle.

FIG. 9 is a diagram showing the number of sampling pixels.

FIG. 10 is a diagram illustrating a relationship between a conversion source image and pixels to be sampled.

FIG. 11 is a diagram showing a black and white image.

FIG. 12 is a diagram illustrating a luminance distribution of a black-and-white image.

FIG. 13 is a diagram showing an image having a frame portion.

FIG. 14 is a diagram illustrating a luminance distribution of an image having a frame portion.

FIG. 15 is a graph showing a relationship between luminance before conversion and luminance after conversion when the assignment of the number of gradations is not changed.

FIG. 16 is a graph in which the luminance before the conversion and the luminance after the conversion correspond to the S-shaped curve by the γ correction.

FIG. 17 is a graph in a case where γ correction of γ <1 is performed in a high-luminance area.

FIG. 18 is a graph in the case where γ correction of γ> 1 is applied in a region on the low luminance side.

FIG. 19 is a graph in which the luminance before conversion and the luminance after conversion are in linear conversion and correspond to an S-shaped curve.

FIG. 20 is a modified example in which the conversion is performed smoothly at the conversion point of the conversion characteristic in the same correspondence.

FIG. 21 is a diagram illustrating a luminance distribution when a median is shifted to a low luminance side.

FIG. 22 is a diagram illustrating a luminance distribution when a median is shifted to a high luminance side.

FIG. 23 is a graph in the case where γ correction of γ <1 is applied in all regions.

FIG. 24 is a graph in the case where γ correction of γ> 1 is applied in all regions.

FIG. 25 is a graph in a case where specified conversion points are connected by an interpolation method.

FIG. 26 is a diagram illustrating a conversion table when expanding a luminance distribution.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Image input device 11 ... Scanner 11b ... Scanner 12 ... Digital still camera 12a ... Digital still camera 12b ... Digital still camera 13b ... Modem 20 ... Image processing device 21 ... Computer 22 ... Hard disk 30 ... Image output device 31 ... Printer 31a ... Printer 31b ... Printer 32 ... Display 32a ... Display

Claims (16)

[Claims] An image processing apparatus according to claim 1, wherein the luminance equivalent value of a predetermined gradation in each pixel unit in the dot matrix image data is inputted, subjected to a predetermined conversion process, and outputted, so that the input and the output are equivalent. An image processing apparatus that performs conversion for enhancing contrast in a relation, wherein a range of a large distribution density is calculated based on the spread amount of the brightness distribution by obtaining a brightness distribution in the image data. An image processing apparatus, wherein a small number of gradations are assigned to a range where the distribution density is small while giving a large number of gradations to the image. 2. The image processing apparatus according to claim 1, wherein the spread amount of the luminance distribution uses a value corresponding to a standard deviation, and the input / output ratio is reduced when the standard deviation of the luminance distribution is large. While making the number of gradations uniform,
An image processing apparatus characterized in that when the standard deviation is small, the input / output ratio is increased to enlarge the change in the assignment of the number of gradations. 3. The image processing apparatus according to claim 2, wherein a value of γ is set so as to perform γ correction on the luminance distribution and reduce a change amount due to γ correction when the standard deviation is large. An image processing apparatus characterized in that the value of γ is set so that the amount of change in γ correction becomes large when is small. 4. The image processing apparatus according to claim 3, wherein the approximate center position of the luminance distribution is determined, and the polarity of the luminance conversion is inverted between the high luminance side and the low luminance side based on the approximate central position. An image processing apparatus for performing approximate luminance conversion of an S-shaped curve based on a relationship between an input and an output. 5. The image processing apparatus according to claim 1, wherein the luminance of the conversion destination is calculated and stored within the range of the luminance of the conversion source, and this correspondence relation is determined at the time of conversion. An image processing apparatus for awakening and converting. 6. The image processing apparatus according to claim 1, wherein, when the image data is represented by a plurality of component values corresponding to the luminance, the luminance calculation is performed by linear addition of the same component values. An image processing apparatus characterized in that it is obtained. 7. The image processing apparatus according to claim 1, wherein a luminance distribution is obtained by performing thinning-out corresponding to a predetermined extraction rate on the image data. 8. The image processing apparatus according to claim 7, wherein a predetermined number of extractions is secured on a shorter side in the vertical and horizontal directions. 9. The image processing apparatus according to claim 1, wherein a limit is set on a degree of enhancement of contrast. 10. The image processing apparatus according to claim 1, wherein the binary image data is determined based on the luminance distribution, and the contrast is not enhanced if the binary image data is binary image data. Characteristic image processing device. 11. The image processing apparatus according to claim 10, wherein when the luminance distribution is concentrated at both ends within a reproducible range, the image data is determined to be monochrome binary image data. Image processing device. 12. The image processing apparatus according to claim 1, wherein a frame portion of the image data is determined based on a prominent luminance distribution, and if there is a frame portion, the data of the frame portion is converted into a contrast data. An image processing apparatus characterized in that it is not used for enhancement. 13. The image processing apparatus according to claim 12, wherein a luminance distribution concentrated on an end portion within a reproducible range is determined to be a frame portion. 14. The image processing apparatus according to claim 1, wherein contrast enhancement is not performed when the image data is not a natural image. 15. The image processing apparatus according to claim 14, further comprising a natural image determining unit that determines that the image data is not a natural image when the luminance distribution exists in a spectrum. apparatus. 16. With respect to a luminance equivalent value of each pixel in dot matrix image data, the same luminance equivalent value is input, subjected to a predetermined conversion process, and output.
In performing the approximate S-curve conversion based on the relationship between input and output, the luminance distribution in the image data is obtained and the spread amount of the luminance distribution is calculated. An image processing method characterized by assigning a small number of gradations to a range where the distribution density is small while giving the number of gradations.