JP5910828B2 - Image evaluation apparatus and image evaluation program - Google Patents

Description

  The present invention relates to an image evaluation apparatus and an image evaluation program.

  For a color image formed on a medium, the quality of the image may be a problem. As a method of evaluating the quality of image quality, there are psychological evaluation for quantifying the quality that humans see and feel, and physical evaluation based on a value obtained by measuring physical characteristics of an image. Psychological evaluation is widely used, but varies depending on the evaluator's difference, fatigue / physical condition, surrounding environment, and the like. Furthermore, the evaluation value is generally a discrete value or only pass / fail, and it is difficult to handle minute differences compared to physical evaluation. For example, it is unsuitable for applications such as finding a sign of occurrence of a malfunction from the tendency of image quality degradation that occurs while the image forming apparatus continues to be used.

Examples of physical evaluation include those described in Patent Document 1, for example. In this patent document 1, tristimulus values X, Y, and Z including two-dimensional position information about an image to be evaluated are captured by a two-dimensional scanning colorimeter, converted into L ^* , a ^* , and b ^* , and brightness. Is converted into information L ^* , C ^* , h ° representing the saturation, hue, brightness difference, saturation difference, hue difference that is the difference from the average value of brightness, saturation, hue for the entire image to be evaluated Information ΔL ^* , ΔC ^* , ΔH is calculated. Then, the image to be evaluated is divided into a plurality of images (data groups), and each image is subjected to two-dimensional Fourier transform on ΔL ^* , ΔC ^* , ΔH to calculate a two-dimensional power spectrum, and then two-dimensional The power spectrum is made one-dimensional, the one-dimensional power spectrum is averaged, multiplied by a function representing the spatial frequency characteristic of the human visual system, and integrated to calculate an image evaluation value.

  As visual characteristics, in addition to characteristics according to spatial frequency, there are characteristics according to lightness and color. For example, there are visual characteristics corresponding to lightness and color such that unevenness and granularity are not conspicuous at low lightness, and unevenness of color becomes less conspicuous than density unevenness as it approaches gray. It is desired to make corrections according to these visual characteristics to match the evaluation values obtained by physical evaluation with human senses. In Patent Document 1 described above, correction is performed according to the visual spatial frequency characteristics, but correction according to the visual characteristics of brightness and color is not performed.

For example, in Patent Document 2, RGB signals from a color scanner are converted into L ^* a ^* b ^* , a spatial frequency component is calculated from the converted signal, and visual spatial frequency characteristics are calculated with respect to the calculated spatial frequency component. A method is described in which an evaluation value of the a ^* and b ^* components is calculated by performing the correction according to the calculation, and the evaluation value of the lightness component is corrected and output. This method corrects the evaluation value of the lightness component, but does not perform correction according to the visual characteristics of the color.

JP 2000-207560 A Japanese Patent Laid-Open No. 10-023191

  SUMMARY OF THE INVENTION An object of the present invention is to provide an image evaluation apparatus and an image evaluation program that can calculate an evaluation value of an image quality suitable for a human sense regardless of color and density.

According to the first aspect of the present invention, the generation means for generating the spatial frequency distribution for the L ^* component and the a ^* and b ^* components of the image to be evaluated, and a predetermined person based on the spatial frequency distribution Calculation means for calculating an integral value of a frequency band sensitive to vision as an image evaluation value, and correcting the image evaluation value of the L ^* component according to the average brightness of the image to be evaluated and monotonously increasing with respect to the average brightness and brightness correction means for correcting an image evaluation value of the b ^* component using a function that monotonically decreases with respect to the a ^* component image evaluation value to correct also the average brightness of using the function, the evaluation object image An image evaluation apparatus comprising color correction means for correcting the image evaluation values of the a ^* and b ^* components in accordance with the average color value.

The invention according to the claims 2, wherein said color correction means in the invention according to the claims 1, a ^*, using a function that increases monotonically with respect to the absolute value of the average of the b ^* component a ^*, b ^* An image evaluation apparatus characterized by correcting an image evaluation value of a component.

  An invention according to claim 3 of the present application is an image evaluation program for causing a computer to execute the function of the image evaluation apparatus according to claim 1 or claim 2.

According to the first aspect of the present invention, the a ^* component and the b ^* component can be corrected for the evaluation value according to the visual characteristic with respect to the brightness, and the human sense can be obtained regardless of the color and density. It is possible to calculate the evaluation value of the image quality suitable for.

