JP2015122616A - Image processor, image processing method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor, an image processing method, and a program capable of obtaining an image having a color based on an observation light source with low noise and high clarity and less brightness unevenness.SOLUTION: An image processor includes: first work photographic image data acquisition means for acquiring first work photographic image data 115 obtained by photographing a work by emitting a flash; second work photographic image data acquisition means for acquiring second work photographic image data 113 by photographing the work without emitting a flash; brightness data acquisition means for acquiring brightness data 202 from the first work photographic image data; chromaticity data acquisition means for acquiring chromaticity data 204 from the second work photographic image data; and synthesis means 205 for synthesizing the brightness data acquired by the brightness data acquisition means with the chromaticity data acquired by the chromaticity data acquisition means.

Description

  The present invention relates to an image processing apparatus, an image processing method, and a program that are effective when creating a copy of a cultural property, a work of art, a painting, or the like that combines a digital camera and a printer.

  In recent years, image processing technology using an ink jet printer has been improved in performance. Specifically, ink to be printed is reduced in droplet size, the head scanning method is improved, and ink is improved. Along with higher performance, the granularity and glossiness of the image are improved, and a high-quality image such as a silver salt photograph can be output. One market where such high-quality images are required is a market called fine art. In the fine art market, digital art images created by a computer are printed using the ink jet printer technology described above. Printed images are treated like paintings and traded as valuable works. Inkjet printers are also used for cultural property protection. Specifically, it creates paintings as valuable cultural assets, replicas such as folding screens, and displays and stores them at museums and schools.

  Conventionally, when creating a copy of this type, a high-performance digital camera is used to shoot the original cultural asset with high definition, and the color of the original image is repeatedly adjusted using image editing software. A duplicate image close to is output with a printer. However, color adjustment using image editing software often depends on the ability of the operator and requires advanced knowledge and skill. Therefore, color matching technology is used to match the color of the original image and the color of the duplicate image by computer processing.

  Conventionally, a color chart for color matching is photographed under the same conditions as the original image, and color conversion parameters are created based on the color data of the color chart photographed image and predetermined target color data (see Patent Document 1). If the original image captured image is color-converted and printed using the created color conversion parameter, a duplicate image color-matched with the original image can be created.

  By the way, in order to create a very high-definition duplicated image, it is necessary to illuminate the original image with sufficient illumination light quantity at the time of capturing the original image and to illuminate the original image without uneven illumination.

  If the amount of illumination light is small, the exposure time at the time of shooting must be lengthened, and electrical noise increases. In addition, if the exposure time is long, even when a digital camera having a very high number of pixels is used, slight vibration also appears as image blur. Therefore, in order to maintain the sharpness of the captured image, it is necessary to secure a sufficient amount of illumination light.

  In addition, when a work is photographed in the presence of uneven illumination, the unevenness is reflected in the photographed image and appears as lightness unevenness of the duplicate image. The work for correcting the brightness unevenness with image editing software takes time and requires high skill from the operator. Therefore, when photographing a work, it is desirable to devise lighting and shoot with as little unevenness as possible.

  For example, there is a method of ensuring the light quantity and uniformity of illumination by preparing a large number of illuminations such as artificial sun lamps and irradiating the work from a plurality of directions. However, cultural properties and paintings with high historical value are forbidden to be exposed to strong lighting in order to protect their works, and the depth of the exhibition place is narrow and there is often little freedom of equipment installation. Therefore, in practice, it is not realistic to shoot with a strong illumination such as an artificial solar light uniformly.

  Therefore, there are many cases where the flash is used for shooting with multiple lights. The flash light can obtain a sufficient amount of light with an irradiation time of a few tens of thousands to a few thousandths of a second, so it does not hurt the work and installation of equipment is easy even in a place with no depth. It is possible to illuminate uniformly without unevenness.

Due to the above background, conventionally, duplicate images are created in the following steps.
(1) Shoot the original work using the flash.
(2) Photograph a color chart using a flash.
(3) A color conversion table for color matching is created from the captured image of the color chart and the original data of the color chart.
(4) The color conversion table for color matching is applied to the work image data for color conversion.
(5) Print the color-converted work image data with a color printer.

Japanese Patent Laid-Open No. 5-103336

  However, the conventional method has the following problems. The spectrum of the flash light source used at the time of photographing differs from the spectral characteristics of the observation light source used when a person actually refers to the work. Since the color conversion table for color matching is created based on the spectral characteristics of the flash light source, the created duplicate image is color-matched with conditions only under the flash light source.

