JP2006277044A - Image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing method capable of correcting gradation well without changing the density of a principal area even after correcting the brightness of an image. <P>SOLUTION: The principal area is extracted according to input image data. Next, a density correction process is carried out on the input image data using a predetermined amount of density correction. The density of the extracted principal area after the density correction process is calculated and set as a reference density value. Since image data corresponding to the density reference value does not change in gradation, other image data are subjected to gradation conversion. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image processing method for digital image data, and more particularly to an image processing method for performing gradation correction processing on image data subjected to density correction processing.

  In order to print digital image data obtained by photographing with a digital still camera (DSC) or digital image data obtained by reading a photographic film with a scanner or display it on a monitor with appropriate image quality These digital image data are subjected to various image processing. For example, in the case of an image obtained by shooting a scene in a dark environment or a backlit scene, the entire image becomes dark, so the digital image data of the image depends on the shooting conditions, the brightness of the subject, etc. Then, image brightness adjustment processing (density correction processing) is performed. In addition, for example, gradation correction processing for increasing the contrast is performed for an image with low contrast, and gradation correction processing for decreasing the contrast is performed for an image with high contrast. In this way, an image with appropriate image quality can be obtained by performing image processing on the digital image data in accordance with the image.

As described above, various methods are known as methods for controlling gradation according to an image. For example, in Patent Document 1, a shooting scene of image data is estimated based on shooting information given to image data and main part information of an image represented by the image data, and gradation processing conditions are set according to the shooting scene. Is set, and image processing is performed on the image data in accordance with the set gradation processing conditions.
As a result, it is possible to set the gradation processing condition suitable for the shooting scene with higher accuracy than the case of setting the gradation processing condition using only the shooting information, and as a result, a high-quality image can be obtained regardless of the shooting scene. It is assumed that processed image data that can be reproduced can be obtained.

  Japanese Patent Laid-Open No. 2004-26853 discloses a scene obtained by subjecting captured image data having a reflection density of 0.7 to a normalization process for each type of imaging device having a reflection density of 0.6 to 0.8 on an output medium. A step of creating reference image data, a step of optimizing predetermined image processing conditions using the scene reference image data, and storing the optimized predetermined image processing conditions for each type of imaging device And the image data obtained by applying the optimized image processing conditions, the average value of the image data becomes a reflection density of 0.6 to 0.8 on the output medium, and the leg portion (shadow portion) And a step of performing gradation correction that is set so that γ of the shoulder portion (highlight portion) is smaller than γ of the abdomen, and outputting the image. This solves the problem of gradation adjustment errors caused by the subject when performing image processing that eliminates differences between models for each imaging device, and not only matches shadows to highlights with high accuracy, but also colors. It is said that high-quality photo prints can be provided by eliminating differences between models such as reproducibility and sharpness.

JP 2003-244620 A JP 2004-88345 A

  By the way, when the gradation correction process is performed after the brightness correction (density correction process) is performed on the image data, the result of the brightness correction may not be appropriate. For example, even if the density of the face area of the main part (main subject) is changed to an appropriate density by density correction processing, if the tone correction processing is performed on the image data after density correction, the obtained image is too bright. Sometimes it was too dark or too dark.

  The present invention has been made to solve the above inconveniences, and the object of the present invention is to achieve a satisfactory level without changing the density of the main area even after image brightness correction. An object of the present invention is to provide an image processing method capable of correcting a tone.

  In order to solve the above problems, the present invention provides an image processing method including a density correction process for correcting density of image data and a gradation correction process for correcting gradation, wherein the density correction process includes: An image processing method for specifying a main area from the image data, calculating a density reference value after density correction processing of the main area, and performing the gradation correction processing without changing the density reference value I will provide a.

  In the gradation correction process, it is preferable to automatically calculate a necessary gradation correction amount by analyzing the image data. Alternatively, it is preferable to manually set the gradation correction amount in the gradation correction processing.

  In the gradation correction process, it is preferable that the gradation correction amount is controlled on the shadow side and the highlight side with the density reference value as a boundary. Further, it is preferable that the main area is a face area of a subject, and an average density of the face area is set to the density reference value. In addition, it is preferable that an upper limit value and a lower limit value of the density reference value are set in advance.

  According to the image processing method of the present invention, even after the brightness of the image is corrected by the density correction process, the gradation can be corrected satisfactorily without changing the density of the main area. Thus, an image in which the density of the main area is appropriately corrected can be obtained.

  Hereinafter, an image processing method of the present invention will be described in detail based on a preferred embodiment shown in the accompanying drawings.

  FIG. 1 shows an image output system 10 including an image processing apparatus that implements the image processing method of the present invention. The image output system 10 includes an image input device 12, an image processing device 14, and an image output device 16. An example of such an image output system 10 is a digital photo print system.

