JP2000134467A - Image processing method, unit and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To conduct image processing suitable for a backlight scene by creating a histogram of an object image, discriminating whether or not the object image is a backlight image based on the created histogram, and setting an image processing condition based on the result of discrimination. SOLUTION: The number of obtained peak areas is counted (S2). Whether or not the number of the peak areas is one is decided (S3). When it is decided that the number of peak areas is not one, there is possibility of a backlight or the like. A mean value I1 of peak values of the peak areas is obtained (S5). Whether or not an image is photographed with standard exposure is judged by using a highlight point(HL) and a shadow point(SD) obtained from a luminance histogram and the mean value I1 (S13). When the conditions above are satisfied, the image is discriminated to be an 'image with standard exposure' (S9). The image not satisfying the conditions is judged to be an 'image in other exposure' such as a backlight scene photographed without using a strobe (S11).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image processing method, apparatus, and recording medium for performing image processing according to a histogram of image data.

[0002]

2. Description of the Related Art In recent years, by increasing the speed of computer CPUs and buses and increasing the capacity of recording media such as memories, hard disks, and magneto-optical disks, it has become possible to handle digital images with many pixels and many gradations. It has become.

Along with this, digital cameras and film scanners as digital image input means have become widespread, and digital images captured by a digital camera and digital images obtained by digitizing a conventional silver halide photograph with a film scanner have been converted to conventional digital images. It is increasingly used for viewing purposes similar to silver halide photography.

[0004] A photographic digital input image as an alternative to the conventional silver halide photography may not be properly exposed for the following reasons.

[0005] For example, taking the photographing with a normal camera as an example, if the exposure is not properly selected, the whole image becomes too bright or the whole image becomes too dark, so that it is difficult to distinguish the details of the subject. In some cases, overexposure or underexposure may occur.

[0006] Further, since a digital camera captures an image using a CCD element, there are cases in which a wavelength such as infrared light which cannot be sensed by human eyes is captured. If the exposure is corrected by visual recognition, it may be possible that an overexposed state occurs by taking in the invisible infrared light even if the exposure is properly selected in the visible light region. Of course, processing such as an infrared cut filter is also performed, but it is not necessarily perfect, so the above situation may occur.

[0007] Further, even if the exposure itself is appropriate, it may appear that the exposure is poor. As a cause thereof, it is known that a human visually recognizes an object while correcting (adapting) the difference in luminance of the surrounding environment. Since the camera records the difference in brightness as it is on a film or a recording medium such as a memory card, even if the camera accurately reproduces the brightness,
When a human compares an image in its memory with a photograph taken by a camera, it may appear as a result of poor exposure.

[0008] Therefore, it is necessary to correct the poor exposure in the digital image according to the exposure state of the image so as to approach human recognition and to convert the image into an image having an appropriate halftone. It is necessary to improve.

The underexposed and overexposed photographic images have a luminance distribution biased toward a low luminance region or a high luminance region, for example, as shown in a luminance histogram of an overexposed image shown in FIG. Conventionally, overexposure and underexposure determination is performed using a reference value considered to represent the distribution state of a luminance histogram.

For example, Japanese Patent Laying-Open No. 09-068764 proposes a method of performing overexposure and underexposure determination using at least one of a reference maximum value, a reference minimum value, and an average value to correct an image. .

[0011]

SUMMARY OF THE INVENTION Using the conventional method,
Examples of images that are difficult to determine appropriately when trying to determine overexposure and underexposure include a person image and a backlit scene image captured using a nighttime flash.

For example, a luminance histogram of a person image generally looks like a luminance histogram when a person is photographed using a strobe at night in FIG. 13, but the reference maximum value or the reference minimum value described in the above publication is disclosed. In the method of comparing with the reference threshold value, the human image is determined to be underexposed in the overexposure and underexposure determination determination. Similarly, it is often determined that an overexposed scene image is overexposed.