According to the invention described in claim 2 of the present application, correction of the a ^* component and b ^* component according to the visual characteristics for the color can be performed on the evaluation value.

  According to the invention described in claim 3 of the present application, the effect of the invention described in claim 1 or claim 2 of the present application can be obtained.

It is a block diagram which shows one Embodiment of this invention. It is a flowchart which shows an example of the operation | movement in one Embodiment of this invention. It is explanatory drawing of the specific example of the production | generation process of the spatial frequency distribution in a spatial frequency distribution production | generation part. It is explanatory drawing of the specific example of the process which calculates the image evaluation value in the evaluation value calculation part. It is explanatory drawing of the specific example of the correction function about each component in the brightness characteristic correction | amendment part. It is explanatory drawing of the specific example of the correction function about each component in the color characteristic correction | amendment part. It is explanatory drawing of an example of the correlation with an image evaluation value and a comprehensive evaluation value, and psychological evaluation. FIG. 3 is an explanatory diagram of an example of a computer program, a storage medium storing the computer program, and a computer when the function described in the embodiment of the present invention is realized by a computer program.

  FIG. 1 is a configuration diagram showing an embodiment of the present invention. In the figure, 11 is a spatial frequency distribution generation unit, 12 is an evaluation value calculation unit, 13 is a visual spatial frequency characteristic correction unit, 14 is an integration unit, 15 is a lightness characteristic correction unit, 16 is a color characteristic correction unit, and 17 is a total. An evaluation value calculation unit. In this example, it is assumed that the image to be evaluated is a CIELAB color space image. For example, for an image in another color space such as an RGB color space image read by an image reading device, color space conversion processing to the CIELAB color space may be performed in advance. Of course, it goes without saying that an image in a color space having lightness and color components other than the CIELAB color space may be an evaluation target. The image to be evaluated may be a part of an image.

The spatial frequency distribution generation unit 11 generates a spatial frequency distribution for the brightness component and the color component of the given evaluation target image, in this example, the L ^* component, the a ^* component, and the b ^* component. When generating the spatial frequency distribution, the frequency component (DC component) having a frequency of 0 indicates the average brightness (L ^* ) and color (a ^* , b ^* ) values of the image to be evaluated. May be used in the brightness characteristic correction unit 15 and the color characteristic correction unit 16 described later.

  Based on the spatial frequency distribution generated by the spatial frequency distribution generation unit 11, the evaluation value calculation unit 12 calculates, as an image evaluation value, an integral value of a predetermined frequency band that is sensitive to human vision. In this example, the evaluation value calculation unit 12 includes a luminous spatial frequency characteristic correction unit 13 and an integration unit 14.

  The luminosity spatial frequency characteristic correction unit 13 corrects each of the spatial frequency distributions generated by the spatial frequency distribution generation unit 11 by imitating visual characteristics with respect to human spatial frequencies.

  The integrating unit 14 uses the spatial frequency distribution corrected by the luminous spatial frequency characteristic correcting unit 13 and integrates it in a predetermined frequency band to obtain an image evaluation value of each component. This frequency band may be a frequency band that is sensitive to human vision.

The lightness characteristic correction unit 15 corrects the image evaluation value of each component calculated by the evaluation value calculation unit 12 according to the average lightness of the image to be evaluated. In this correction, the image evaluation value calculated by the evaluation value calculation unit 12 is corrected according to the visual characteristics for human brightness. As an example of the correction, the image evaluation value of the brightness component may be corrected using a function that monotonously increases with respect to the average brightness. Among the color components, a function that monotonically increases with respect to the average brightness for the image evaluation value of the a ^* component, and a function that monotonously decreases with respect to the average brightness for the image evaluation value of the b ^* component. It is good to correct using.

The color characteristic correction unit 16 corrects the image evaluation value of the color component according to the average color value of the image to be evaluated. In this example, according to the average value of the average value and the b ^* component of the a ^* component, to correct the image evaluation values of a ^* component and the b ^* component. In this correction, correction according to visual characteristics for human color components is performed. An example of a correction, the a ^* component, using a function that increases monotonically with respect to the absolute value of the average of b ^* components, it is preferable to correct the image evaluation values of a ^* component and the b ^* component.

