JP2006050390A - Image processing device, imaging device, recording medium and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing device with structure for calculating an optimum image processing parameter, an imaging device, a recording medium and an image processing program. <P>SOLUTION: A proofreading image is stored in a memory unit 100, and a feature amount calculating unit 102 calculates the feature amount from the proofreading image transmitted from the memory unit 100 by the control of a control unit 105. In a target value setting unit 101, a target value is set through an exterior I/F 106 by the control of the control unit 105. In a parameter calculating unit 103, a parameter for image processing is calculated by the control of the control unit 105 by using the feature amount calculated by the feature amount calculating unit 102 and the target value set by the target value setting unit 101, and the calculated parameter is transferred to an output unit 104. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus that determines an image processing parameter using a calibration image, and more specifically, calculates a noise amount, frequency characteristics, color information, and a color difference using a calibration image, and the image processing parameter The present invention relates to an image processing apparatus, an imaging apparatus, a recording medium, and an image processing program.

  Conventionally, a method of performing correction using a calibration image has been proposed in order to remove individual differences in an image processing apparatus and an imaging apparatus. Patent Document 1 discloses a method for improving various aberrations, chromatic aberration, image sensor color reproducibility, pixel defects and the like of an imaging optical system in a digital image input device such as a digital camera or a video camera. Japanese Patent Application Laid-Open No. H10-228688 discloses a method of calculating calibration function information using a signal obtained by capturing a calibration image including three or more types of gray charts having different reflectivities from each other using an image sensor. Has been.

Japanese Patent Laid-Open No. 11-205652 JP 2004-128985 A

  In Patent Document 1, it is necessary to store a desired image in advance in addition to the calibration image, and there is a problem that the amount of memory becomes enormous. Further, Patent Document 2 has a problem that information other than noise cannot be dealt with.

  In view of the above problems, the present invention sets an image quality target value in advance, and uses the target value and a calibration image to calculate an optimal image processing parameter, an image processing apparatus and an imaging apparatus, It is an object to provide a recording medium and an image processing program.

(1) In order to achieve the above object, an image processing apparatus according to the first embodiment of the present invention includes:
In an image processing apparatus for processing a digitized image, a storage means for storing a calibration image in the image processing apparatus, a target value setting means for setting a target value, and a feature quantity for obtaining a feature quantity from the calibration image The image processing apparatus includes: a calculation unit; a parameter calculation unit that calculates an image processing parameter using the target value and the feature amount information; and an output unit that outputs the parameter.

  The embodiment relating to the present invention corresponds to the first embodiment shown in FIGS. 1 and FIG. 4 is the storage means configured as described above, the target value setting means is the target value setting section 101 shown in FIG. 1 and FIG. 4, and the feature amount calculation means is the FIG. 2 and 4, the parameter calculation means is the parameter calculation section 103 shown in FIGS. 1, 3 and 4, and the output means is the output section 104 shown in FIGS. 1 and 4. Each is applicable.

  In a preferred application example of the present invention, the calibration image is stored in the storage unit 100 shown in FIG. 1, the target value is set in the target value setting unit 101, and the calibration image is stored in the feature amount calculation unit 102. This is an image processing apparatus that calculates feature amounts, calculates parameters using information of the target value setting unit 101 and the feature amount calculation unit 102 in the parameter calculation unit 103, and outputs the parameters in the output unit 104.

  The image processing apparatus according to the first embodiment of the present invention outputs image processing parameters using a calibration image and a target value. For this reason, an image processing parameter that satisfies the target value can be obtained.

  (2) In the image processing apparatus according to the first embodiment of the present invention, the feature amount calculating unit includes a luminance value calculating unit that calculates a luminance value of the calibration image, and a frequency with respect to the luminance value. Frequency calculating means for converting and calculating a frequency component. The embodiment relating to the present invention corresponds to the first embodiment shown in FIGS. The luminance value calculating means having the above configuration corresponds to the luminance signal calculating section 200 shown in FIG. 2, and the frequency calculating means corresponds to the DCT section 201 shown in FIG. A preferred application example of the present invention is an image processing apparatus that calculates a luminance value by a luminance signal calculation unit 200 shown in FIG. 2 and performs frequency conversion on the luminance value by a DCT unit 201. The present invention calculates the frequency characteristics of a calibration image. For this reason, a frequency characteristic can be used as the feature amount of the calibration image.

  (3) Further, in the image processing apparatus according to the first embodiment of the present invention, the feature amount calculating means includes a chart extracting means for extracting a plurality of chart portions included in the calibration image, and each chart. And color information calculation means for calculating color information. The embodiment relating to the present invention corresponds to the first embodiment shown in FIGS. The chart extracting unit having the above-described configuration corresponds to the chart extracting unit 300 shown in FIG. 2, and the color information calculating unit corresponds to the color information calculating unit 302 shown in FIG. A preferred application example of the present invention is an image processing apparatus in which a chart extraction unit 300 shown in FIG. 2 extracts a calibration image chart and a color information calculation unit 302 calculates color information of the chart. The present invention calculates color information of the chart portion of the calibration image. For this reason, color information can be used as the feature amount of the calibration image.

  (4) Further, in the image processing apparatus according to the first embodiment of the present invention, the parameter calculation unit includes a calculation unit that performs calculation using information of the target value setting unit and the feature amount calculation unit, And filter calculation means for calculating a filter used for edge enhancement processing using information obtained from the calculation means. The embodiment relating to the present invention corresponds to the first embodiment shown in FIGS. The calculation means having the above configuration corresponds to the calculation section 400 shown in FIG. 3, and the filter calculation means corresponds to the filter coefficient calculation section 401 shown in FIG. In a preferred application example of the present invention, the calculation unit 400 shown in FIG. 4 calculates information used for filter calculation using the information of the target value setting unit 101 and the feature amount calculation unit 102, and the filter coefficient calculation unit 401 An image processing apparatus that calculates filter coefficients used in edge enhancement processing. The present invention calculates a filter used for edge enhancement processing. Therefore, the edge enhancement filter can be calculated from the feature amount and the target value, and a high-quality edge enhancement filter can be calculated.

  (5) Further, the image processing apparatus according to the first embodiment of the present invention is characterized in that the parameter calculation means includes conversion matrix calculation means for calculating a conversion matrix used for color conversion. The embodiment relating to the present invention corresponds to the first embodiment shown in FIGS. The conversion matrix calculation means having the above configuration corresponds to the matrix coefficient calculation unit 501 shown in FIG. A preferred application example of the present invention is to calculate the XYZ chromaticity value using the information of the target value setting unit 101 in the XYZ calculation unit 500 shown in FIG. This is an image processing apparatus that calculates matrix coefficients using information of the quantity calculation unit 102. The present invention calculates a matrix used for color conversion. Therefore, a highly accurate color conversion matrix can be calculated using the color information of the calibration image.

(6) An image forming apparatus according to the second embodiment of the present invention includes:
In an image processing apparatus for processing a digitized image, a storage means for storing a calibration image in the image processing apparatus, a target value setting means for setting a target value, and a feature quantity for obtaining a feature quantity from the calibration image A calculation unit; a comparison unit that compares the target value with the feature quantity; a parameter calculation unit that calculates an image processing parameter using information of the comparison unit; an output unit that outputs the parameter; and the parameter And determining means for determining whether to transfer to the feature quantity calculating means or to the output means.

  The embodiment relating to the present invention corresponds to the second embodiment shown in FIGS. The storage means configured as described above is the storage section 100 shown in FIG. 6, the target value setting means is the target value setting section 101 shown in FIG. 6, and the feature quantity calculation means is the feature quantity calculation shown in FIGS. 6 is a comparison unit 150 shown in FIG. 6, a parameter calculation unit is a parameter calculation unit 1003 shown in FIGS. 6 and 8, and an output unit is an output unit 104 shown in FIG. Corresponds to the control unit 105 shown in FIG. In a preferred application example of the present invention, the calibration image is stored in the storage unit 100 shown in FIG. 6, the target value is set in the target value setting unit 101, and the calibration image is stored in the feature amount calculation unit 1002. The feature value is calculated, the comparison unit 150 compares the target value with the feature value, the parameter calculation unit 1003 calculates the parameter using the information of the comparison unit 150, and the control unit 105 calculates the parameter. The image processing apparatus determines whether to transfer the parameter calculated by the unit 1003 to the feature amount calculation unit 1002 and outputs the parameter by the output unit 104.