[0013] In this regard, the above-mentioned publication describes a method of using an average value for a person image that would take into account the luminance value of the person region. However, as shown in FIG. 13, when the image area corresponding to a person who is a high-luminance area in the image is smaller than the area of the image, that is, the peak of the frequency distribution on the high-luminance side in the histogram indicates a relatively low frequency. In such a case, the average value is pulled to the peak of the frequency distribution on the low luminance side where the cumulative frequency is absolutely large and located on the low luminance side, and the exposure for the person who is the main subject was appropriate. However, it is determined that the image is underexposed.

In addition, when performing image correction on a backlight scene image, if a look-up table (LUT) used to correct the luminance value of an overexposed image is used, a person region located in a low luminance region becomes too dark. I will.

An object of the present invention is to enable appropriate image processing to be performed according to an exposure state.

It is an object of the present invention to determine an image of a backlight scene and perform image processing suitable for the backlight scene.

It is an object of the present invention to determine an image picked up by using a strobe at night and perform image processing suitable for a backlight scene.

[0018]

According to a first aspect of the present invention, there is provided an image processing method comprising the steps of: inputting image data representing a target image; creating a histogram of the target image; It is characterized in that it is determined whether or not the target image is a backlight image based on the obtained histogram, and an image processing condition is set based on the determination result.

In the image processing method according to the present invention, image data representing a target image is input, a histogram of the target image is created, and the target image is stored in a nighttime strobe based on the created histogram. It is characterized in that it is determined whether the image is an image captured by using the image processing method, and image processing conditions are set based on the determination result.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [First Embodiment] The present embodiment will be described below in detail with reference to the drawings.

FIG. 1 shows a schematic example of a system according to the present embodiment.

The host computer 100 includes a printer 106 such as an ink jet printer and a monitor 1.
05 is connected. The host computer 100
Application software 101 such as a word processor, spreadsheet, Internet browser, etc., and an operating system (OS)
stem) 102, depending on the application OS 102
The printer driver 103 which processes various rendering commands (image rendering commands, text rendering commands, graphics rendering commands) indicating output images issued to the printer driver 103, and processes various rendering commands issued by the application. And a monitor driver 104 for displaying on the monitor 106 as software.

The host computer 100 includes a central processing unit CPU 108 and a hard disk driver HD1 as various hardware on which the software can operate.
07, a random access memory 109, a read only memory ROM 110, and the like.

As an embodiment shown in FIG. 1, for example, a Windows 95 of Microsoft Corporation is used as an OS on a personal computer of an AT compatible type widely used by IBM Corporation, an arbitrary printable application is installed, and a monitor is installed. And a printer are considered as one embodiment.

In the host computer 100, based on the display image displayed on the monitor, the application 101
, Text data classified as text such as characters,
Output image data is created using graphics data classified as graphics such as figures, image image data classified as natural images, and the like. When printing out the output image data, the application 1
01 issues a print output request to the OS 102, and issues to the OS 102 a rendering command group indicating an output image composed of a graphics rendering command for the graphics data portion and an image rendering command for the image data portion. OS102
Receives an output request from the application and issues a rendering command group to the printer driver 103 corresponding to the output printer. The printer driver 103 processes a print request and a drawing command group input from the OS 102, creates print data printable by the printer 105, and transfers the print data to the printer 105.
When the printer 105 is a raster printer, the printer driver 103 sequentially performs image correction processing on a drawing command from the OS 102, and rasterizes the drawing command sequentially into a 24-bit RGB page memory. The contents of the page memory are converted into a data format printable by the printer 105, for example, CMYK data, and transferred to the printer.

The processing performed by the printer driver 103 will be described with reference to FIG.

The image correction processing unit 120 performs an image correction process on the color information included in the drawing command group input from the OS 102. In this image correction processing, the RGB color information is converted into luminance / color difference signals, exposure correction processing is performed on the luminance signals, and the corrected luminance / color difference signals are inversely converted into RGB color information. The printer correction processing unit 121 first rasterizes a drawing command using the color information subjected to the image correction processing, and generates a raster image on an RGB 24-bit page memory. Then, CMYK data depending on the color reproducibility of the printer is generated for each pixel and transferred to the printer 105.

Next, the procedure relating to the exposure correction processing performed by the image correction processing section will be described with reference to the flowchart shown in FIG.