The comprehensive evaluation value calculation unit 17 uses the image evaluation value of the lightness component corrected by the lightness characteristic correction unit 15 and the image evaluation value of the color component (a ^* component, b ^* component) corrected by the color characteristic correction unit 16. The overall evaluation value is calculated. This comprehensive evaluation value may be an evaluation value related to the image quality of the image to be evaluated. Note that the image evaluation value of the lightness component corrected by the lightness characteristic correction unit 15 and the image evaluation value of the a ^* component and b ^* component corrected by the color characteristic correction unit 16 may be used as final evaluation values. In this case, the overall evaluation value calculation unit 17 may be omitted.

FIG. 2 is a flowchart showing an example of the operation in the embodiment of the present invention. An example of the operation of the above configuration will be described using specific examples. In the following description, as an example, it is assumed that the image to be evaluated is an image in the CIELAB color space, and processing is performed for each of the L ^* (lightness) component, the a ^* component, and the b ^* component. As described above, for example, in the case of an image in another color space such as an image in the RGB color space read by the image reading device, the color space conversion to the CIELAB color space may be performed. Various processes may be performed in advance. For example, when the size of the image area to be evaluated is insufficient, the image may be expanded by an image edge folding process or the like. If the image read by the image reader is a halftone image with halftone dots and stripes, adjust the reading resolution or use multiple resolutions to read the interference between the resolution and halftone dots and stripes. Moire fringes may be avoided.

The spatial frequency distribution generation unit 11 generates a spatial frequency distribution for each component. Here, in S1, a discrete two-dimensional Fourier transform is performed as a two-dimensional orthogonal transform on each of the L ^* component, a ^* component, and b ^* component to generate a two-dimensional spatial frequency distribution. Further, in this example, in S2, a one-dimensional spatial frequency distribution is generated from the two-dimensional spatial frequency distribution. Since the frequency characteristic in the human visual system does not depend on the direction in the two-dimensional spatial frequency distribution, it is preferable to generate a one-dimensional spatial frequency distribution. Of course, it goes without saying that the subsequent processing may be performed using a two-dimensional spatial frequency distribution. Further, an orthogonal transformation technique other than the Fourier transformation may be used.

  FIG. 3 is an explanatory diagram of a specific example of spatial frequency distribution generation processing in the spatial frequency distribution generation unit. FIG. 3A shows an image to be evaluated for each component, and the image portion is shown with diagonal lines. FIG. 3B schematically shows an example of a two-dimensional spatial frequency distribution obtained by performing discrete two-dimensional Fourier transform on each component of the image to be evaluated.

  In the two-dimensional spatial frequency distribution, the distance from the origin corresponds to the frequency. As shown in FIG. 3C, by integrating in concentric circles, an integrated value is obtained for each frequency component. Furthermore, since the length of a circumference changes with frequency components, it cannot be overemphasized that you may normalize by a frequency component.

  A one-dimensional frequency distribution is generated by the value for each frequency component thus obtained. FIG. 3D shows an example of the generated one-dimensional frequency distribution, where the horizontal axis is the frequency component, and the vertical axis is the integrated value or a value obtained by normalizing the integrated value. In this example, frequencies below a certain value are omitted. When the frequency is 0, it is called a DC component or the like, and is handled separately from the frequency component (AC component). The DC component indicates the average of the respective components, and this may be used in the lightness characteristic correction unit 15 and the color characteristic correction unit 16.

  Returning to FIG. 2, the evaluation value calculation unit 12 calculates an image evaluation value based on the spatial frequency distribution generated by the spatial frequency distribution generation unit 11. As the processing, first in S3, the visual spatial frequency characteristic correcting unit 13 corrects each of the spatial frequency distributions generated by the spatial frequency distribution generating unit 11 by imitating visual characteristics for human spatial frequencies. Do. In S4, the integration unit 14 uses the spatial frequency distribution corrected by the visual spatial frequency characteristic correction unit 13 in S3, and integrates in a predetermined frequency band to obtain an image evaluation value of each component.

  FIG. 4 is an explanatory diagram of a specific example of processing for calculating an image evaluation value in the evaluation value calculation unit 12. FIG. 4A shows a certain component again in the example of the one-dimensional spatial frequency distribution shown in FIG. Human visual characteristics differ depending on the spatial frequency, and the degree of visual recognition at a certain frequency is maximized. The higher the frequency is, the less likely it is to be visually recognized. An example of such visual characteristics with respect to human spatial frequency is shown in FIG. Here, visual characteristics with respect to human spatial frequency for each component are shown.