  The image forming apparatus according to the second embodiment of the present invention outputs image processing parameters by comparing feature amounts with target values. In this way, since the feature amount and the target value are compared and the image processing parameter is calculated until the target value is satisfied, an image processing parameter that satisfies the target value can be obtained.

  (7) Further, in the image forming apparatus according to the second embodiment of the present invention, the feature amount calculating means includes a chart extracting means for extracting a plurality of chart portions included in the calibration image, and each chart Color information extracting means for extracting color information, color information true value setting means for setting the true value of color information of each chart, and color difference calculating means for calculating a difference between the color information and the true value It is characterized by that. The embodiment relating to the present invention corresponds to the second embodiment shown in FIGS. The chart extracting unit configured as described above is the chart extracting unit 600 shown in FIG. 7, the color information extracting unit is the chart Lab calculating unit 602 shown in FIG. 7, and the color information true value setting unit is the Lab setting unit shown in FIG. Reference numeral 604 denotes the color difference calculation means, which corresponds to the color difference calculation unit 603 shown in FIG. A preferred application example of the present invention is that the chart extraction unit 600 shown in FIG. 7 extracts the chart portion of the calibration image, the chart Lab calculation unit 602 calculates the color information of the chart unit, and the Lab setting unit 604 In this image processing apparatus, the true value of the color information in the chart part is set, and the difference between the color information calculated in the chart part and the true value is obtained by the color difference calculating part 603. The present invention calculates the color difference from the difference between the color information and the true value. For this reason, the color difference can be used as the feature amount of the calibration image.

  (8) In the image forming apparatus according to the second embodiment of the present invention, the parameter calculation unit includes a conversion matrix calculation unit that calculates a conversion matrix used for color conversion, and a matrix order that sets the order of the conversion matrix. And setting means. The embodiment relating to the present invention corresponds to the second embodiment shown in FIGS. The matrix order setting means having the above configuration corresponds to the matrix order setting unit 700 shown in FIG. A preferred application example of the present invention is an image processing apparatus in which the order of the color conversion matrix is set using the information of the comparison unit 150 in the matrix order setting unit 700 shown in FIG. In the present invention, the order of a matrix used for color conversion is set. For this reason, the conversion accuracy is increased by increasing the order of the color conversion matrix.

(9) An image forming apparatus according to the third embodiment of the present invention includes:
In an image processing apparatus for processing a digitized image, a storage means for storing a calibration image in the image processing apparatus, a target value setting means for setting a target value, and a feature quantity for obtaining a feature quantity from the calibration image A calculation unit; a comparison unit that compares the target value with the feature quantity; a parameter calculation unit that calculates an image processing parameter using information of the comparison unit; and an image processing unit that performs image processing using the parameter And an output means for outputting the parameter, and whether to transfer the image processed by the image processing means to the feature quantity calculating means or to transfer the parameter calculated by the parameter calculating means to the output means And a judging means.

  The embodiment relating to the present invention corresponds to the third embodiment shown in FIGS. The storage means configured as described above is the storage section 100 shown in FIG. 10, the target value setting means is the target value setting section 101 shown in FIG. 10, and the feature quantity calculation means is the feature quantity calculation section shown in FIGS. 1102 is the comparison means 150 shown in FIG. 10, the parameter calculation means is the image processing section 151 shown in FIGS. 10 and 12, and the image processing means is the image processing section 151 shown in FIGS. However, the output means corresponds to the output section 104 shown in FIG. 10, and the judgment means corresponds to the control section 105 shown in FIG. In a preferred application example of the present invention, the calibration image is stored in the storage unit 100 shown in FIG. 10, the target value is set in the target value setting unit 101, and the calibration image is stored in the feature amount calculation unit 1102. The feature value is calculated, the comparison unit 150 compares the target value and the feature value, the image processing unit 151 calculates the parameter using the information of the comparison unit 150, and the image processing unit 151 calibrates the feature value. An image for performing image processing on the image for use, determining whether to transfer the image processed by the image processing unit 151 to the feature amount calculation unit 1102 by the control unit 105, and outputting the above parameters by the output unit 104 It is a processing device.

  The image forming apparatus according to the third embodiment of the present invention outputs image processing parameters by comparing feature amounts with target values. As described above, the feature amount is compared with the target value, and the image processing parameter is calculated until the target value is satisfied. Therefore, an image processing unit parameter satisfying the target value can be obtained.

  (10) In the image forming apparatus according to the third embodiment of the present invention, the feature amount calculation unit includes a chart extraction unit that extracts a plurality of chart portions included in the calibration image, and a chart extraction unit. Noise amount calculating means for calculating the noise amount. The embodiment relating to the present invention corresponds to the third embodiment shown in FIGS. The chart extracting unit having the above configuration corresponds to the chart extracting unit 800 shown in FIG. 11, and the noise amount calculating unit corresponds to the noise calculating unit 802 shown in FIG. A preferred application example of the present invention is an image processing apparatus in which a chart extraction unit 800 shown in FIG. 11 extracts a chart portion of a calibration image, and a noise calculation unit 802 calculates a noise amount of the chart unit. The present invention calculates the noise amount of the calibration image. For this reason, the noise amount can be used as the feature amount of the calibration image.

  (11) In the image forming apparatus according to the third embodiment of the present invention, the image processing unit includes a local region extracting unit that extracts a local region centered on the target pixel of the image, and a noise amount with respect to the target pixel. A noise amount calculating means for calculating the noise amount, a gain setting means for setting a gain for the noise amount, and a noise reducing means for performing noise reduction processing on the local region using the information on the noise amount and the gain. It is characterized by having. The embodiment relating to the present invention corresponds to the third embodiment shown in FIGS. The local region extracting unit configured as described above is a local region extracting unit 900 shown in FIG. 12, the noise amount calculating unit is a noise amount calculating unit 904 shown in FIG. 12, and the gain setting unit is a gain setting unit 906 shown in FIG. However, the noise reduction means corresponds to the noise reduction unit 903 shown in FIG. In a preferred application example of the present invention, a local region of an image is extracted by a local region extraction unit 900 shown in FIG. 12, a noise amount of a target pixel is calculated by a noise amount calculation unit 904, and a gain setting unit 906 The image processing apparatus sets a gain used for noise reduction and performs noise reduction processing in a noise reduction unit 903. The present invention calculates the noise amount of the target pixel and performs noise reduction using the noise amount and a set gain. For this reason, since noise reduction is performed based on the calculated amount of noise, a high-quality image can be obtained. Further, since the noise reduction gain can be set, the strength of noise reduction can be changed.

  (12) In the image forming apparatuses according to the first to third embodiments of the present invention, the storage unit includes an external storage unit that can be attached to and detached from the image processing apparatus. The first to third embodiments shown in FIGS. 1, 6, and 10 correspond to the embodiments relating to the present invention. The external storage means configured as described above corresponds to the storage unit 100 shown in FIGS. A preferred application example of the present invention is an image processing apparatus that stores a calibration image in the storage unit 100 shown in FIGS. In the present invention, the storage unit can be removed. For this reason, the same calibration image can be used in other image processing apparatuses by removing the storage unit in which the calibration image is stored.

(13) The imaging apparatus according to the first embodiment of the present invention includes:
An imaging means for taking an image, a preprocessing means for converting the taken image signal into a digital signal, a signal processing means for the digital signal, and an output means for transferring an output signal of the signal processing means Shooting mode setting section, and
Storage means for storing a calibration image photographed by the photographing means, target value setting means for setting a target value, feature quantity calculation means for obtaining a feature quantity from the calibration image, the target value and the feature quantity An image processing parameter mode setting unit including a parameter calculation unit that calculates image processing parameters using the information and an output unit that outputs the parameters, and the parameter calculation unit converts the parameters for shooting Image processing parameters are transferred to the signal processing means.

  The embodiment relating to the present invention corresponds to the first embodiment shown in FIG. The photographing means for photographing the image having the above-described configuration corresponds to the lens system 111 and the CCD 112, the preprocessing means corresponds to the preprocessing section 113, the signal processing means corresponds to the signal processing section 114, and the output means corresponds to the output section 104. A preferred application example of the present invention is an image pickup apparatus having a shooting mode setting unit and an image processing parameter mode setting unit shown in FIG. According to the present invention, means for outputting image processing parameters using a calibration image and a target value is provided in the imaging apparatus. For this reason, it is possible to obtain an image processing parameter that satisfies the target value in the imaging apparatus. In addition, since the calibration image is taken, it is possible to cope with a change with time of the imaging apparatus and to calculate the image processing parameter with high accuracy.