  However, it is practically impossible to observe the work with a flash light source, and there is a problem that the color of the original image does not match the color of the duplicate image when observed under the spectral characteristics of the actual observation light source.

  On the other hand, if a color chart is photographed under an observation light source and a color conversion table for color matching is created, the color of the duplicate image is color-matched under the observation light source. However, as described above, the work photographed image data photographed under the observation light source has a problem that it does not become a high-definition duplicate image because there are many noises, blurs, and illumination unevenness.

  An object of the present invention is to provide an image processing apparatus, an image processing method, and a program capable of obtaining an image having low noise, clearness, little brightness unevenness, and having color based on an observation light source.

  The image processing apparatus of the present invention includes a first work photographed image data acquisition unit that obtains first work photographed image data obtained by photographing a work by emitting a flash, and a first photographer that photographs the work without emitting a flash. From the second work photographed image data acquisition means for obtaining the second work photographed image data, from the first work photographed image data, the lightness data acquisition means for obtaining lightness data, and from the second work photographed image data Chromaticity data acquisition means for acquiring chromaticity data, lightness data acquired by the lightness data acquisition means, and combining means for combining the chromaticity data acquired by the chromaticity data acquisition means And

  According to the present invention, the work is photographed with both the flash light and the observation light, and the lightness signal of the image obtained with the flash light and the chromaticity signal of the image obtained with the observation light are combined. A clear image with little brightness unevenness and a color based on the observation light source can be obtained.

FIG. 3 is a diagram illustrating an entire duplicate image creation method according to the first embodiment. 6 is a flowchart showing an overall flow in Examples 1 and 2; 3 is a flowchart showing a flow of a preliminary preparation process in the first embodiment. 6 is a flowchart showing a flow of duplicate image generation in the first and second embodiments. FIG. 10 is a diagram illustrating a flow of color conversion parameter generation processing in Examples 1 and 2. It is a figure which shows the concept of the color conversion parameter production | generation process in Example 1,2. 3 is a flowchart illustrating a flow of work image generation processing in the first embodiment. It is a figure explaining the work image generation process in Example 1. FIG. FIG. 10 is a diagram for explaining an entire duplicate image creation method according to a second embodiment. 10 is a flowchart illustrating a flow of a preliminary preparation process in the second embodiment. 10 is a flowchart illustrating a flow of work image generation processing according to the second embodiment. It is a figure explaining the work image generation process in Example 2. FIG.

  The modes for carrying out the invention are the following examples.

  FIG. 1 is an explanatory diagram showing the overall flow of a duplicate image creation method to which the image processing method of the present invention is applied.

  In the figure, a range indicated by a dotted rectangle 100 is a processing portion executed by a computer. In the computer 100, color chart original data 101 and color chart image data 102 are stored in association with each other.

  The color chart image data 102 is data in which a plurality of color patches for generating color conversion parameters are laid out. The color chart original data 101 stores RGB values of a plurality of patches laid out in the color chart image data 102.

  FIG. 2 is a diagram showing the overall flow when creating a duplicate image.

  In step S001, the operator performs advance preparation. The advance preparation includes, for example, output of a color chart, shooting of a chart, shooting of a work, and data input to a computer.

  In step S002, the processing of the above embodiment is executed inside the computer 100 to generate image data 119 of a duplicate image. In step S003, duplicate image data is sent from the computer 100 to the printer 103, and the duplicate image 120 is printed.

  Details of each step will be described below.

  FIG. 3 is a flowchart showing the operation performed by the operator in step S001.

  In step S101, the operator operates the computer 100, transmits the color chart image data 102 to the color printer 103, and outputs the color chart 104. Here, it is desirable that the paper on which the color chart is printed and various printing conditions are the same as the conditions for printing a duplicate of the work later.

  In step S102, the color chart 104 is photographed using the digital camera 105 in a state where the operator does not use the flash. Color chart photographed image data 107 obtained by photographing is temporarily stored in the digital camera 105. In this shooting, a color chart is placed under the observation light source of the work. As a result, the color of the color chart photographed image data 104 conforms to the spectrum of the observation light source.

  In step S <b> 103, the work 111 is photographed using the digital camera 105 without using the flash. The photographed image data 113 with the flash OFF obtained by photographing is temporarily stored in the digital camera 105. In this photo, the photo is taken under the observation light source. As a result, the color of the captured image data 113 conforms to the spectrum of the observation light source. However, if the amount of light from the observation light source is insufficient, the exposure time may be long, resulting in a noisy image. It may also include some blurring.