  The one-dimensional spatial frequency distribution in each component obtained from the image to be evaluated shown in FIG. 4A is multiplied by the function indicating the visual characteristic shown in FIG. 4B corresponding to each component. For example, the corrected spatial frequency distribution shown in FIG. 4C is obtained. Such a corrected one-dimensional spatial frequency distribution is obtained for each component. The obtained spatial frequency distribution after correction is a frequency distribution according to the visual characteristics with respect to the human spatial frequency. Although correction may be performed according to visual characteristics for the two-dimensional spatial frequency distribution, the frequency characteristics in the human visual system as described above do not depend on the direction in the two-dimensional spatial frequency distribution. The calculation time is shortened when the one-dimensional spatial frequency distribution is generated and then corrected.

  Using the one-dimensional spatial frequency distribution obtained and corrected for each component, the integrating unit 14 integrates in a predetermined frequency band to obtain an image evaluation value of each component. As a specific example, the predetermined frequency band may be in the range of 0.1 cycle / mm to 1.0 cycle / mm. This frequency band is a region that is perceived as a sense of unevenness rather than a sense of roughness in the observer, and is called, for example, a mottle. This is a frequency band that is sensitive to human vision. FIG. 4C shows this frequency band. When the corrected one-dimensional spatial frequency distribution is integrated in this frequency band, an image evaluation value related to mottle is calculated. Of course, the frequency band to be integrated is not limited to this example, and the integration start and end frequencies may be determined in advance. The frequency range may be determined according to the item to be evaluated. For example, in a frequency range higher than the frequency range in this example, an image evaluation value related to graininess, such as a rough feeling, is calculated.

Returning to FIG. 2, in S <b> 5, the lightness characteristic correction unit 15 uses the average lightness of the image to be evaluated and visually compares the image evaluation values of the respective components calculated by the evaluation value calculation unit 12 with respect to human lightness. Perform correction according to the characteristics. When the image evaluation value calculated by the evaluation value calculation unit 12 is NV and the average brightness of the image to be evaluated is Lave, the image evaluation value LNV after the correction of the brightness characteristic is
LNV = (Lave / 100) ^α · NV ((Lave / 100) ^α <β)
LNV = β · NV ((Lave / 100) ^α ≧ β)
It can be obtained by α is a lightness correction coefficient, which is determined by the visual characteristics of each component with respect to human lightness. Β is a value indicating the upper limit of (Lave / 100) ^α . The correction process may be performed without providing β.

FIG. 5 is an explanatory diagram of a specific example of a correction function for each component in the brightness characteristic correction unit 15. 5A shows an example of a lightness (L ^* ) component correction function, FIG. 5B shows an example of an a ^* component correction function, and FIG. 5C shows an example of a b ^* component correction function. , Respectively. This correction function is the above-mentioned (Lave / 100) ^α , and the lightness correction coefficient α differs depending on each component.

The lightness component has a characteristic that the lower the lightness, the harder it is to be seen by human senses. The image evaluation value of the lightness component may be corrected using a function that monotonously increases with respect to the average lightness. An example of the lightness (L ^* ) component correction function shown in FIG. 5A shows a case where the lightness correction coefficient 0 <α <1, and monotonously increases with respect to the average lightness shown on the horizontal axis.

The a ^* component also has a characteristic that it becomes difficult to be seen by human senses as the lightness is low, and the image evaluation value of the a ^* component is corrected using a function that monotonously increases with respect to the average lightness. Good. An example of the correction function for the a ^* component shown in FIG. 5B shows a case where the brightness correction coefficient α> 1, and monotonously increases with respect to the average brightness shown on the horizontal axis.

On the other hand, the b ^* component has a characteristic that the lower the brightness, the easier it is to be seen by human senses. Therefore, the image evaluation value of the b ^* component may be corrected using a function that monotonously decreases with respect to the average brightness. An example of the b ^* component correction function shown in FIG. 5C shows a case where the lightness correction coefficient α <0, which basically decreases monotonously with respect to the average lightness shown on the horizontal axis. In this example, the upper limit is β, and is β when the value of the correction function is equal to or greater than β.