(14) An imaging apparatus according to a second embodiment of the present invention includes an imaging unit that captures an image, a preprocessing unit that converts the captured image signal into a digital signal, and a signal processing unit that processes the digital signal. An imaging mode setting unit comprising output means to which an output signal of the signal processing means is transferred, and
Storage means for storing a calibration image photographed by the photographing means; target value setting means for setting a target value; feature quantity calculating means for obtaining a feature quantity from the calibration image; the target value and the feature quantity Comparing means, parameter calculating means for calculating parameters of image processing using information of the comparing means, output means for outputting the parameters, and transferring the parameters to the feature amount calculating means or the output And an image processing parameter mode setting unit configured to determine whether to transfer to the means, wherein the parameter calculation means transfers the image processing parameters converted for photographing to the signal processing means. And

  The embodiment of the present invention corresponds to the configuration of the shooting mode setting unit of the first embodiment shown in FIG. 4 and the image processing parameter mode setting unit shown in FIG. According to the present invention, since the image processing parameter is calculated until the feature value is compared with the target value and the target value is satisfied, the image processing parameter satisfying the target value can be obtained in the imaging apparatus. In addition, since the calibration image is taken, it is possible to cope with a change with time of the imaging apparatus and to calculate the image processing parameter with high accuracy.

(15) An imaging apparatus according to a third embodiment of the present invention includes an imaging unit that captures an image, a preprocessing unit that converts the captured image signal into a digital signal, and a signal processing unit that processes the digital signal. An imaging mode setting unit comprising output means to which an output signal of the signal processing means is transferred, and
Storage means for storing a calibration image photographed by the photographing means; target value setting means for setting a target value; feature quantity calculating means for obtaining a feature quantity from the calibration image; the target value and the feature quantity Comparing means, parameter calculating means for calculating image processing parameters using information of the comparing means, image processing means for performing image processing using the parameters, output means for outputting the parameters, Image processing parameter mode setting comprising: determination means for determining whether to transfer the image processed by the image processing means to the feature amount calculation means or to transfer the parameter calculated by the parameter calculation means to the output means The parameter calculation means transfers the image processing parameters converted for photographing to the signal processing means.

  The embodiment of the present invention corresponds to the configuration of the shooting mode setting unit of the first embodiment shown in FIG. 4 and the image processing parameter mode setting unit of the third embodiment shown in FIG. . According to the present invention, the feature amount is compared with the target value, and the image processing parameter is calculated until the target value is satisfied. Therefore, the image processing parameter that satisfies the target value can be obtained in the imaging apparatus. In addition, since the calibration image is taken, it is possible to cope with a change with time of the imaging apparatus and to calculate the image processing parameter with high accuracy.

  (16) A recording medium according to the first embodiment of the present invention is a recording medium in which image signal data of an optimal image processing parameter is recorded by a preset target value of image quality and a calibration image. Storage processing, target value setting processing for setting a target value, feature amount calculation processing for obtaining a feature amount from the calibration image, and image processing parameters using information on the target value and the feature amount. Means is provided for holding image signal data that has been subjected to parameter calculation processing to be calculated.

  The embodiment relating to the present invention corresponds to the first embodiment shown in FIG. The storage processing for storing the calibration image having the above-described configuration is the processing of the storage unit 100, the target value setting processing for setting the target value is the processing of the target value setting unit 101, and the feature amount is calculated from the calibration image. The processing is performed by the feature amount calculation unit 102, the parameter calculation processing for calculating the image processing parameters using the target value and the feature amount information is performed by the parameter calculation unit 103, and the recording medium is performed by the output unit 104. Correspond. According to the present invention, image signal data subjected to image processing parameter calculation processing using a calibration image and a target value is held in a recording medium. Therefore, image processing by an information processing device such as a PC (Personal Computer) can be performed using the recording medium.

  (17) A recording medium according to the second embodiment of the present invention is a recording medium in which image signal data of an optimal image processing parameter is recorded based on a preset image quality target value and a calibration image. Are stored in the image processing apparatus, target value setting processing for setting a target value, feature amount calculation processing for obtaining a feature amount from the calibration image, and comparison processing for comparing the target value with the feature amount A parameter calculation process that calculates image processing parameters using the information of the comparison process, an output process that outputs the parameters, and whether the parameters are transferred to the feature quantity calculation process or the output process Means is provided for holding image signal data that has been subjected to determination processing.

  The embodiment relating to the present invention corresponds to the second embodiment shown in FIG. The storage processing for storing the calibration image having the above-described configuration is the processing of the storage unit 100, the target value setting processing for setting the target value is the processing of the target value setting unit 101, and the feature amount is calculated from the calibration image. The processing is performed by the feature amount calculation unit 1002, and the parameter calculation processing for calculating the image processing parameters using the target value and the feature amount information is performed by the parameter calculation unit 1003, and the parameters are converted into the feature amount calculation processing. The determination processing for determining whether to transfer or to output processing corresponds to the processing of the control unit 105. According to the present invention, the feature amount is compared with the target value, the image processing parameter is calculated until the target value is satisfied, and the image signal data subjected to the calculation process is held in the recording medium. Therefore, it is possible to obtain a recording medium that holds image processing parameters that satisfy the target value. Therefore, image processing by an information processing device such as a PC can be performed using the recording medium.

  (18) A recording medium according to the third embodiment of the present invention is a recording medium in which image signal data of an optimal image processing parameter is recorded by a preset target value of image quality and a calibration image. Are stored in the image processing apparatus, target value setting processing for setting a target value, feature amount calculation processing for obtaining a feature amount from the calibration image, and comparison processing for comparing the target value with the feature amount A parameter calculation process for calculating image processing parameters using the information of the comparison process, an image process using the parameters, an output process for outputting the parameters, and an image processed by the image process. The image signal data that has been subjected to determination processing for determining whether to transfer to the feature amount calculation processing or to transfer the parameter calculated in the parameter calculation processing to the output processing. Wherein the means for lifting are provided.

  The embodiment relating to the present invention corresponds to the third embodiment shown in FIG. The storage processing for storing the calibration image having the above-described configuration is the processing of the storage unit 100, the target value setting processing for setting the target value is the processing of the target value setting unit 101, and the feature amount is calculated from the calibration image. The processing is a feature amount calculation unit 1102, the comparison process for comparing the target value and the feature amount is a comparison unit 150 processing, and the parameter calculation processing for calculating image processing parameters using the information of the comparison processing is The process of the image processing unit 151 is an image process using the parameter, and the process of the image processing unit 151 determines whether the process transfers the parameter to the feature amount calculation process or the output process. These processes correspond to each. According to the present invention, the feature amount is compared with the target value, and the image processing parameter is calculated until the target value is satisfied. Therefore, it is possible to obtain a recording medium that holds the image processing parameter that satisfies the target value. Therefore, image processing by an information processing device such as a PC can be performed using the recording medium.

  (19) An image processing program according to the first embodiment of the present invention includes a procedure for reading a calibration image into a computer, a procedure for calculating a feature amount of the calibration image, a set target value, and the calculation The method is characterized in that a procedure for calculating a parameter for image processing using the feature amount thus obtained and a procedure for outputting a calculation result of the parameter are executed.

The embodiment relating to the present invention corresponds to the first embodiment shown in FIG.
The procedure for reading the calibration image is S1 in FIG. 5, the procedure for calculating the feature value of the calibration image is S2, the target value setting is S4, and the target value and the calculated feature value are used for image processing. The procedure for calculating the parameters corresponds to S3, and the procedure for outputting the parameter calculation results corresponds to S5.

  (20) An image processing program according to the second embodiment of the present invention includes a procedure for reading a calibration image into a computer, a procedure for calculating a feature amount of the calibration image, a set target value, and the calculation A procedure for comparing the calculated feature values, a procedure for calculating an image processing parameter if the calculated feature value does not satisfy a target value, a procedure for feeding back the parameter to the feature value calculation process, And a step of outputting the parameter when the calculated feature amount satisfies a target value.