  In step S <b> 104, the work 111 is photographed using the digital camera 106 with the operator using the flash. The photographed image data 115 with flash ON obtained by photographing is temporarily stored in the digital camera 106. The color of the captured image data 115 conforms to the flash light source. Since the flash is used, a sufficient amount of light from the observation light source can be secured and an image with less noise can be obtained. Moreover, it does not include blurring.

  The digital cameras 105 and 106 shown in FIG. 1 differ only in turning on / off the flash, and it is desirable to use the same digital camera. In FIG. 1, the digital camera 106 is a digital camera equipped with a clip ON type flash. Shooting that is difficult to receive is possible.

  In step S105, the photographed color chart photographing data 107, the photographed image data 113 when the flash is turned off, and the photographed image data 115 when the flash is turned on are input to the computer 100.

  When the digital camera 105/106 is connected to the computer 100, the image data is automatically taken into the computer 100 immediately after shooting by connecting via a USB or wireless LAN.

  Of course, a small storage medium such as a compact flash (registered trademark) card may be replaced from the digital camera 105 to the computer 100 to copy the image data to the computer 100.

  Thus, the advance preparation by the operator in step S001 is completed.

  Next, in step S002, a duplicate image generation process by the computer 100 is automatically performed. The operation of the duplicate image generation process will be described.

  FIG. 4 is a flowchart illustrating an operation of generating a duplicate image according to the first and second embodiments.

  In step S201, the computer 100 generates color conversion parameters. This color conversion parameter is a three-dimensional lookup table 109. The three-dimensional lookup table is hereinafter referred to as “color conversion 3D LUT”. This color conversion 3D LUT 109 is a look-up table for color-converting RGB values (hereinafter referred to as “camera RGB values”) captured by a camera into RGB values (hereinafter referred to as “printer RGB values”) input to a printer. is there. The color conversion 3D LUT 109 is data for color matching between a work photographed by a camera and an output material printed by a printer.

  The color conversion parameters are generated based on the color chart original data 101 and the color chart photographed image data 107.

  Next, generation of the color conversion 3D LUT 109 (step S201) will be described.

  FIG. 5 is a flowchart illustrating a flow of processing for generating a color conversion parameter in the first and second embodiments.

  FIG. 6 is a diagram illustrating a concept of processing for generating a color conversion parameter in the first and second embodiments.

  In step S301, printer RGB values and camera RGB values are acquired. The printer RGB value is the RGB value input to the printer when the printer 103 prints the color chart 102, that is, the color chart original data 101 itself.

  The color chart original data 101 is data in which RGB values of patches are recorded in association with each patch number, as shown in the table on the left side of FIG.

  The camera RGB value is an RGB value of each patch extracted from the color chart photographed image data 107. In the color chart photographed image 107, one patch is composed of a plurality of pixels. A value obtained by averaging RGB values of a plurality of pixels constituting a certain patch is set as a camera RGB value of the patch.

  By executing this process for all patches, the RGB values of all patches in the chart are obtained as camera RGB values. As a result, as shown in the table on the right side of FIG. 6A, the RGB data of the patch is acquired as the camera RGB value in association with each patch number.

  In color chart photography, since a flash is not used, the color chart photographed image 10 may have noise components or some blurring. However, since a plurality of pixels are averaged in units of patches, the effects of noise and blur can be greatly reduced. In addition, since the color chart having a size of about A4 to A3 is sufficient, the influence of illumination unevenness is small.

  As described above, even when the color chart is photographed under an observation light source, practical data can be obtained.

  In step S302, a color conversion 3D LUTα from a printer RGB value to a camera RGB value is generated. FIG. 6B shows the concept of a 3D LUT for converting printer RGB values into camera RGB values.

  Since the printer RGB values are lined up at equal intervals, the printer RGB value is used as a point of a grid point, and a 3D LUT having a camera RGB value may be created as a conversion value of the grid point. For example, if the grid points are configured so that the RGB values of the printer RGB values are 0, 32, 64,..., 255, the 3D LUT configures a 3D LUT composed of a total of 729 (= 9 × 9 × 9) grid points. be able to.

  In step S303, using the 3D LUTα created in step S302, a 3D LUTβ that reversely converts camera RGB values into printer RGB values is created.