  In FIG. 1, the image input device 12 can be configured using any device as long as it is a device that acquires an image as digital data. For example, a digital still camera, a digital video camera, a scanner, or image data It is configured using a reading device or the like that reads image data from a recorded recording medium. For example, when the image output system 10 is configured as a digital print system, the image input device 12 can be configured as a scanner that photoelectrically reads an image recorded on a photographic film. In such a scanner, reading light emitted from a light source is incident on a film to obtain projection light that carries an image photographed on the film, and this projection light is applied to an image sensor such as a CCD sensor by an imaging lens. An image can be read by forming an image and performing photoelectric conversion. The read image data is subjected to various types of image processing as necessary, and then sent to the image processing apparatus as film image data. Here, in such a scanner, the frames are fed by a carrier mounted on the scanner, whereby the images of the frames shot on the film are sequentially read one by one.

In FIG. 1, an image output device 16 is a device for outputting the image data image-processed by the image processing device 14 in a visible form as a visible image. Examples of such an image output device 16 include a printer and a display.
For example, when the image output system 10 is configured as a digital print system, the image output device 16 scans and exposes a photosensitive material in accordance with image data output from the image processing device 14 and performs development processing to obtain a print. It can be configured as a printer (image recording apparatus).

  The image output device 16 may be a device for converting image data into a format corresponding to output to a printer or display and outputting the converted image data to a recording medium.

The image processing device 14 is a device for performing various image processing on the image data output from the image input device 12 so that an appropriate image corresponding to the image output device 16 is output.
In addition to image data of an image captured by a digital camera or the like, and image data of an image read by a scanner, data related to an image may be input to the image processing device 14. For example, exposure time, shutter speed, aperture value, ISO sensitivity, brightness value, subject distance range, light source, presence / absence of strobe emission, white balance, zoom magnification, type of shooting scene, amount of reflected light from strobe light source, shooting saturation, etc. Data such as information relating to shooting condition settings and information relating to the type of subject may be input. The image processing apparatus 14 can execute various types of image processing using these pieces of information.

  In FIG. 1, the image processing apparatus 14 includes a density correction processing unit 22 for correcting density, and a tone correction processing unit 24 that performs tone correction processing on image data whose density has been corrected by the density correction processing unit 22. It has. In FIG. 1, only the density correction processing unit 22 and the gradation correction processing unit 24 are shown as the processing units constituting the image processing device 14, but the image processing device 14 includes various processing units other than these processing units. A processing unit that executes image processing can be provided. For example, the image data output from the image input device 12 is subjected to log conversion by a data correction unit that performs predetermined correction such as DC offset correction, dark correction, and shading correction, an LUT (lookup table), and the like, and is converted into a digital image. A log conversion unit for (density) data, an enlargement / reduction unit that enlarges / reduces image data to a size (number of pixels) corresponding to output, and an image correction unit that performs predetermined image correction such as sharpness processing Further, a processing unit that executes various processes can be provided. The image processing apparatus 14 can execute these image processes as necessary.

  In FIG. 1, the density correction processing unit 22 can adjust the entire image to an appropriate brightness by converting input image data using a predetermined density correction LUT (lookup table). The gradation correction processing unit 24 can appropriately control the contrast by performing gradation correction on the data whose density has been corrected by the density correction processing unit 22 using the gradation correction LUT. Hereinafter, the processing of the density correction processing unit 22 and the gradation correction processing unit 24 will be described in detail with reference to FIGS.

  FIG. 2 is a flowchart showing a processing flow of the density correction processing unit 22 and the gradation correction processing unit 24. FIG. 3 is a diagram for explaining a method of creating a gradation correction LUT used in the gradation correction processing unit 24.

  First, after image data is input (step S1), the density correction processing unit 22 uses the image data to extract a main part (main subject) area (hereinafter referred to as a main part area) in the image. Is executed (step S2). As a method for extracting the main part region, an already known extraction method (specific subject extraction method) can be used. Examples of the extracted main area include a person, a person's face, a building, and the like.

Next, the density correction processing unit 22 performs density correction processing for adjusting the brightness of the image with respect to the input image data (step S3). In the density correction process, for example, when the image data acquired by the image input device 12 in FIG. 1 is output as a visible image by the image output device 14, the entire output image or the subject included in the image is displayed. A density correction amount (density correction LUT) is set according to the brightness of the image, and the input image data is converted based on the set density correction amount. The density correction amount is set for each image to be processed, and density correction processing is performed with the density correction amount set for each image. The method for setting the density correction amount is not particularly limited. For example, the density correction amount may be automatically calculated by analyzing image data. As a method of calculating the density correction amount by analyzing the image data, for example, it can be performed as follows. The density correction amount is calculated so that the area extracted as the main area is finished to an appropriate density. At this time, since the appropriate density of the main area varies depending on the shooting conditions such as the flash scene and the backlight scene, various image feature amounts are calculated from the image data. Then, a scene is estimated based on these image feature amounts, and a density correction amount is calculated so as to be finished to a density corresponding to the scene.
Also, the image before density correction is displayed on an image display device such as a display, the density correction amount is manually adjusted to various values, and the density correction amount when an image with appropriate brightness is obtained is input. May be set.