  The embodiment relating to the present invention corresponds to the second embodiment shown in FIG. The procedure for reading the calibration image is S1, the procedure for calculating the feature value of the calibration image is S10, the target value setting is S4, and the set target value is compared with the calculated feature value Is S12, S11 is the procedure for calculating the image processing parameters if the calculated feature quantity does not satisfy the target value, and S11 and S10 are the procedures for returning the parameters to the feature quantity calculation process. S5 corresponds to the procedure for outputting the parameter if the calculated feature value satisfies the target value.

  (21) An image processing program according to the third embodiment of the present invention includes: a procedure for reading a calibration image into a computer; a procedure for performing predetermined image processing on the calibration image; A procedure for calculating the feature value of the calibration image, a procedure for comparing the set target value with the calculated feature value of the calibration image, and if the calculated feature value does not satisfy the target value, the image A procedure for returning to the processing and a procedure for outputting the parameter if the calculated feature value satisfies a target value are executed.

  The embodiment relating to the present invention corresponds to the third embodiment shown in FIG. The procedure for reading the calibration image is S1, the procedure for performing predetermined image processing on the calibration image is S20, the procedure for calculating the feature value of the calibration image after the image processing is S21, The setting is S4, the procedure for comparing the target value with the calculated feature value of the calibration image is S12, and the procedure for returning to the image processing if the calculated feature value does not satisfy the target value is S12, The procedure for outputting the parameter if S20 satisfies the calculated feature value satisfies the target value corresponds to S5.

  In the inventions according to the embodiments (19) to (21), image signal processing is realized by a computer using an image processing program. For this reason, it is possible to realize image signal processing that is quick and accurate and highly accurate.

  In the present invention, an image processing apparatus, an imaging apparatus, and a recording medium having a configuration capable of calculating an optimum image processing parameter by setting a target value of image quality in advance and using the target value and a calibration image. And an image processing program is obtained.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. 1 to 5 show a first embodiment of the present invention. FIG. 1 is a configuration diagram of the first embodiment, FIG. 2 is a configuration diagram of the feature quantity calculation unit 102, FIG. 3 is a configuration diagram of the parameter calculation unit 103, FIG. 4 is another configuration diagram of the first embodiment, FIG. 5 is a flowchart relating to the software of the first embodiment.

  The configuration of the first embodiment will be described with reference to FIG. The storage unit 100 stores a calibration image in advance. The storage unit 100 is connected to the feature amount calculation unit 102, and the feature amount calculation unit 102 is connected to the parameter calculation unit 103. The target value setting unit 101 is connected to the parameter calculation unit 103, and the parameter calculation unit 103 is connected to the output unit 104. A control unit 105 such as a microcomputer is bidirectionally connected to the storage unit 100, the target value setting unit 101, the feature amount calculation unit 102, the parameter calculation unit 103, and the output unit 104. Further, an external I / F unit 106 having an interface for setting a target value in the target value setting unit 101 is also connected to the control unit 105 in both directions.

  Next, the signal flow in FIG. 1 will be described. The storage unit 100 stores a calibration image such as a Macbeth chart. This calibration image is stored in a ROM existing in the image processing apparatus or stored in a removable external memory card. The feature amount calculation unit 102 calculates a feature amount from the calibration image transferred from the storage unit 100 based on the control of the control unit 105. A method for calculating these feature values will be described later.

  The target value setting unit 101 sets a target value through the external I / F 106 based on the control of the control unit 105. Based on the control of the control unit 105, the parameter calculation unit 103 calculates image processing parameters using the feature amount calculated from the feature amount calculation unit 102 and the target value set by the target value setting unit 101. The parameter is transferred to the output unit 104. The output unit 104 records and stores the image processing parameters in a memory card or the like.

  Folders and files for holding various data are set on a recording medium such as a memory card. In this case, the recording medium holds image signal data of an optimal image processing parameter by a preset target value of image quality and a calibration image. The storage means for the recording medium includes a storage process for storing a calibration image performed by the storage unit 100 in FIG. 1, a target value setting process for setting a target value performed by the target value setting unit 101, and a feature amount calculating unit. A feature amount calculation process for obtaining a feature amount from the calibration image performed in 102, a parameter calculation process for calculating a parameter for image processing using the target value and the information on the feature amount, performed by a parameter calculation unit 103, The image signal data subjected to is held.

  FIG. 2 shows an example of the configuration of the feature quantity calculation unit 102. The feature quantity calculation unit 102 includes a luminance signal calculation unit 200, a DCT unit 201, a chart extraction unit 300, a buffer 301, and a color information calculation unit 302. The storage unit 100 is connected to the luminance signal calculation unit 200 and the chart extraction unit 300. The luminance signal calculation unit 200 is connected to the DCT unit 201, and the DCT unit 201 is connected to the parameter calculation unit 103. The chart extraction unit 300 is connected to the buffer 301.

  The buffer 301 is connected to the color information calculation unit 302, and the color information calculation unit 302 is connected to the parameter calculation unit 103. The control unit 105 is bi-directionally connected to the luminance signal calculation unit 200, the DCT unit 201, the chart extraction unit 300, and the color information calculation unit 302. The frequency information is calculated by the configuration from the luminance signal calculation unit 200 to the DCT unit 201, and the color information is calculated by the configuration from the chart extraction unit 300 to the color information calculation unit 302.

Based on the control of the control unit 105, one or both of the frequency information and the color information of the calibration image stored in the storage unit 100 can be calculated. The calibration image stored in the storage unit 100 is transferred to one or both of the luminance signal calculation unit 200 and the chart extraction unit 300 based on the control of the control unit 105. Here, the calibration image stored in the storage unit 100 is a three-plate RGB image. The luminance signal calculation unit 200 calculates the luminance signal of the calibration image transferred from the storage unit 100 based on the control of the control unit 105. The luminance signal calculation unit 200 converts the RGB signal into the luminance signal Y based on the conversion equation shown in equation (1).
Y = 0.29900 R +0 .58700 G +0.11400 B (1)
In equation (1), R, G, and B represent R, G, and B signal values, respectively. Next, the calculated luminance signal Y is transferred to the DCT unit 201.

  Here, when the calibration image is a single plate image, for example, the G signal may be the luminance signal Y. The DCT unit 201 converts the Y signal of the calibration image into a frequency space by a known DCT (Discrete Cosine Transform) conversion. The signal converted into the frequency space is transferred to the parameter calculation unit 103 based on the control of the control unit 105. The chart extraction unit 300 separates each chart existing in the calibration image transferred from the storage unit 100 based on the control of the control unit 105, and transfers it to the buffer 301 as a block area of a predetermined size that is slightly smaller than each chart. The separation operation can be automatically performed if the calibration image is shown in the entire image and the arrangement state of the chart is known. The color information calculation unit 302 averages the RGB values of the block area transferred from the buffer 301 based on the control of the control unit 105 for each chart, and transfers the average to the parameter calculation unit 103.

  FIG. 3 shows an example of the configuration of the parameter calculation unit 103. The parameter calculation unit 103 includes a calculation unit 400, a filter coefficient calculation unit 401, an XYZ calculation unit 500, and a matrix coefficient calculation unit 501. Target value setting unit 101 is connected to calculation unit 400 and XYZ calculation unit 500. The feature quantity calculation unit 102 is connected to the calculation unit 400 and the matrix coefficient calculation unit 501. The calculation unit 400 is connected to the filter coefficient calculation unit 401, and the filter coefficient calculation unit 401 is connected to the output unit 104. The XYZ calculation unit 500 is connected to the matrix coefficient calculation unit 501, and the matrix coefficient calculation unit 501 is connected to the output unit 104.

  The control unit 105 is bidirectionally connected to the calculation unit 400, the filter coefficient calculation unit 401, the XYZ calculation unit 500, and the matrix coefficient calculation unit 501. When the feature amount calculated by the feature amount calculation unit 102 and the target value set by the target value setting unit 101 are frequency information, the information of the feature amount calculation unit 102 and the target value setting unit 101 is stored in the calculation unit 400. Forwarded to When the feature amount calculated by the feature amount calculation unit 102 and the target value set by the target value setting unit 101 are color information, the information of the feature amount calculation unit 102 is sent to the matrix coefficient calculation unit 501. Information of the setting unit 101 is transferred to the XYZ calculation unit 500.

First, the calculation unit 400 and the filter coefficient calculation unit 401 will be described. If the converted image obtained by applying a predetermined filter f (x, y) to the pixel value i (x, y) of the calibration image is g (x, y), a convolution like equation (2) Can be expressed as
f (x, y) * i (x, y) = g (x, y) (2)
However, * represents convolution operation, and x and y represent coordinates in the image.