  In step S101, the camera RGB values acquired from the color chart photographed image data 107 are not arranged at equal intervals, unlike the printer RGB values. Therefore, a conversion value corresponding to the lattice point of the 3D LUTβ is calculated using the 3D LUTα. Specifically, the 3D LUTα is interpolated to calculate a value between the lattice points, and a conversion value corresponding to the lattice point of the 3DLUTβ is obtained. With this process, conversion values corresponding to the grid points of 3DLUTβ are determined, and a 3D LUT for converting camera RGB values into printer RGB values can be created. The number of grid points of the 3D LUTβ may be any number, but it is preferable to increase the number of grid points as much as possible in order to perform color matching with high accuracy. When interpolating the 3D LUTα, various interpolation methods such as tetrahedral interpolation and cube interpolation may be used.

  When the RGB value acquired from the color chart photographed image data 107 is input to the generated 3D LUTβ and color conversion processing is performed, it is ideally the same as the color chart original data 101 as shown in FIG. RGB values are calculated. However, in practice, there may be some errors due to the influence of the interpolation calculation process.

  Through the above processing, a color conversion 3D LUT for converting camera RGB values into printer RGB values is generated, and this color conversion 3D LUT calculates the color of the subject photographed by the camera and the color of the output material output by the printer. This is a color conversion parameter for matching. However, the colors most closely match when the output is observed under the same light source used when the color chart is taken.

  The generated color conversion parameter (color conversion 3D LUT) is the color conversion 3D LUT 109 shown in FIG.

  Next, the process proceeds to step S202 (FIG. 4). In step S202, the computer 100 executes the work image generation process 116, and the work image data 117 is generated. The work image generation operation performed in step S202 will be described.

  FIG. 7 is a flowchart illustrating the flow of the work image generation process in the first embodiment.

  FIG. 8 is a diagram illustrating the work image generation process in the first embodiment.

  In step S401, brightness data is acquired from the work photographed image data 115 obtained by photographing with the flash ON. When acquiring the brightness data, the brightness signal Y is calculated by applying the following mathematical formula to the RGB values of the work photographed image data.

Y = 0.29891 × R + 0.58661 × G + 0.11448 × B
FIG. 8 shows a state in which the brightness data 202 is obtained by applying the brightness data acquisition process 201 shown in the previous equation to the photographed work image data 115.

  In step S402, chromaticity data is acquired from the photographed image data 113 obtained by photographing with the flash off.

  The chromaticity data is composed of two values (Cb, Cr) for one pixel, and the chromaticity data Cb, Cr can be calculated by applying the following formulas to the RGB values of the photographed image data of the work.

Cb = −0.16874 × R−0.33126 × G + 0.50000 × B
Cr = 0.50000 × R−0.41869 × G−0.0811 × B
FIG. 8 shows how chromaticity data 204 is obtained by applying the chromaticity data acquisition processing 203 shown in the previous equation to the photographed image data 113 of the work.

  In step S403, the brightness data 202 obtained in step S401 and the chromaticity data 204 obtained in step S402 are combined.

  The synthesizing process is executed by the following expression.

R = Y + 1.40200 x Cr
G = Y−0.34414 × Cb−0.71414 × Cr
B = Y + 1.777200 × Cb
FIG. 8 shows how the image data 117 is obtained by applying the lightness / chromaticity composition processing 205 to the lightness data 202 and the chromaticity data 204.

The generated work image data 117 has the following characteristics.
-Since the brightness signal is obtained from an image taken with the flash ON, there is little noise, there is little blurring, and there is little illumination unevenness.
The chromaticity signal is data reflecting the color of the observation light source because it is obtained from a photographed image with the flash off.

  Human visual characteristics are mainly sensitive to noise and blurring on the lightness component, and insensitive to noise and blurring on the chromaticity component. For this reason, even if some noise components are included in the chromaticity signal obtained under the flash OFF condition, a practical image quality can be obtained.

  The generation of the work image data 117 is completed by the above processing. In step S203, the color conversion process 118 is executed to generate duplicate image data 119. Here, the color conversion parameter 109 generated in step S201 is applied to the work image data 117 to obtain image data 119 obtained by color conversion. The color conversion parameter 109 is designed to perform color matching under the observation light source, and the work image data 117 is also generated from the chromaticity component obtained with the observation light source. The data has RGB values with matching colors.

  With the above processing, the duplicate image generation processing in step S002 is completed.