The exposure correction process is performed on the image data portion relating to the same image indicated by the image drawing command. Therefore, for example, when a graphics image and an image image are included in the output image, an image image portion related to the same image is extracted and exposure correction processing is performed.

In S32, luminance data is created from RGB image data constituting each pixel value for all pixels in the image,
The frequency is sequentially counted based on the luminance value, and a luminance histogram is created.

When creating a luminance histogram, it is not always necessary to examine the luminance values of all the pixels constituting the image to create a luminance histogram, and it is also possible to appropriately select a pixel whose luminance value is to be examined. .

In step S33, the highlight point (HL) of the image is determined in step S33 based on the created luminance histogram.
And determine the shadow point (SD).

Next, the details of the method for determining the highlight point and the shadow point will be specifically described.

For example, in Japanese Patent Application Laid-Open No. 60-57594, a cumulative frequency histogram is created for a luminance signal obtained by weighting and adding each of the color signals R, G, and B of an input signal, and a predetermined cumulative frequency is set in the cumulative frequency histogram. For example, a method has been proposed in which an upper limit value of a luminance signal corresponding to 1% and 99% is determined as a highlight point, and a lower limit value is determined as a shadow point.

In the present embodiment, for example, with respect to image data having 350,000 pixels, 3500 corresponding to about 1% thereof
Are set as threshold values, and points at which the cumulative luminance value reaches the threshold value from the ends of the luminance value 0 and the luminance value 255 toward the center are defined as shadow points and highlight points, respectively.

Specifically, when the frequency of the pixel having the luminance Yn is set to n, the cumulative frequency is calculated as n0 + n1 +..., And the luminance value (Yk) when the cumulative frequency exceeds 3500 is calculated as the shadow point. The luminance value (Yk) is used.

In this embodiment, the luminance value is 10 and the luminance value is 24.
Although the cumulative frequency has been obtained from the five luminance positions, it may have a predetermined offset, for example, from the luminance value 1 and the luminance value 254.

In S34, a process is performed on the obtained luminance histogram to smooth the shape of the luminance histogram, for example, by taking a moving average, and to improve the accuracy of the determination described later.

Using the luminance histogram obtained above,
An image exposure determination process and an exposure correction process according to the determination result are performed.

First, a peak of a luminance histogram required for exposure determination processing and exposure correction processing to be described later is detected (S35).

(Detection of Peak) The method of detecting the peak of the luminance histogram (S35) will be described with reference to the flowchart of the processing for detecting the peak of the luminance histogram in FIG.

First, in S82, a threshold value Th_f for the frequency is set for the obtained luminance histogram, and a luminance value region having a frequency exceeding this threshold value is regarded as a region where the peak of the luminance histogram exists (A, FIG. 14). B, C).

Here, for example, (the total number of pixels constituting the luminance histogram) / (HL-SD + 1) can be used as the threshold Th_f.

In the present embodiment, a 256-width labeling table (one-dimensional array) is prepared, and when the frequency of the luminance histogram exceeds the threshold value Th_f, a label, for example, from the low luminance side.
If it is the first peak area, start to label 1.
This is achieved by successively shifting the same label to the corresponding position on the table while shifting the luminance value to the higher luminance side.

(If histogram [j]> Th_f then label_table [j] = 1)

That is, for example, referring to FIG. 14, labels 1, 2, and 3 are assigned to the peak areas A, B, and C in FIG. 14, respectively. A number string having a length corresponding to the width of each area is written.

(Label_table [] = {0,0,0,1,1,1,1,1,0,0,0,2,2,0,0,0,3,3,3 ...)

However, in the method for detecting the peak area, FIG.
In the case of a distribution like the luminance histogram shown in FIG. 14, a region that is not so important, that is, a small mountain (B in FIG. 14) that cannot be said to represent the characteristics of the entire image is detected. Therefore, in the present embodiment, in S83, which is a process of detecting the peak of the luminance histogram, when each class value is compared with the threshold value Th_f, the cumulative frequency of the region that simultaneously saw the threshold value, that is, the region Count the area inside. Then, the cumulative frequency is equal to the threshold Th_a
The following regions are excluded from the region where the peak of the luminance histogram exists (FIG. 15).