When DCT processing is performed on both sides of equation (2), equations (3) and (4)
F (u, v) × I (u, v) = G (u, v) (3)
F (u, v) = G (u, v) / I (u, v) (4)
Is obtained. Where F (u, v), I (u, v), G (u, v) are DCT transforms of f (x, y), i (x, y), g (x, y), respectively. And u, v represent coordinates in the frequency space.

  The calculation unit 400 uses the frequency characteristic I (u, v) of the calibration image transferred from the feature amount calculation unit 102 and the target frequency characteristic G (u, v) transferred from the target value setting unit 101 ( 4) Calculate F (u, v) using the formula. Next, the filter coefficient calculation unit 401 uses the frequency characteristic F (u, v) calculated from the calculation unit 400 to calculate the filter coefficient f (x, y) using inverse DCT or a known window function method. calculate. The calculated filter coefficient is transferred to the output unit 104.

Next, the XYZ calculation unit 500 and the matrix coefficient calculation unit 501 will be described. The target value setting unit 101 is set with the spectral reflectance corresponding to each chart of the calibration image set via the external I / F unit 106, and the spectral reflectance of each chart is sent to the XYZ calculation unit 500. Transferred. Next, the XYZ calculation unit 500 calculates an XYZ signal value from the spectral reflectance o _i (λ) of the chart i based on the equation (5).

Where λ is a wavelength, K is a constant, E (λ) is illumination light, ￣x (λ), ￣y (λ), and ￣z (λ) are XYZ color matching functions, and X _i , Y _i , Z _i represent XYZ signal values of chart i.

The matrix coefficient calculation unit 501 uses the RGB signal value of each chart transferred from the feature amount calculation unit 102 and the XYZ signal value of each chart transferred from the XYZ calculation unit 500 to convert the RGB signal into the XYZ signal. Create Here, a method for creating a transformation matrix using least square estimation will be described. Assuming that the number of charts is N, the conversion equation for the RGB signal R _k G _k B _k and the XYZ signal X _k Y _k Z _{k of} a certain k-th chart is as shown in equation (6).

  If equation (6) is modified and expressed for all signals, it is expressed by equation (7).

(7) If the matrix of the expression is A, p, b from the left,
Ap = b (8)
It becomes.

Estimating p using least squares estimation gives
p = [A ^T A] ^-1 A ^T b (9)
Therefore, the conversion matrix coefficient can be obtained by using the equation (9). Here, the RGB signal is converted to the XYZ signal using the RGB primary equation as shown in Equation (6), but it may be a quadratic equation, a cubic equation or a higher order. For example, the quadratic expression is expressed as (10).

  FIG. 4 shows an example of another configuration of the first embodiment. 1 differs from FIG. 1 in that a calibration image 110, a lens system 111, a CCD 112, a preprocessing unit 113, a signal processing unit 114, and a compression unit 115 are added. This configuration diagram assumes that a function for calculating image processing parameters is added to the camera device. The same names and numbers are assigned to the same components as those in the first embodiment. Only the points different from the first embodiment will be described below. The calibration image 110 is taken through the lens system 111 and the CCD 112. The captured image is converted into a digital signal through a preprocessing unit 113 that performs processing such as Gain amplification, A / D conversion, AF and AE control. The preprocessing unit 113 is connected to the storage unit 100 and the signal processing unit 114.

  The signal processing unit 114 is connected to the compression unit 115, and the compression unit 115 is connected to the output unit 104. The signal processing unit 114 performs edge enhancement, color conversion processing, and the like on the image transferred from the preprocessing unit 113 and transfers the image to the compression unit 115. At this time, the signal processing unit 114 performs signal processing on the image transferred from the preprocessing unit 113 by using the edge enhancement filter coefficient and the color conversion matrix information calculated by the parameter calculation unit 103. You can also.

  The compression unit 115 compresses the transferred image by publicly known JPEG or the like, and transfers the compressed image to the output unit 104 such as a memory card. Depending on the setting of the external I / F unit 106, processing from the preprocessing unit 113 via the storage unit 100 to the output unit 104 is performed in the image processing parameter calculation mode, and from the preprocessing unit 113 in the normal shooting mode. Processing from the signal processing unit 114 to the output unit 104 is performed. The image processing parameters transferred from the parameter calculation unit 103 to the output unit 104 are transferred in a general-purpose form so that they can be used in software processing such as a PC, but the parameters transferred to the signal processing unit 114 are transferred to the camera device. It is converted into a suitable form and transferred. As described above, the configuration shown in FIG. 4 includes a lens system 111, a CCD 112, a preprocessing unit 113, a signal processing unit 114 and a compression unit 115, an output unit 104, a storage unit 100, a target value setting unit 101. It functions as an imaging apparatus having an image processing parameter mode setting unit including a feature amount calculation unit 102, a parameter calculation unit 103, an output unit 104, and a control unit 105.

  With the above configuration, since the image processing parameter is calculated using the calibration image and the target value, it is possible to calculate a highly accurate image processing parameter that satisfies the target value. In addition, since image processing parameters are assumed to be used in post-processing such as a PC, general-purpose parameters that do not depend on the camera system can be calculated. In addition, since a calibration image is taken and image processing parameters are calculated, it is possible to flexibly cope with changes over time and system changes.

  In the above-described embodiment, processing by hardware is assumed, but it is not necessary to be limited to such a configuration. A configuration in which processing is performed separately by software is also possible. FIG. 5 shows a flowchart relating to the software of the first embodiment. The flowchart in FIG. 5 will be described in correspondence with the processing in FIG. The calibration image is read in S1 corresponding to the storage unit 100, and the feature amount of the calibration image is calculated in S2 corresponding to the feature amount calculation unit 102. Next, in S3 corresponding to the parameter calculation unit 103, the parameter is calculated using the target value set in S4 corresponding to the target value setting unit 101 and the feature amount calculated in S2. The parameter calculation result is output in S5 corresponding to the output unit 104.

  6 to 9 correspond to the second embodiment of the present invention. 6 is a configuration diagram of the second embodiment example, FIG. 7 is a second configuration diagram of the feature amount calculation unit, FIG. 8 is a second configuration diagram of the parameter calculation unit, and FIG. 9 is an example of the second embodiment example. It is a flowchart regarding software.

  A second embodiment will be described with reference to the block diagram of FIG. In the first embodiment shown in FIG. 1, the feature quantity calculation unit 102 and the parameter calculation unit 103 are changed to a feature quantity calculation unit 1002 and a parameter calculation unit 1003, respectively. Further, the difference is that a comparison unit 150 is added. The same names and numbers are assigned to the same components as those in the first embodiment. Only the points different from the first embodiment will be described below. The storage unit 100 is connected to the feature amount calculation unit 1002. The feature amount calculation unit 1002 and the target value setting unit 101 are connected to the comparison unit 150, and the comparison unit 150 is connected to the parameter calculation unit 1003. The parameter calculation unit 1003 is connected to the feature amount calculation unit 1002 and the output unit 104.

  Based on the control of the control unit 105, the comparison unit 150 compares the feature amount transferred from the feature amount calculation unit 1002 with the target value transferred from the target value setting unit 101, and transfers the comparison result to the parameter calculation unit 1003. . Thereafter, using the information of the comparison unit 150, the control unit 105 controls the parameter calculation unit 1003 to transfer the calculated image processing parameter to the feature amount calculation unit 1002 or to the output unit 104. Judging.

  The output unit 104 records and saves the image processing parameters in a recording medium such as a memory card. The storage means for the recording medium includes a storage process for storing a calibration image performed by the storage unit 100 in FIG. 6, a target value setting process for setting a target value performed by the target value setting unit 101, and a feature amount calculation unit. A feature amount calculation process for obtaining a feature amount from the calibration image performed in 102, a parameter calculation process for calculating a parameter for image processing using the target value and the information on the feature amount, performed by a parameter calculation unit 103, Image signal data that has been subjected to a determination process for determining whether to transfer the parameter to the feature amount calculation process or the output process performed by the control unit 105 is held.

  FIG. 7 shows an example of the configuration of the feature quantity calculation unit 1002. The feature amount calculation unit 1002 includes a chart extraction unit 600, a matrix calculation unit 601, a chart Lab calculation unit 602, a color difference calculation unit 603, and a Lab setting unit 604. The storage unit 100 is connected to the chart extraction unit 600. The parameter calculation unit 1003 is connected to the matrix calculation unit 601. The chart extraction unit 600 is connected to the matrix calculation unit 601.