  In step S003, a duplicate image is output. This process is automatically executed by the computer 100 after the end of step S002. In step S <b> 003, duplicate image data 119 is sent from the computer 100 to the printer 103, and the printer 103 prints the duplicate image 120.

  When the work 111 and the duplicated image 120 are observed side by side under the observation light source 121, the color of the duplicated image is perceived as being close to the color of the work. Furthermore, it is perceived as a duplicate image with little noise, blur, and uneven illumination.

  As described above, according to the first embodiment, it is possible to generate a duplicate image having low noise, high definition, little brightness unevenness, and a color close to the original image.

  In other words, human visual characteristics have low sensitivity to the high frequency component of the chromaticity signal, and it is difficult to perceive even if there is some noise, so even if the chromaticity signal is generated from the information of the observation light source, Practical image quality can be obtained. Further, by applying noise reduction processing only to the chromaticity signal, it is possible to obtain a high quality photographed image with little noise without further visually reducing sharpness.

  In the above embodiment, since the color conversion table is generated from the chart image obtained with the observation light source, a duplicate image color-matched under the observation light source can be obtained.

  The second embodiment is an embodiment that further reduces brightness shift and noise. This is especially useful when using digital cameras with significantly different sensor sensitivity characteristics when comparing flash on and off shooting, or when the illuminance of the observation light source is low and noise with respect to chromaticity components cannot be ignored. This is an example.

  FIG. 9 is a diagram for explaining an entire duplicate image creation method according to the second embodiment.

  FIG. 10 is a flowchart illustrating the flow of the preliminary preparation process in the second embodiment.

  In the second embodiment, the same numbers are assigned to the components common to the first embodiment. The explanation of common elements is omitted. The overall processing operation is the same as the processing operation in the first embodiment shown in FIG. However, the processing contents in the preliminary preparation step by the operator in step S001 and the duplicate image generation step in step S002 are different.

  In the second embodiment, an operation performed by the operator in step S001 will be described.

  In step S501, the operator operates the computer 100, transmits the color chart image data 102 to the color printer 103, and outputs the color chart 104. Here, it is desirable that the paper on which the color chart is printed and various printing conditions are the same as the conditions for printing a duplicate of the work later.

  In step S502, the color chart 104 is photographed using the digital camera 105 in a state where the operator does not use the flash. Color chart photographed image data 107 obtained by photographing is temporarily stored in the digital camera 105. In this shooting, a color chart is placed under the observation light source of the work. Thus, the color chart photographed image data 104 has a color according to the spectrum of the observation light source.

  In step S503, the work 111 is photographed using the digital camera 105 in a state where the operator does not use the flash. The photographed image data 113 with the flash OFF obtained by photographing is temporarily stored in the digital camera 105. In this photo, the photo is taken under the observation light source. Thereby, the color of the photographed image data 113 becomes a color according to the spectrum of the observation light source. However, when the amount of light from the observation light source is insufficient, the exposure time may be long, resulting in a noisy image. In addition, there may be some blurring.

  In step S504, the brightness chart 110 is photographed using the digital camera 105 in a state where the operator does not use the flash.

  The brightness chart 110 is a gray scale chart having a plurality of levels of brightness from black solid to paper white. This chart may be printed in advance by the printer 103 or may be a chart prepared separately. Of course, the color chart 104 may include a gray patch having multi-level brightness, and may also be used for photographing the color chart 104.

  Brightness chart photographed image data 112 obtained by photographing with the flash OFF is temporarily stored in the digital camera 105. In this photo, the photo is taken under the observation light source.

  In step S504, the work 111 is photographed by using the digital camera 106 in a state where the operator uses the flash. The photographed image data 115 with flash ON obtained by photographing is temporarily stored in the digital camera 106. The color of the captured image data 115 is a color according to the flash light source. Since a flash can be used, a sufficient amount of light from the observation light source can be secured and an image with less noise can be obtained. Moreover, there is no blur.

  The digital cameras 105 and 106 shown in FIG. 9 differ only in turning on / off the flash, and it is desirable to use the same digital camera. In FIG. 9, the digital camera 106 is a digital camera equipped with a clip-on type flash. However, in practice, if multiple external flashes are used, uneven illumination is reduced and Shooting that is less susceptible to the negative effects of reflection is possible.

  In step S506, the brightness chart 110 is photographed by using the digital camera 105 in a state where the operator uses the flash. Lightness chart photographed image data 114 with flash ON obtained by photographing is temporarily stored in the digital camera 105.

  In step S507, the following shooting data is input to the computer 100.