The threshold value Th_a is a threshold value for removing small hills that cannot be said to represent the characteristics of the entire image, and is set according to the characteristics of the image. In the present embodiment, a value twice the threshold Th_f is used.

In the present embodiment, a peak elimination method uses an equivalent table method. In advance, an equivalent table (equiv_table [j] =) storing a label value, for example, label value 3 at a position corresponding to 3, and 4 at a position corresponding to 4
Prepare j). Next, the value of the table corresponding to the label value to be excluded on the equivalence table is set to 0. For example, in FIG. 14, equiv_table [2] = 0 since B (label 2) is a label value to be excluded. Finally, it can be performed by rewriting the label value of the labeling table using the equivalent table. For example, using a group of tables, tmp = label_table [j], lable_table [j]
= equiv_table [tmp], rewrite the value from 0 to 255.

Further, in the method using the threshold value Th_f for the luminance value and the threshold value Th_a for the cumulative frequency of the luminance value, no peak of the luminance histogram may be obtained depending on the shape of the luminance histogram. In such a case, for example, instead of Th_f, Th_f ′ = 0.5 * Th_
By repeating the peak detection method using f and sequentially lowering the threshold, the peak of the luminance histogram can always be found.

The brightness values at both ends of the obtained peak area are set as characteristic points Pk (k = 1, 2, 3,..., Max) of the brightness histogram, and the following processing is performed.

(Exposure Judgment Processing) Exposure judgment processing S36 is performed using the luminance values Pk at both ends of each of the plurality of peaks obtained in S35.

In the exposure determination processing of this embodiment, the input image can be classified into overexposure S8, underexposure S10, standard exposure S9, and other exposure states S11 such as backlight.

Hereinafter, a flowchart of the image processing performed by the exposure determination processing section will be described with reference to FIG.

First, the number of obtained peak areas is counted.
(S2). Determine whether the number of peak areas is one (S
3).

Determination of other exposure state images such as backlight
If it is determined that there are two or more peak areas determined in S3, the image may be a backlight scene, a person image using a nighttime strobe, or the like. Therefore,
If it is determined in S3 that the number of peak areas is not one, there is a possibility that the image is another exposure state image such as backlight.

If an image of standard exposure, in which the luminance histogram expresses halftones richly as a whole, contains a bright sky or black clothing, a peak also occurs in a high luminance portion and a low luminance portion. In some cases, a plurality of peak areas may occur in the same luminance histogram. It is not preferable to perform exposure correction on a standard exposure image that expresses halftones richly in the same manner as a backlight scene image or a person image using a nighttime strobe.

Therefore, the following determination processing is performed to detect a standard exposure image, and control is performed so as to perform exposure correction suitable for the standard exposure image.

In S5, the average value I1 of P1 and P2 is obtained. The following processing is performed on the assumption that this average value I1 is the peak value of the peak area defined by P1 and P2.

In this embodiment, it is determined whether or not the image is a standard exposure image by using the highlight point (HL) and the shadow point (SD) obtained in S33 and the average value I1 obtained in S5. This determination prevents an image of standard exposure having a plurality of peak areas from being erroneously determined to be a backlight image.

More specifically, there is a sufficient width between the shadow point (SD) and the highlight point (HL). For example, in the present embodiment, the width exceeds 160 and the peak defined by P1 and P2. The region is located substantially at the center of the luminance range of 0 to 255, for example, in this embodiment, the average value I1 exceeds 50
On condition that the value is less than 150, when the condition is satisfied, the image is determined to be a “standard exposure image” (S9).

For an image which does not satisfy this condition, the image is captured without using a strobe having a luminance distribution such as a luminance histogram of a backlight image shown in FIG. 11, or as shown in FIG. It is determined that the image is an “other exposure state image” having a different exposure state depending on a region in the image, such as a person image using a nighttime strobe having a luminance distribution like a luminance histogram (S11).