  The matrix calculation unit 601 is connected to the chart Lab calculation unit 602. The chart Lab calculation unit 602 and the Lab setting unit 604 are connected to the color difference calculation unit 603, and the color difference calculation unit 603 is connected to the comparison unit 150. The control unit 105 is bi-directionally connected to the chart extraction unit 600, matrix calculation unit 601, chart Lab calculation unit 602, color difference calculation unit 603, and Lab setting unit 604. The storage unit 100 stores calibration images made up of charts having several different spectral reflectances.

  In this embodiment, it is assumed that the calibration image is an RGB image. Based on the control of the control unit 105, the chart extraction unit 600 separates each chart using the calibration image transferred from the storage unit 100, and extracts a block region having a predetermined size that is slightly smaller than each chart. The extracted signals of the respective block areas are averaged and sequentially transferred to the matrix calculation unit 601. The matrix calculation unit 601 uses the matrix coefficients a to i transferred from the parameter calculation unit 1003 for the RGB average signal of each chart transferred from the chart extraction unit 600 based on the control of the control unit 105 (11). Performs conversion to the XYZ signal indicated by the equation.

Since there is no output from the parameter calculation unit 1003 in the first processing, the initial value of the matrix coefficient can be arbitrarily set by the user via the external I / F unit 106, and the information is calculated by the matrix calculation through the control unit 105. Transferred to part 601. The calculated XYZ signal is transferred to the chart Lab calculation unit 602. The chart Lab calculation unit 602 converts the XYZ signal value of each chart transferred from the matrix calculation unit 601 based on the control of the control unit 105 into an L ^* a ^* b ^* signal expressed by equation (12). I do.

However, X ₀ , Y ₀ , Z ₀ are values on a completely diffusing white surface under a standard light source, and the function f is, for example, f (X / X ₀ ) as shown in Equation (13).

The L ^* a ^* b ^* signal value of the calculated chart is transferred to the color difference calculation unit 603. In the Lab setting unit 604, the true L ^* a ^* b ^* of each chart of the calibration image is set via the external I / F unit 106. The color difference calculation unit 603 controls the L ^* a ^* b ^* signal value of each chart transferred from the chart Lab calculation unit 602 based on the control of the control unit 105 and the true L ^* a ^* b transferred from the Lab setting unit 604. ^* Color difference ΔE (L ^* a ^* b ^* ) is calculated from the signal value. The calculated color difference is transferred to the comparison unit 150. The average RGB value of each chart is also transferred to the comparison unit 150.

  FIG. 8 shows an example of the configuration of the parameter calculation unit 1003. The parameter calculation unit 1003 includes a matrix order setting unit 700, a matrix coefficient calculation unit 701, and an XYZ setting unit 702. The comparison unit 150 is connected to the matrix coefficient calculation unit 701. The matrix order setting unit 700 and the XYZ setting unit 702 are connected to the matrix coefficient calculation unit 701, and the matrix coefficient calculation unit 701 is connected to the feature amount calculation unit 1002 and the output unit 104. The control unit 105 is bidirectionally connected to the comparison unit 150, the matrix order setting unit 700, the matrix coefficient calculation unit 701, and the XYZ setting unit 702.

  The comparison unit 150 compares the color difference of the calibration image calculated by the feature amount calculation unit 1002 with the target color difference transferred from the target value setting unit 101. In the first comparison of the comparison unit 150, if the color difference of the calibration image is larger than the target color difference, the matrix order setting unit 700 is controlled based on the control of the control unit 105, and the matrix order is set to the first order. The information is transferred to the matrix coefficient calculation unit 701. If the color difference of the calibration image is smaller than the target color difference, the initial matrix coefficient is transferred to the output unit 104 based on the control of the control unit 105. The average RGB value of each chart of the calibration image is transferred from the comparison unit 150 to the matrix coefficient calculation unit 701.

  In the XYZ setting unit 702, the true XYZ value of the calibration image chart is set via the external I / F unit 106. Based on the control of the control unit 105, the matrix coefficient calculation unit 701 uses the RGB value of each chart of the calibration image transferred from the comparison unit 150 and the XYZ value of each chart transferred from the XYZ setting unit 702, 9) The conversion matrix coefficient is calculated as shown in equation (9). The calculated conversion matrix coefficient is transferred to the feature amount calculation unit 1002.

  The comparison unit 150 compares the color difference of the calibration image calculated by the feature amount calculation unit 1002 again with the target color difference transferred from the target value setting unit 101. If the color difference of the calibration image is larger than the target color difference, based on the control of the control unit 105, the matrix order setting unit 700 is controlled, the matrix order is set to secondary, and the information is sent to the matrix coefficient calculation unit 701. Forward. If the color difference of the calibration image is smaller than the target color difference, the matrix coefficient calculated by the matrix coefficient calculation unit 701 based on the control of the control unit 105 is transferred to the output unit 104. The matrix order setting unit 700 sequentially increases the order of the matrix according to the comparison result of the comparison unit 150 as described above.

  In the configuration of FIG. 6 as well, a function for calculating image processing parameters can be provided in the camera apparatus as shown in FIG. In this case, a lens system for capturing a calibration image, a CCD, a preprocessing unit that performs processing such as Gain amplification, A / D conversion, AF, and AE control on the captured image, edge enhancement and color for the image A signal processing unit that performs conversion processing and the like, and a compression unit that performs compression such as publicly known JPEG on the transferred image are provided.

  With the above configuration, the feature amount calculated from the calibration image is compared with the set target value, and the image processing parameter is sequentially updated. Therefore, the image processing parameter that satisfies the target value condition can be calculated. In the above embodiment, processing by hardware is assumed, but it is not necessary to be limited to such a configuration. FIG. 9 shows a flowchart relating to the software of the second embodiment.

  The flowchart in FIG. 9 will be described in correspondence with the processing in FIG. The calibration image is read in S1 corresponding to the storage unit 100, and the feature amount of the calibration image is calculated in S10 corresponding to the feature amount calculation unit 1002. Next, in S12 corresponding to the comparison unit 150, the target value set in S4 corresponding to the target value setting unit 101 is compared with the feature amount calculated in S10. If the calculated feature value does not satisfy the target value, the image processing parameter is calculated in S11 corresponding to the parameter calculation unit 1003, and the process returns to the feature value calculation process in S10. If the calculated feature value satisfies the target value, the parameter is output in S5 corresponding to the output unit 104.

  10 to 14 are diagrams showing a third embodiment. FIG. 10 is a configuration diagram of the third embodiment, FIG. 11 is a third configuration diagram of the feature amount calculation unit, FIG. 12 is a configuration diagram of the image processing unit, FIG. 13 is an explanatory diagram regarding the functionalization of the noise amount, These are the flowcharts regarding the software of the third embodiment.

  A third embodiment will be described with reference to the block diagram of FIG. 1 differs from FIG. 1 in that the feature quantity calculation unit 102 and the parameter calculation unit 103 are changed to a feature quantity calculation unit 1102 and an image processing unit 151, respectively. The same names and numbers are assigned to the same configurations as those in the first embodiment and the second embodiment. Only different points will be described below. The storage unit 100 is connected to the feature amount calculation unit 1102 and the image processing unit 151. The image processing unit 151 is connected to the feature amount calculation unit 1102 and the output unit 104.

  Only portions different from FIG. 1 and FIG. 6 in FIG. 10 will be described. The image processing unit 151 performs predetermined image processing on the calibration image transferred from the storage unit based on the control of the control unit 105, using the information of the comparison unit 150, and the image of the processing result is the feature amount calculation unit 1102. Forward to. Based on the control of the control unit 105, the comparison unit 150 compares the feature amount transferred from the feature amount calculation unit 1102 with the target value transferred from the target value setting unit 101, and transfers the comparison result to the image processing unit 151. . Thereafter, using the information of the comparison unit 150, the control unit 105 controls the image processing unit 151 to retransfer the image on which the predetermined image processing has been performed to the feature amount calculation unit 1002, or to perform image processing parameters. Is transferred to the output unit 104.