Color chart shooting data 107
・ Shooting image data 113 with flash off
・ Brightness chart data 112 with flash off
・ Picture data 115 of the work taken with flash ON
Brightness chart photographed image data 114 when the flash is on.

  When the digital camera 105/106 is connected to the computer 100, it is desirable that the image data is automatically taken into the computer 100 immediately after shooting by connecting via a USB or wireless LAN. Of course, a small storage medium such as a compact flash (registered trademark) card may be replaced with the computer 100 from the digital camera 105, and image data may be copied to the computer 100.

  Thus, the advance preparation (step S001) by the operator is completed.

  In step S002, the computer 100 automatically executes a duplicate image generation process.

  Next, the operation of the duplicate image generation process will be described with reference to FIG. In step S201, the computer 100 generates color conversion parameters. Since the processing content of this step S201 is the same processing as Example 1, the description is omitted.

  Next, in step S202, the computer 100 executes a work image generation process 116a to generate work image data 117. An operation for generating a work image performed in step S202 in the second embodiment will be described.

  FIG. 11 is a flowchart illustrating an operation for generating a work image in the second embodiment.

  FIG. 12 is an explanatory diagram of an operation for generating a work image in the second embodiment.

  In step S601, the lightness value 215 in the flash-on state is acquired from the lightness chart photographed image data 114 photographed in the flash-on state.

  In the lightness chart photographed image 114, one patch is composed of a plurality of pixels. An RGB value of a plurality of pixels constituting a patch is averaged, and the RGB value is converted into a brightness value Yx by the following equation.

Yx = 0.29891 × R + 0.58661 × G + 0.11448 × B
The lightness value Yx is obtained for N patches included in the lightness chart, and the lightness values Yx (0) to Yx (N-1) are the lightness values when the flash is on. The brightness value Yx (n) is sorted in order of brightness.

  The patch of the brightness chart 110 may be included in the color chart 104, the color chart 104 may be photographed with the flash turned on, and a patch area corresponding to the brightness chart 110 may be extracted from the photographed image. Even in this case, the same processing as described above can be performed.

  FIG. 12 shows obtaining the lightness value Yx (n) 215 when the flash is on from the lightness chart photographed image data 114 photographed when the flash is on.

  In step S602, the brightness value 214 in the flash OFF state is acquired from the brightness chart captured image data 112 captured in the flash OFF state.

  In the lightness chart photographed image 112, one patch is composed of a plurality of pixels. The RGB values of a plurality of pixels constituting a patch are averaged, and the RGB values are converted into a brightness signal Y by the following equation.

Yy = 0.29891 × R + 0.58661 × G + 0.11448 × B
The lightness signal Y is obtained for N patches included in the lightness chart, and Yy (0) to Yy (N-1) values are lightness values when the flash is off. Sort Yy (n) in order of brightness.

  The patch of the lightness chart 110 is included in the color chart 104, the color chart 104 is photographed with the flash OFF, and the patch area corresponding to the lightness chart 110 is extracted from the photographed image, as described above. Can be processed.

  FIG. 12 shows obtaining the lightness value Yy (n) 214 when the flash is off from the lightness chart photographed image data 112 photographed when the flash is off.

  In step S603, the brightness correction parameter 208 is generated from the brightness value in the flash-on state and the brightness value in the flash-off state.

  The lightness correction parameter 208 plots the lightness value Yx (n) when the flash is on on the horizontal axis and the lightness value Yy (n) when the flash is off on the vertical axis, and this is plotted from n = 0 to n = N−1. It can be obtained by connecting the points plotted for multiple patches. The brightness correction parameter 208 is a one-dimensional lookup table (1D LUT).

  In the one-dimensional lookup table, the lightness value Yy (n) 214 when the image is taken with the flash OFF corresponding to the lightness value Yx (n) 215 obtained with the flash ON can be obtained.

  FIG. 12 shows that the brightness correction parameter generation processing 207 is executed based on the brightness value 214 when the flash is turned off and the brightness value 215 when the flash is turned on to obtain the brightness correction parameter 208.

  In step S604, a lightness signal Y is acquired from the photographed image data 115 with the flash ON.

  The lightness signal Y can be calculated by applying the following formula to the RGB values of the photographed image data.

Y = 0.29891 × R + 0.58661 × G + 0.11448 × B
FIG. 12 shows how the brightness data 202 is obtained by applying the brightness data acquisition process 201 according to the previous equation to the photographed image data 115 of the work.