Judgment of underexposure and overexposure If it is determined that the number of peak areas obtained in S3 is one, the image may be a backlight scene or a person image using a nighttime strobe. There is.

Therefore, in S6, it is determined whether the input image is overexposed, and then in S7, the underexposure is determined.

First, in S4, an average value I1 is obtained based on P1 and P2, and an average value Imax is obtained based on Pmax- ₁ and Pmax.

For example, when the subject is directly exposed to the sun under the clear sunlight, the overall brightness is high,
An overexposed image is obtained. As can be seen from the luminance histogram shown in FIG. 10, for example, the luminance histogram of such an image has a distribution in which the luminance distribution as a whole is biased toward the high luminance side.

In this embodiment, in order to easily determine an overexposed image, when the relative relationship between the average value I1 and the threshold value Th_over for the luminance value is I1> Th_over, the input image is Assume overexposed image
(S8). In the present embodiment, for example, 190 is used as the threshold Th_over.

Next, in this embodiment, underexposure is determined in step S10 for an input image that has not been determined to be overexposed.

Even when the subject is under cloudy weather or under direct sunlight, the subject may be in a space covered with shade such as a forest, or of course, as a result of imaging using the automatic photographing function of the camera. AE was mostly due to the bright blue sky.
Under a situation where aperture correction by (auto exposure) is applied, a dark, underexposed image is obtained as a whole. The luminance histogram for the image in such a case is, for example, a luminance histogram of an underexposed image in FIG. 9 or a luminance histogram of a shaded image without using a strobe in FIG. The distribution becomes skewed.

Therefore, in the present embodiment, in order to easily determine an underexposed image, the relative relationship between Imax and the threshold value Th_under for the luminance value is expressed as Imax <Th_unde
When r, the input image is determined to be an underexposed image (S10). In the present embodiment, the threshold Th_under
For example, 60 is used.

The underexposure determination and the overexposure determination may be performed in reverse order.

Standard Exposure In the present embodiment, the image not determined to be underexposed in S7 and the image determined to be standard exposure in S13 are determined to be standard exposure (S9).

(Exposure Correction Processing) An LUT suitable for the result of the determination processing in S36 performed by the processing shown in FIG. 4 is selected or created, and the exposure correction according to the exposure state of the image is performed using the LUT. (S37).

In the present embodiment, the selected one-dimensional LUT
Is used to convert the input luminance values 0 to 255 to output luminance values 0 to 25.
Convert to 5.

In this embodiment, a fixed LUT (for example, FIG. 5 for an overexposed image) is used for an image determined to be standard exposure, overexposure, and underexposure. Suitable for each time image
Automatically generate LUT and perform exposure compensation.

More specifically, a look-up table (LUT), which is a correspondence table of output signals with respect to input signals, is prepared in advance or each time so as to be on a gradation curve, and the pixel value of the input image is used as an index. The image signal is converted by a procedure of obtaining a corresponding output signal with reference to the LUT.

An underexposed image has a halftone that is too dark as apparent from a luminance histogram such as an underexposed image luminance histogram shown in FIG. Therefore, the underexposure LUT performs exposure correction so that the halftone is brightened and the halftone gradation is good.

Further, in the overexposed image, the halftone is too bright as apparent from the luminance histogram such as the luminance histogram of the overexposed image shown in FIG. Therefore, the overexposure LUT performs exposure correction so that the halftone is darkened and the halftone gradation is improved.

Exposure correction for other exposed images such as backlighting As shown in the brightness histogram of the backlighting image in the backlighting scene image, the high brightness region is too bright, the low brightness region is too dark, and the halftone is expressed. I'm terrible.

As an image determined to be “other exposure state”, for example, there is a backlight scene image in which a person who is the main subject has a low luminance value while the background has a high luminance value.

In a backlight scene image, an overexposed background region and an underexposed main subject region are mixed, but in a backlight scene, exposure correction processing for brightening the main subject region is desired.

Therefore, in the present embodiment, as shown in FIG. 7, an LUT is created based on the luminance histogram of the image, and exposure correction is performed on the image in the “other exposure state”.