  The output unit 104 records and saves the image processing parameters in a recording medium such as a memory card. The storage means for the recording medium includes a storage process for storing a calibration image performed by the storage unit 100 in FIG. 10, a target value setting process for setting a target value performed by the target value setting unit 101, and a feature amount calculating unit. A feature amount calculation process for obtaining a feature amount from the calibration image performed at 1102, a comparison process for comparing the target value with the target amount performed by the comparison unit 105, and a comparison process performed by the image processing unit 151. A parameter calculation process for calculating image processing parameters using information; an image process using the parameters performed by the image processing unit 151; and a feature amount calculation process for the image processed image performed by the control unit 105 The image signal data subjected to the determination process for determining whether to transfer to the output process or the parameter calculated in the parameter calculation process is held.

  FIG. 11 shows an example of the configuration of the feature quantity calculation unit 1102. The feature amount calculation unit 1102 includes a chart extraction unit 800, an average variance calculation unit 801, and a noise calculation unit 802. The storage unit 100 and the image processing unit 151 are connected to the chart extraction unit 800. The chart extraction unit 800 is connected to the average variance calculation unit 801. The average variance calculation unit 801 is connected to the noise calculation unit 802. The noise calculation unit 802 is connected to the comparison unit 150. The control unit 105 is bidirectionally connected to the chart extraction unit 800, the average variance calculation unit 801, and the noise calculation unit 802. The storage unit 100 stores calibration images made up of charts having several different spectral reflectances. This image may be an RGB image or a monochrome image.

  The chart extraction unit 800 separates each chart existing in the calibration image transferred from the storage unit 100 based on the control of the control unit 105, and sequentially calculates an average variance calculation unit 801 as a block region of a predetermined size that is slightly smaller than each chart. Forward to. The separation operation can be automatically performed if the calibration image is shown in the entire image and the arrangement state of the chart is known. Here, the signal value of the block area transferred from the chart extraction unit 800 may be an RGB signal or a luminance signal converted according to the above equation (1).

  The average variance calculation unit 801 calculates the average and variance of the block area signals transferred from the chart extraction unit 800 based on the control of the control unit 105. A standard deviation is calculated as a noise amount from the calculated variance, and is sequentially transferred to the noise calculation unit 802. Based on the control of the control unit 105, the noise calculation unit 802 averages the standard deviation of each transferred chart, that is, the noise amount, and calculates the average noise amount of the calibration image. The calculated noise amount of the calibration image is transferred to the comparison unit 150.

  FIG. 12 shows an example of the configuration of the image processing unit 151. The image processing unit 151 includes a local area extraction unit 900, a buffer 901, an average luminance calculation unit 902, a noise reduction unit 903, a noise amount calculation unit 904, a parameter ROM 905, and a gain setting unit 906. The storage unit 100 is connected to the local region extraction unit 900. The local area extraction unit 900 is connected to the buffer 901. The buffer 901 is connected to the average luminance calculation unit 902.

  The average luminance calculation unit 902 is connected to the noise reduction unit 903 and the noise amount calculation unit 904. The parameter ROM 905 is also connected to the noise amount calculation unit 904. The noise amount calculation unit 904 and the gain setting unit 906 are connected to the noise reduction unit 903, and the noise reduction unit 903 is connected to the feature amount calculation unit 1102. The gain setting unit 906 is connected to the output unit 104. The control unit 105 is bidirectionally connected to the local region extraction unit 900, the average luminance calculation unit 902, the noise reduction unit 903, the noise amount calculation unit 904, and the gain setting unit 906.

  The local region extraction unit 900 extracts a local region centered on a predetermined target pixel of the calibration image transferred from the storage unit 100 based on the control of the control unit 105, for example, a 5 × 5 region, and transfers the extracted region to the buffer 901. To do. The average luminance calculation unit 902 converts the local area transferred from the buffer 901 into a luminance signal based on the equation (1), and calculates the average luminance signal of the local area. The calculated average luminance signal is transferred to the noise amount calculation unit 904 and the noise reduction unit 903. The local region signal in the buffer 901 is also transferred to the noise reduction unit 903. The noise amount calculation unit 904 obtains function information used for noise amount calculation from the parameter ROM 905 based on the control of the control unit 105.

As the function of the noise amount N, for example, a function that increases in the form of a power to the luminance value Y as shown in FIG. 13 can be considered. This is in the form of equation (14).
N = αY ^β + γ (14)
Here, α, β, and γ are constants. In addition, the amount of noise increases and decreases depending on the ISO sensitivity of the calibration image.

When formula (14) is expanded, formula (15)
^{N = αiY βi + γi (15} )
become that way. Here, i is a parameter indicating the amplification factor. The parameter ROM 905 records constant terms αi, βi, and γi. The noise amount calculation unit 904 reads the constant terms of αi, βi, and γi from the parameter ROM 604 based on the amplification factor.

  The noise amount calculation unit 904 calculates the noise amount based on the equation (15) from the average luminance transferred from the average luminance calculation unit 902. The noise amount is assumed to be the noise amount of the central pixel of the region extracted by the local region extraction unit 900. The calculated noise amount is transferred to the noise reduction unit 903. The noise reduction unit 903 uses the central pixel I of the local area transferred from the average luminance calculation unit 902, the noise amount N transferred from the noise amount calculation unit 904, and the gain value μ transferred from the gain setting unit 906, It is determined whether the central pixel in the local area is noise based on the equation (16).

  Here, ￣I represents an average pixel in the local region. If it is determined to be noise, the center pixel is replaced with the average value of the local area, otherwise it is determined to have a structure such as an edge or texture, and the process for the center pixel is not performed. As the value of the gain μ increases, the number of portions determined to be noise increases, and the effect of noise reduction processing increases. The comparison unit 150 compares the target noise amount set by the target value setting unit 101 with the noise amount of the calibration image calculated by the feature amount calculation unit 1102, and if the target noise amount is not satisfied, the value of the gain μ is compared. Increase it gradually. When the noise amount of the calibration image is less than or equal to the target noise amount, the gain μ set by the gain setting unit 906 is transferred to the output unit 104.

  Also in the configuration of FIG. 10, a function for calculating image processing parameters can be provided in the camera apparatus as shown in FIG. In this case, a lens system for capturing a calibration image, a CCD, a preprocessing unit that performs processing such as Gain amplification, A / D conversion, AF, and AE control on the captured image, edge enhancement and color for the image A signal processing unit that performs conversion processing and the like, and a compression unit that performs compression such as publicly known JPEG on the transferred image are provided.

  With the above configuration, the feature amount calculated from the calibration image is compared with the set target value, and the image processing parameter is sequentially updated. Therefore, the image processing parameter that satisfies the target value condition can be calculated. In the above embodiment, processing by hardware is assumed, but it is not necessary to be limited to such a configuration. FIG. 14 shows a flowchart regarding the software of the third embodiment.

  The flowchart in FIG. 14 will be described in correspondence with the processing in FIG. A calibration image is read in S1 corresponding to the storage unit 100, and predetermined image processing is performed on the calibration image in S20 corresponding to the image processing unit 151. Next, in S21 corresponding to the feature amount calculation unit 1102, the feature amount of the calibration image after image processing is calculated, and in S12 corresponding to the comparison unit 150, S4 corresponding to the target value setting unit 101 The target value set in step S2 is compared with the feature amount calculated in step S21. If the calculated feature amount does not satisfy the target value, image processing is performed again in step S20. If the calculated feature value satisfies the target value, the parameter is output in S5 corresponding to the output unit 104.

  As described above, according to the present invention, an image processing device, an imaging device, a recording medium, and an image processing program capable of calculating noise amounts, frequency characteristics, color information, and color differences and determining image processing parameters. Can be provided.