  Next, the density correction processing unit 22 calculates the density of the main part region extracted by the main part extraction process, and sets the density as a density reference value (step S4). For example, when the main area is a face area, the average density of the face area can be calculated, and the calculated average density can be set as the density reference value. This density reference value is output to the gradation correction processing unit 24 together with the image data after the density correction process.

  In the present invention, a limit range such as an upper limit value or a lower limit value may be set in advance for the density reference value. In this way, when the limit range is set for the density reference value, when the calculated density of the main region exceeds the preset upper limit value or lower limit value, the upper limit value or lower limit value is set as the density reference value. Can be set as a value. In such a limited range, for example, when the main region is a face, it is known that the appropriate density is generally 0.4 to 0.5. Therefore, in this case, the upper limit value and the lower limit value may be set so as to include the range of the appropriate concentration. For example, the upper limit value may be set to 0.6 and the lower limit value may be set to 0.3. By doing so, it is possible to suppress the change in the density of the main region even if the gradation correction process is performed.

  Next, the gradation correction processing unit 24 performs gradation correction processing on the image data after the density correction processing so that the set density reference value is not changed (step S5). In the gradation correction process, a gradation correction LUT is created so that the set density reference value is not changed by the gradation correction process. A method of creating such a gradation correction LUT will be described later. In the gradation correction process, image data is analyzed, and a gradation correction amount necessary for the gradation correction process of the image is automatically obtained. For example, if the image has a low overall contrast, a tone correction amount that makes the overall tone harder is calculated. Conversely, if the image has a high contrast, the overall tone is softened. A gradation correction amount is calculated. Also, depending on the image, it may be better to increase or decrease only the gradation on the shadow side. Similarly, it may be better to harden or soften only the gradation on the highlight side. These determinations are made by image analysis, and the gradation correction amount is automatically obtained.

The appropriate density of the main subject varies depending on the shooting scene. For example, the main subject in a backlight scene tends to be dark and the main subject in a strobe scene tends to be bright. Women tend to be brighter and blacks tend to be darker. In the present invention, an appropriate density correction amount of the main subject is obtained by analyzing the image data, and further, the gradation of the image is controlled so that the density of the subject that has been properly adjusted is not changed by the gradation correction processing. be able to. The tone-corrected image data obtained in this way is output to, for example, another image processing unit or an image output device (step S6).
As described above, according to the image processing method of the present invention, desired gradation correction can be performed while maintaining the density of these main subjects appropriately.

  In the above example, the main part region is extracted based on the image data that has not been subjected to the density correction process. However, the main part area may be extracted based on the image data after the density correction process. In this case, it is not necessary to extract the main area in the density correction processing unit 22, and the main area may be extracted upstream of the density correction process in the flow of image data.

  Here, an example of a method for creating a gradation correction LUT in which the density reference value is not changed by the gradation correction processing will be described with reference to FIG. In the following description, an example of a process for increasing the density as the density correction process will be described.

The first quadrant in Fig. 3, density correction line M ₁ to density correction as the image density increases represented by the input image data is shown. The vertical axis is image data before the density correction process, and the horizontal axis is image data after the density correction process. Density correction straight lines M ₁ is obtained by translating the linear M ₀ shown by a broken line outputs without changing any density of an image represented by the input image data. Such density correction according to the density correction line _{M 1,} for example, LUT (density correction LUT) is used. Density correction amount of the density correction straight line M ₁ may be automatically created based on the input image data may be created by externally input.

From the image data after the density-corrected density correction using a density correction straight lines M _1, determine the concentration of the main portion region extracted by such specific object extraction processing (e.g., average density). Then, the density is set as a density reference value. In FIG. 3, the density reference value in the image data after density correction is indicated by P1. The density reference value P1 is specified for each image data to be image processed.