The LUT includes the calculated shadow point SD, highlight point HL, minimum luminance point P1 of the peak area indicating the minimum luminance value of the luminance histogram, and maximum luminance point of the area.
It is represented by the following equation using P2 and the average value I1 of the area.

Note that x is an input luminance value and y is an output luminance value. Here, the average value I1 is used as a substitute for the true value of the peak of the luminance histogram based on the assumption that the peak of the luminance histogram is located at the center of the obtained peak area. Here, when the true peak of the luminance histogram is not located at the center of the obtained peak region and is greatly biased toward the end of the peak region, the luminance values at both ends and the center of the peak region are manually given. Thereby, a more preferable result can be obtained.

Y = 0 (0 <= x <SD) y-0 = (P1-0) / (P1-SD) * (x-SD) where (SD <= x <= P1) y-P1 = A * (I'-P1) / (I1-P1) * (x-P1) (P1 <= x <= P2) yy (P2) = (255-y (P2)) / (HL-P2) * (x -P2) (P2 <= x <= HL) y = 0 (HL <x)

By constructing the LUT as described above and calculating x in order from the low luminance side, it is possible to obtain a gradation curve for mainly correcting the luminance value in the region having the luminance from P1 to P2. is there.

Here, I ′ uses a value obtained by the following equation.

I '= 1.5 * (I1-SD)

For example, when the luminance histogram of an image in which HL <170 is greatly biased toward the low luminance side, for example, A = A term that emphasizes the slope of 1.2 can also be applied to the above equation.

Further, in the section of (P1 <= x <= P2), the slope of the straight line is determined by the above equation. If the slope is too steep, the difference is close to 1, for example, before conversion. After conversion, the difference between the pixels having the brightness value becomes, for example, 10, and a smooth gradation change is lost. Therefore, in the present embodiment, the gradient of the gradation curve in the section (P1 <= x <= P2) determined by x = P1 is limited from 0.6 to 2.0. In addition to this, the interval of (P2 <= x <= HL) is too short due to the value of the slope, and the calculated slope is used to prevent the smooth gradation expression on the high luminance side from being lost. P2) is 220
When y exceeds (2), the inclination is adjusted so that y (P2) becomes 220.

According to the first embodiment, it is possible to easily determine the exposure state of an image.

Further, optimal exposure correction can be automatically performed according to the exposure state.

[Second Embodiment] A second embodiment of the present invention will be described.

The system used in the second embodiment (FIG. 1),
The general flow of the image processing (FIG. 3) is the same as that of the first embodiment, and therefore the description is omitted, and only the differences will be described.

(Detection of Peak) In detecting the peak of the luminance histogram, in the first embodiment of the present invention, only the region having the frequency equal to or higher than the threshold Th_f is targeted. However, it is necessary to examine the distribution of the peak in more detail. In some cases, a method of dividing the luminance value range of 0 to 255 into, for example, 16 equal parts, comparing the heights of the respective areas, and deeming a region having a higher frequency than both adjacent regions as a peak region, or the like is considered.

When a group of luminance values such that the class value of the cumulative luminance histogram is equal to a power of 2 is obtained from the cumulative luminance histogram, the area of the luminance histogram in the luminance value section divided by the luminance value group is determined. All have the same number of pixels, so the shorter the section, the higher the frequency in that section. Therefore, it is possible to detect the peak of the luminance histogram by comparing the lengths of the sections.

(Exposure Judgment) When performing the exposure judgment, for example, only the overexposure module is prepared without preparing both the overexposure judgment module and the underexposure judgment module and the threshold value for the underexposure judgment is inverted. By setting the input image to the module as the input image to the module, by using a bit-inverted version of the original image for each pixel, one module can be used for both overexposure and underexposure detection. When implemented using resources, resources can be used effectively.

Since the threshold value Th_under indicates the degree of underexposure of the image to some extent, by appropriately changing the threshold value, it is possible to further increase the degree of darkness of the image, that is, the degree of underexposure. It is possible to adjust whether an image is determined to be underexposed. For example, the adjustment may be made based on a user's manual instruction.