1 is a configuration diagram of a first embodiment example. FIG. It is a block diagram of the feature-value calculation part. It is a block diagram of a parameter calculation part. FIG. 5 is another configuration diagram of the first embodiment. 3 is a flowchart relating to software of the first embodiment. FIG. 6 is a configuration diagram of a second exemplary embodiment. FIG. 6 is a second block diagram of the feature quantity calculation unit. FIG. 6 is a second block diagram of the parameter calculation unit. 10 is a flowchart relating to software of a second exemplary embodiment. FIG. 10 is a configuration diagram of a third exemplary embodiment. FIG. 10 is a third block diagram of the feature quantity calculation unit. It is a block diagram of an image processing part. It is explanatory drawing regarding functionalization of the amount of noise. 12 is a flowchart relating to software of the third embodiment.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100 ... Memory | storage part, 101 ... Target value setting part, 102 ... Feature-value calculation part, 103 ... Parameter calculation part, 104 ... Output part, 105 ... Control part, 106 ... External I / F unit, 110 ... calibration image, 111 ... lens system, 112 ... CCD, 113 ... pre-processing unit, 114 ... signal processing unit, 115 ... compression unit , 150 ... comparison unit, 151 ... image processing unit

Claims (21)

In an image processing apparatus for processing a digitized image, a storage means for storing a calibration image in the image processing apparatus, a target value setting means for setting a target value, and a feature quantity for obtaining a feature quantity from the calibration image An image processing apparatus comprising: a calculation unit; a parameter calculation unit that calculates an image processing parameter using the target value and the feature amount information; and an output unit that outputs the parameter. The feature amount calculating unit includes: a luminance value calculating unit that calculates a luminance value of the calibration image; and a frequency calculating unit that performs frequency conversion on the luminance value to calculate a frequency component. The image processing apparatus according to claim 1. The feature amount calculating means includes a chart extracting means for extracting a plurality of chart portions included in the calibration image, and a color information calculating means for calculating color information from each chart. The image processing apparatus according to 1. The parameter calculation means includes a calculation means for calculating using information of the target value setting means and the feature amount calculation means, and a filter for calculating a filter used for edge enhancement processing using information obtained from the calculation means 2. The image processing apparatus according to claim 1, further comprising a calculation unit. 2. The image processing apparatus according to claim 1, wherein the parameter calculation means includes conversion matrix calculation means for calculating a conversion matrix used for color conversion. In an image processing apparatus for processing a digitized image, a storage means for storing a calibration image in the image processing apparatus, a target value setting means for setting a target value, and a feature quantity for obtaining a feature quantity from the calibration image A calculation unit; a comparison unit that compares the target value with the feature quantity; a parameter calculation unit that calculates an image processing parameter using information of the comparison unit; an output unit that outputs the parameter; and the parameter An image processing apparatus comprising: a determination unit that determines whether to transfer to the feature amount calculation unit or to the output unit. The feature amount calculating means includes a chart extracting means for extracting a plurality of chart portions included in the calibration image, a color information extracting means for extracting color information from each chart, and a true value of the color information of each chart. The image processing apparatus according to claim 6, comprising color information true value setting means for setting, and color difference calculation means for calculating a difference between the color information and the true value. The image processing apparatus according to claim 6, wherein the parameter calculation unit includes a conversion matrix calculation unit that calculates a conversion matrix used for color conversion, and a matrix order setting unit that sets the order of the conversion matrix. In an image processing apparatus for processing a digitized image, a storage means for storing a calibration image in the image processing apparatus, a target value setting means for setting a target value, and a feature quantity for obtaining a feature quantity from the calibration image A calculation unit; a comparison unit that compares the target value with the feature quantity; a parameter calculation unit that calculates an image processing parameter using information of the comparison unit; and an image processing unit that performs image processing using the parameter And determining whether to transfer the image processed by the image processing unit to the feature amount calculating unit or to transfer the parameter calculated by the parameter calculating unit to the output unit. And an image processing apparatus. The feature amount calculating unit includes a chart extracting unit that extracts a plurality of chart portions included in the calibration image, and a noise amount calculating unit that calculates a noise amount from each chart. The image processing device according to claim 9. The image processing means includes a local area extracting means for extracting a local area centered on a target pixel of an image, a noise amount calculating means for calculating a noise amount for the target pixel, and a gain setting for setting a gain for the noise amount. The image processing apparatus according to claim 9, further comprising: means; and noise reduction means for performing noise reduction processing on the local region using the noise amount and the gain information. The image processing apparatus according to claim 1, wherein the storage unit includes an external storage unit that can be attached to and detached from the image processing apparatus. An imaging means for taking an image, a preprocessing means for converting the taken image signal into a digital signal, a signal processing means for the digital signal, and an output means for transferring an output signal of the signal processing means Shooting mode setting section, and
Storage means for storing a calibration image photographed by the photographing means, target value setting means for setting a target value, feature quantity calculation means for obtaining a feature quantity from the calibration image, the target value and the feature quantity An image processing parameter mode setting unit including a parameter calculation unit that calculates image processing parameters using the information and an output unit that outputs the parameters, and the parameter calculation unit converts the parameters for shooting An image pickup apparatus for transferring image processing parameters to the signal processing means. An imaging means for taking an image, a preprocessing means for converting the taken image signal into a digital signal, a signal processing means for the digital signal, and an output means for transferring an output signal of the signal processing means Shooting mode setting section, and
Storage means for storing a calibration image photographed by the photographing means; target value setting means for setting a target value; feature quantity calculating means for obtaining a feature quantity from the calibration image; the target value and the feature quantity Comparing means, parameter calculating means for calculating parameters of image processing using information of the comparing means, output means for outputting the parameters, and transferring the parameters to the feature amount calculating means or the output And an image processing parameter mode setting unit configured to determine whether to transfer to the means, wherein the parameter calculation means transfers the image processing parameters converted for photographing to the signal processing means. An imaging device. An imaging means for taking an image, a preprocessing means for converting the taken image signal into a digital signal, a signal processing means for the digital signal, and an output means for transferring an output signal of the signal processing means Shooting mode setting section, and
Storage means for storing a calibration image photographed by the photographing means; target value setting means for setting a target value; feature quantity calculating means for obtaining a feature quantity from the calibration image; the target value and the feature quantity Comparing means, parameter calculating means for calculating image processing parameters using information of the comparing means, image processing means for performing image processing using the parameters, output means for outputting the parameters, Image processing parameter mode setting comprising: determination means for determining whether to transfer the image processed by the image processing means to the feature amount calculation means or to transfer the parameter calculated by the parameter calculation means to the output means And the parameter calculation means transfers the image processing parameters converted for photographing to the signal processing means. Location. In a recording medium on which image signal data of an optimal image processing parameter is recorded by a preset image quality target value and a calibration image,
Storage processing for storing a calibration image, target value setting processing for setting a target value, feature amount calculation processing for obtaining a feature amount from the calibration image, and image processing using information on the target value and the feature amount A recording medium comprising: means for holding image signal data that has been subjected to parameter calculation processing for calculating the parameters of In a recording medium on which image signal data of an optimal image processing parameter is recorded by a preset image quality target value and a calibration image,
A storage process for storing a calibration image in the image processing apparatus, a target value setting process for setting a target value, a feature quantity calculation process for obtaining a feature quantity from the calibration image, and a comparison between the target value and the feature quantity Comparison processing, parameter calculation processing for calculating image processing parameters using the information of the comparison processing, output processing for outputting the parameters, and transfer of the parameters to the feature amount calculation processing or to the output processing A recording medium comprising: means for holding image signal data subjected to a determination process for determining whether to transfer. In a recording medium on which image signal data of an optimal image processing parameter is recorded by a preset image quality target value and a calibration image,
A storage process for storing a calibration image, a target value setting process for setting a target value, a feature quantity calculation process for obtaining a feature quantity from the calibration image, a comparison process for comparing the target value and the feature quantity, A parameter calculation process that calculates image processing parameters using the information of the comparison process; an image process that uses the parameters; and the image processed image is transferred to the feature amount calculation process or the parameter calculation process A recording medium, comprising: means for holding image signal data subjected to a determination process for determining whether to transfer the calculated parameter to the output process. A procedure for reading a calibration image into a computer; a procedure for calculating a feature quantity of the calibration image; a procedure for calculating a parameter for image processing using the set target value and the calculated feature quantity; An image processing program for executing a procedure for outputting a parameter calculation result. A procedure for reading a calibration image into a computer, a procedure for calculating a feature quantity of the calibration image, a procedure for comparing a set target value with the calculated feature quantity, and the calculated feature quantity A procedure for calculating an image processing parameter if it does not satisfy the target value, a procedure for feeding back the parameter to the feature amount calculation processing, and a procedure for outputting the parameter if the calculated feature amount satisfies the target value; An image processing program for executing A procedure for reading a calibration image into a computer, a procedure for performing predetermined image processing on the calibration image, a procedure for calculating a feature amount of the calibration image after the image processing, a set target value, A procedure for comparing the calculated feature values of the calibration image, a procedure for returning to the image processing if the calculated feature value does not satisfy the target value, and the calculated feature value not satisfying the target value. An image processing program for executing the procedure for outputting the parameter.

Images