  In step S605, the lightness correction parameter 208 obtained in step S603 is applied to the lightness data 202 to obtain lightness data 210 after correction. The corrected brightness data 210 is brightness characteristics obtained under flash OFF, that is, under an observation light source.

  In step S606, chromaticity data is acquired from the photographed image data 113 with the flash off.

  The chromaticity data is composed of two values (Cb, Cr) for one pixel, and the chromaticity data Cb, Cr can be calculated by applying the following formula to the RGB values of the photographed image data.

Cb = −0.16874 × R−0.33126 × G + 0.50000 × B
Cr = 0.50000 × R−0.41869 × G−0.0811 × B
FIG. 12 shows how chromaticity data 204 is obtained by applying the chromaticity data acquisition process 203 according to the previous formula to the photographed image data 113 of the work.

  In step S607, the smoothing filter 211 is applied to the chromaticity data Cb and Cr in order to reduce noise generated by photographing with the observation light source.

  Human visual characteristics are not sensitive to high frequency components for chromaticity components. Therefore, even if the light smoothing filter 211 is applied to the chromaticity component, the sharpness of the image is not greatly lowered, and a practical image quality can be obtained. When the noise of the original image is large, applying the smoothing filter 211 to the chromaticity component is effective for reducing the tint noise.

  Even if the smoothing filter 211 is a light smoothing filter composed of a 3 × 3 or 5 × 5 matrix, a great effect can be obtained.

  FIG. 12 shows how the smoothing filter 211 is applied to the chromaticity data 204 to obtain chromaticity data 212 with reduced noise. That is, the smoothing filter 211 performs noise reduction processing.

  In step S608, the corrected brightness data 210 obtained in step S605 and the noise-reduced chromaticity data 212 obtained in step S607 are synthesized by the brightness / chromaticity synthesis process 205.

  The synthesizing process is performed by the following formula.

R = Y + 1.40200 x Cr
G = Y−0.34414 × Cb−0.71414 × Cr
B = Y + 1.777200 × Cb
FIG. 12 shows how the lightness / chromaticity synthesis process 205 is applied to the lightness data 210 and the chromaticity data 212 to obtain the work image data 117.

The generated work image data 117 has the following characteristics.
-Since the brightness signal is obtained from an image taken with the flash ON, there is little noise, there is little blurring, and there is little illumination unevenness.
-Since the brightness signal is further converted from the flash-on characteristic to the off-characteristic, even if there is a difference in the light-receiving sensitivity characteristic of the camera in each flash on / off, the difference in the light-receiving sensitivity characteristic is effective Can be absorbed.
The chromaticity signal is data reflecting the color of the observation light source because it is obtained from a photographed image with the flash off.
Since the chromaticity signal is further subjected to noise reduction processing, an image with little tint noise can be obtained even when shooting at low illuminance.

  The generation of the work image data 117 is completed by the above processing.

  Next, in step S203, the color conversion process 118 shown in FIG. 9 is performed, and duplicate image data 119 is generated.

  Here, the color conversion parameter 109 generated in step S201 is applied to the work image data 117 to obtain image data 119 obtained by color conversion. The color conversion parameter 109 is designed to perform color matching under an observation light source. In addition, the brightness component of the photographed image data is converted into the characteristics of the observation light source, and the chromaticity component is also generated from the data obtained with the observation light source. This is data having RGB values that match the color of the duplicate image 120.

  With the above processing, the duplicate image generation processing in step S002 is completed.

  Next, in step S003, a duplicate image is output. This process is automatically executed by the computer 100 after the end of step S002.

  In step S003, the duplicate image data 119 is sent from the computer 100 to the printer 103, and the duplicate image 120 is printed.

  When the work 111 and the duplicated image 120 are observed side by side under the observation light source 121, it is perceived that the color of the duplicated image 120 approximates the color of the work. In addition, the reproduced image has little noise, blur, and uneven illumination.

  As described above, according to the second embodiment, it is possible to generate a duplicate image 120 with low noise, high definition, little brightness unevenness, and a color close to the original image.

  In particular, even in a digital camera in which the light receiving sensitivity characteristic of the sensor is greatly different between flash ON and OFF, it has a mechanism that absorbs the characteristic difference between the flash OFF and ON. As a result, the lightness characteristic after combining the lightness / chromaticity signals approaches the characteristic originally obtained under the observation light source, and the color matching performance in the color conversion process can be enhanced.