When judging whether an image is an underexposed image, in addition to the peak position, the cumulative frequency of the peak area having the peak, that is, the peak located on the lowest luminance side in the entire image is determined. It is possible to express the degree of underexposure by using the number of constituent pixels together. It is also possible to use the width of the peak for a similar purpose. In addition, using the threshold Th_over, the peak position located at the highest luminance value, the cumulative frequency of the peak region having the peak position, the width of the peak, and the like, the degree of overexposure can be represented as in the case of the underexposure determination. It is possible. Then, processing such as selecting a corresponding one of a plurality of underexposure correction LUTs may be performed according to the obtained degrees of underexposure and overexposure.

(Exposure Correction) In the first embodiment, image correction is performed without considering the γ characteristic of the input / output device. However, if the γ characteristic is known, the γ characteristic is considered. It is desirable to perform the exposure correction of the present invention.

When the highlight point and the shadow point are known, corresponding known values may be used as the highlight point and the shadow point.

It is also possible to change, for example, a threshold value for judging whether overexposure or underexposure is performed in conjunction with an image capturing mode when an image captured using a digital camera or a scanner is captured. is there.

It is also possible to divide the original image into appropriate regions and to perform the exposure judgment or the exposure correction process separately for each of them.

When creating a luminance histogram,
For example, in order to prevent an image in which the blue sky occupies most of the constituent pixels from being determined to be overexposed, for example, a process for weakening an image signal for blue may be performed.

In determining whether the exposure is over or under, at least one of the R, G, and B image signals, or
The sum of the weights of the R, G, and B image signals can be used as an input signal for the determination process.

Further, in the first embodiment, the average value of the detected peak area is used as the peak position of the luminance histogram, but an intermediate value or an appropriate weighted sum can be used.

In the first embodiment, a gradation curve having two break points is used as a gradation curve like the LUT shown in FIG. 7, but a gradation curve of one break point is more simply used as in the LUT of FIG. The use of a curve can reduce the amount of calculation. Alternatively, when a smooth change in the halftone is required, it is possible to smoothly connect the gradation curves using an approximate curve such as a spline curve.

In the first embodiment, a fixed LUT is used in exposure correction for images with standard exposure, underexposure, and overexposure, but the HL, SD obtained in S33 are used.
Exposure correction may be performed using an LUT created based on the above. For example, LUT is S, considering SD and HL.
It can be created by setting a curve that converts D to a luminance value of 10 and HL to a luminance value of 245.

In the first embodiment, only the mode for automatically judging the exposure state based on the histogram is provided, but a mode for judging the exposure state based on the manual instruction of the user may be provided.

(Other Embodiments) The present invention is for implementing the functions of the embodiments in an apparatus or system connected to the various devices so as to operate various devices so as to realize the functions of the above-described embodiments. The program code itself and means for supplying the program code to a computer, for example, a storage medium storing the program code constitute the present invention.

As a storage medium for storing such a program code, for example, a floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM, magnetic tape, nonvolatile memory card, ROM and the like can be used.

When the computer executes the supplied program code, not only the functions of the above-described embodiments are realized, but also the OS (operating system) running on the computer or another program code. Needless to say, the program code is also included in the embodiment of the present invention when the functions of the above-described embodiment are realized in cooperation with application software or the like.

Further, the supplied program code is stored in a memory provided in a function expansion board of a computer or a function expansion unit connected to the computer, and then stored in the function expansion board or the function storage unit based on an instruction of the program code. It is needless to say that the present invention includes a case where a provided CPU or the like performs part or all of actual processing, and the processing realizes the functions of the above-described embodiments.

Further, the above embodiments may be combined.

[0116]

According to the present invention, appropriate image processing can be performed according to the exposure state.

According to the first aspect of the present invention, it is possible to determine an image of a backlight scene and perform image processing suitable for the backlight scene.

According to the invention of claim 8 of the present application, it is possible to judge an image picked up by using a strobe at night and perform image processing suitable for a backlight scene.

[Brief description of the drawings]

FIG. 1 is a diagram showing a system configuration.

FIG. 2 is a diagram illustrating processing performed by a printer driver.