Using this density reference value P1, a gradation correction LUT is created as follows.
The second quadrant of FIG. 3, the gradation correction line N ₁ for correcting the gradation of an image represented by the input image data is shown. The vertical axis represents image data before gradation correction processing (image data after density correction), and the horizontal axis represents image data after gradation correction processing. Gradation correction line N _1, the concentration reference value P1 is created so as not to be converted to a different density value by the tone conversion. For example, in the second quadrant of FIG. 3, on the gradation correction line N ₀ such that the output without changing any tone of the image, corresponding to the density reference value P1 of the input image data (image data before the density correction) The point P2 to be specified is specified. Then, the gradient (gradation correction amount) of the gradation correction line N ₀ is changed with the point P2 as the center (pivot point). The straight line thus obtained is identified as the gradation conversion line N _1. Then, LUT as the input image data is gradation conversion in accordance with such gradation conversion line N ₁ is a gradation correction LUT. By using such a gradation correction LUT, the image data at the density reference value is not converted after the gradation correction process, and only the image data other than the image data at the density reference value is converted. The gradation correction amount (slope) is automatically calculated by analyzing image data before or after density correction.

Here, the gradation conversion line N ₁ gradient showed a constant linear, not limited to this, without being density reference value is fixed also converted, image data other than the concentration reference value is converted stepwise Such a straight line or a curved line may be used. For example, a gradation conversion line or a gradation conversion curve that controls the gradation correction amount separately on the highlight side and the shadow side from the threshold value using the density reference value P1 as a threshold value may be used.
Here, tone correction is controlled so that the density reference value of the main area is not converted, but tone correction can also be controlled so that a predetermined density reference range including the density reference value is not converted.

  In this embodiment, the gradation correction amount is automatically obtained based on the image data. However, the gradation correction amount can be manually set. In the tone correction processing, for example, depending on the image, the shadow tone is softened in order to obtain more shadow details (details), or the highlight tone is softened to suppress highlight skipping. Is done. In such gradation correction processing, the image data after density correction is displayed as a visible image on a display device such as a display, and the gradation correction amount of the entire image can be manually set based on the displayed image. it can. In the present invention, when the gradation correction amount is manually set, the gradation correction amount range that can be manually set is set so that the density reference value of the main region obtained by the above-described method is not changed. Can be limited. As described above, even when the range of the gradation correction amount is limited, the gradation correction amount can be set separately on the shadow side and the highlight side using the density reference value as a threshold value.

  Conventionally, when manually setting the gradation correction amount in the gradation correction processing, when the gradation correction processing is executed with the set gradation correction amount, the main part represented by the appropriate density by the density correction processing is appropriate. However, in the present invention, it is very difficult to set the gradation correction amount while maintaining the density of the main part properly. By limiting the range of gradation correction amount that can be manually set so that the density reference value of the partial area is not changed, gradation correction can be easily performed without affecting the density of the main part. it can.

  As a method for manually setting the gradation correction amount, for example, the density reference value or the density reference range of the main area is set as a threshold value, and the shadow side image data and the highlight side image data with respect to the threshold value, γ that is used for γ correction is set separately, and the density reference value or density reference range of the main area is fixed without change, and each γ is increased or decreased to increase or decrease the shadow side and highlight side. A method of individually setting the gradation correction amount can be used. Also, create multiple gradation conversion lines or gradation conversion curves for each image data so that the density reference value of the main area is not changed, and set either of these gradation conversion lines or gradation conversion curves manually. The gradation correction may be executed based on the set gradation conversion line or gradation conversion curve. It is also possible to manually change the slope of such a gradation conversion line and the curve shape of the gradation conversion curve. In this way, by manually setting the gradation correction amount without changing the density of the main part region, it is possible to adjust the gradation with higher accuracy while maintaining the appropriate density of the main part. A high-quality image can be obtained.

  The image processing method of the present invention has been described in detail above. However, the present invention is not limited to the above embodiment, and various improvements and modifications may be made without departing from the gist of the present invention. Of course.

1 is a block diagram of an image output system including an image processing apparatus that implements an image processing method of the present invention. It is a flowchart which shows the flow of a process of the image processing method of this invention. It is a figure for demonstrating an example of the preparation method of the gradation correction LUT used for the gradation correction process of the image processing method of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Image output system 12 Image input apparatus 14 Image processing apparatus 16 Image output apparatus 22 Density correction process part 24 Gradation correction process part

Claims (6)

An image processing method including a density correction process for correcting density for image data and a gradation correction process for correcting gradation,
The main area is identified from the image data during the density correction process, and the density reference value after the density correction process of the main area is calculated.
An image processing method for performing the gradation correction processing so as not to change the density reference value.   The image processing method according to claim 1, wherein the gradation correction processing automatically calculates a necessary gradation correction amount by analyzing the image data.   The image processing method according to claim 1, wherein a gradation correction amount in the gradation correction process is manually set.   The image processing method according to claim 1, wherein the gradation correction processing controls a gradation correction amount on a shadow side and a highlight side with the density reference value as a boundary.   The image processing method according to claim 1, wherein the main area is a face area of a subject, and an average density of the face area is set as the density reference value.   The image processing method according to claim 1, wherein an upper limit value and a lower limit value of the density reference value are set in advance.

Images