  Further, even when the illumination intensity of the observation light source is low and noise with respect to the chromaticity component cannot be ignored, noise reduction processing is performed only on the chromaticity component, so that a reduction in sharpness in the duplicate image 120 is minimized. The color noise can be reduced while suppressing the noise.

  Furthermore, since the brightness signal can be corrected between the flash ON / OFF states, the difference in brightness characteristics between the observation light source and the flash light source can be absorbed.

  In the first and second embodiments, even a storage medium in which the processing content executed by the computer 100 is stored as a computer program on a CD-ROM or a floppy (registered trademark) disk is a subject of the present invention.

  In addition, the function of the said Example is realizable also with the following structures. That is, it is also achieved by supplying a program code for performing each process in the above-described embodiment to a system or apparatus, and a computer (or CPU or MPU) of the system or apparatus executing the program code. In this case, the program code itself read from the storage medium realizes the function of the above embodiment, and the storage medium storing the program code also realizes the function of the above embodiment.

  Further, the program code for realizing the functions of the above embodiments may be executed by one computer (CPU, MPU), or may be executed by a plurality of computers cooperating. Also good. Further, the program code may be executed by a computer, or hardware such as a circuit for realizing the function of the program code may be provided. Alternatively, a part of the program code may be realized by hardware and the remaining part may be executed by a computer.

201 ... lightness data acquisition,
203 ... chromaticity data acquisition,
205 ... Lightness / chromaticity synthesis,
207 ... Lightness correction parameter generation,
209: Lightness correction.

Claims (8)

First photographed image data acquisition means for acquiring first photographed image data by photographing a photograph using a flash;
Second work photographed image data acquisition means for acquiring second work photographed image data by photographing the above work without using a flash;
Brightness data acquisition means for acquiring brightness data from the first photographed image data;
Chromaticity data acquisition means for acquiring chromaticity data from the second work photographed image data;
Combining means for combining the lightness data acquired by the lightness data acquisition means and the chromaticity data acquired by the chromaticity data acquisition means;
An image processing apparatus comprising: Claim 1 wherein
First brightness value acquisition means for acquiring a first brightness value from a brightness chart by photographing the work using a flash;
Second brightness value acquisition means for acquiring a second brightness value from the brightness chart by shooting without using a flash;
Brightness correction parameter generation means for generating a brightness correction parameter from the first brightness value and the second brightness value;
Brightness correction means for correcting the brightness data extracted from the first photographed image data by applying the brightness correction parameter;
Have
The image processing apparatus according to claim 1, wherein the synthesizing unit is a unit that synthesizes the lightness data corrected by the lightness correction unit and the chromaticity data acquired by the chromaticity data acquisition unit. Claim 1 or claim 2, wherein
The image processing apparatus, wherein the chromaticity data acquisition means includes noise reduction means for performing noise reduction processing on chromaticity data acquired from the second work photographed image data. Claim 2 or claim 3, wherein
The color chart and the brightness chart are composed of the same color chart,
The first lightness value acquisition means acquires the first lightness value from at least a part of an image of a color chart photographed using a flash,
The second lightness value acquisition means acquires the second lightness value from at least a part of an image of a color chart taken without using a flash. It is one of Claims 1-4, Comprising:
An image processing apparatus comprising output means for outputting work image data color-converted by the color conversion means with a printer. A first work photographed image data acquisition step of acquiring first work photographed image data by photographing the work using a flash;
A second photographed image data acquisition step of acquiring second photographed image data by photographing the above work without using a flash;
A brightness data acquisition step of acquiring brightness data from the first photographed image data;
A chromaticity data acquisition step of acquiring chromaticity data from the second work photographed image data;
A synthesis step of combining the brightness data acquired in the brightness data acquisition step and the chromaticity data acquired in the chromaticity data acquisition step;
An image processing method comprising: Claim 6.
A first brightness value acquisition step of acquiring a first brightness value from a brightness chart by photographing the work using a flash;
A second brightness value acquisition step of acquiring a second brightness value from the brightness chart by photographing without using a flash;
A brightness correction parameter generating step of generating a brightness correction parameter from the first brightness value and the second brightness value;
A brightness correction step of correcting the brightness data extracted from the first photographed image data by applying the brightness correction parameter;
Have
The image processing apparatus according to claim 1, wherein the combining step is a step of combining the lightness data corrected in the lightness correction step and the chromaticity data acquired in the chromaticity data acquisition step.   A program for causing a computer to execute the image processing method according to claim 6.

Images