FIG. 3 is a flowchart illustrating a flow of image processing performed by an image correction processing unit.

FIG. 4 is a flowchart illustrating a flow of image processing performed by an exposure determination processing unit.

FIG. 5 is a diagram of an LUT used when overexposure is corrected for overexposure.

FIG. 6 shows LUs used for exposure compensation for other exposure states.
It is a figure of T (one break point).

FIG. 7 shows LUs used for exposure compensation for other exposure states.
It is a figure of T.

FIG. 8 is a flowchart of a process for detecting a peak of a luminance histogram.

FIG. 9 is a luminance histogram of an underexposed image.

FIG. 10 is a luminance histogram of an overexposed image.

FIG. 11 is a luminance histogram of a backlight image.

FIG. 12 is a luminance histogram of a shaded image without using a flash.

FIG. 13 is a diagram of a luminance histogram when a person is photographed using a strobe at night.

FIG. 14 is a diagram of a peak of a luminance histogram detected in a peak detection process of the luminance histogram.

FIG. 15 is a diagram of a peak from which small hills have been removed in consideration of the cumulative frequency of the peak region in the peak detection process of the luminance histogram.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Tetsuya Suwa 3-30-2 Shimomaruko, Ota-ku, Tokyo Inside Canon Inc. (72) Inventor Manabu Yamazoe 3-30-2 Shimomaruko, Ota-ku, Tokyo Canon In-house F term (reference) 5B057 AA20 BA02 BA30 CA01 CA08 CA12 CA16 CB01 CB08 CB12 CB16 CC01 CE11 CE18 CH01 DA20 DB02 DB06 DB09 DC04 DC23 DC36 5C077 LL01 MP08 PP05 PP15 PP32 PP34 PP43 PP46 PP48 PP52 PP53 PP58 PP61 PP65 PP68 PQ19 PQ20 PQ23 RR14 TT02

Claims (11)

[Claims] 1. An image data representing a target image is input, a histogram of the target image is created, and it is determined whether the target image is a backlight image based on the created histogram. Based on the determination result, An image processing method comprising setting image processing conditions. 2. The image processing method according to claim 1, wherein the image processing condition is calculated based on a peak area of the histogram. 3. The image processing method according to claim 1, further comprising determining whether the target image is underexposed or overexposed based on the generated histogram. 4. The histogram according to claim 1, wherein the histogram is created based on component data indicating brightness, and a peak area of the histogram is detected using a predetermined detection method based on the created histogram. The image processing method described in the above. 5. The image processing method according to claim 4, wherein the predetermined detection method regards an area of the histogram exceeding a predetermined frequency as a peak area. 6. The method according to claim 4, wherein the predetermined detection method divides the input value range into n equal parts, compares the frequencies of the respective regions, and regards a region having a frequency higher than both adjacent regions as a peak region. Image processing method. 7. The method according to claim 1, wherein the predetermined detection method is performed by dividing a class value of the histogram obtained for the histogram so as to be a power of 2 and comparing the lengths of the divided sections. Item 5. The image processing method according to Item 4. 8. Inputting image data indicating a target image, creating a histogram of the target image, and determining whether the target image is an image captured using a nighttime strobe based on the created histogram. An image processing method, wherein an image processing condition is set based on the determination result. 9. An input unit for inputting image data indicating a target image; a generating unit for generating a histogram of the target image; and determining whether the target image is a backlight image based on the generated histogram. Means, setting means for setting image processing conditions based on the result of the determination means, correction means for correcting the image data based on the set image processing conditions, and a recording medium based on the corrected image data An image processing apparatus comprising: an image forming unit that forms an image thereon. 10. A recording medium storing a computer-readable program, inputting image data indicating a target image, creating a histogram of the target image, and setting the target image to backlight based on the created histogram. A recording medium for determining whether the image is an image and storing a program for setting image processing conditions based on the determination result. 11. A recording medium storing a computer-readable program, inputting image data representing a target image, creating a histogram of the target image, and generating a histogram of the target image at night based on the created histogram. And a storage medium for storing a program for setting image processing conditions based on the determination result.