  Using such a correction function corresponding to each component, the image evaluation value obtained for each component is corrected by visual characteristics with respect to human brightness. Note that the correction function shown in FIG. 5 is an example, and needless to say, it is not limited to this example.

Returning to FIG. 2, in S <b> 6, the color characteristic correction unit 16 uses the average color value of the image to be evaluated and uses the average color value of the image to be evaluated for the image evaluation value of the color component corrected by the lightness characteristic correction unit 15. Perform correction according to the characteristics. Assuming that the image evaluation values of the color components corrected by the lightness characteristic correcting unit 15 are LNVa and LNVb, and the average of the a ^* and b ^* components of the image to be evaluated are aave and bave, the image evaluation value CNVa after correcting the color characteristics. , CNVb is
CNVa = ((1 / (1 + exp (−g | aave |))) − 0.5) · LNVa
CNVb = ((1 / (1 + exp (−g | bave |))) − 0.5) · LNVb
It can be obtained by g is a color correction coefficient, and is determined by the visual characteristics of the human color in each of the a ^* component and b ^* component.

FIG. 6 is an explanatory diagram of a specific example of a correction function for each component in the color characteristic correction unit 16. 6A shows an example of the correction function for the a ^* component, and FIG. 6B shows an example of the correction function for the b ^* component. The correction functions are ((1 / (1 + exp (−g | aave |))) − 0.5) and ((1 / (1 + exp (−g | bave |))) − 0.5) described above. The color correction coefficient g differs depending on each component.

  The color component has a characteristic that as the deterioration of the color image quality becomes closer to an achromatic color, it becomes harder to be seen by human senses and uneven density is more visible. Therefore, even if the image evaluation value is a color close to an achromatic color and the color image quality is deteriorated, the observer may not recognize the deterioration. In this case, the influence of the color component on the image evaluation value is suppressed, and the evaluation of the image quality by giving priority to the image evaluation value of the lightness component may correspond to the judgment based on the sense of the observer. The color characteristic correction unit 16 corrects the difference between the evaluation value and the observer's feeling due to such a color difference.

As an example of correction, here, the image evaluation values of the a ^* component and b ^* component are corrected using a function that monotonically increases with respect to the average value of a ^* and the average absolute value of b ^* . Using the correction function corresponding to each of the a ^* component and the b ^* component, the image evaluation value obtained by correcting the lightness for each component is corrected by the visual characteristic with respect to the human color. Even so, the image evaluation value matches the observer's feeling. Note that the correction function shown in FIG. 6 is an example, and needless to say, the correction function is not limited to this example.

Returning to FIG. 2, in S <b ^> 7, the comprehensive evaluation value calculation unit 17 calculates the image evaluation value of the brightness component corrected by the brightness characteristic correction unit 15 and the color component (a ^* component, b) corrected by the color characteristic correction unit 16. ^A comprehensive evaluation value is calculated from the image evaluation value of ⁽ component). For example, if the image evaluation value of the lightness component corrected by the lightness characteristic correction unit 15 is LNV1, and the image evaluation values of the a ^* component and b ^* component corrected by the color characteristic correction unit 16 are CNVa and CNVb, the total evaluation value TNV is TNV = p · LNVl + q · CNVa + r · CNVb + s
It can be obtained by The comprehensive evaluation value TNV may be an evaluation value indicating whether the image quality is good or bad. Note that p, q, r, and s are coefficients and may be determined in advance. Of course, the present invention is not limited to this example, and the comprehensive evaluation value TNV may be calculated by various methods.

  The comprehensive evaluation value is used for evaluating the image quality in the frequency range integrated in S4, and indicates the deterioration characteristic of the image quality indicated by the spatial frequency component in the integrated frequency band. For example, when the frequency band is in the range of 0.1 cycle / mm or more and 1.0 cycle / mm or less as a specific example of the frequency band, it becomes an evaluation value for unevenness called mottle. Further, in a frequency band higher than this frequency band, an evaluation value related to graininess such as a feeling of roughness is obtained. In any case, the image evaluation value is corrected by the lightness characteristic correction unit 15 and the color characteristic correction unit 16, and the obtained comprehensive evaluation value does not depend on the lightness, but also on the human sense regardless of the color. The evaluation value is a suitable image quality.

  FIG. 7 is an explanatory diagram of an example of the correlation between the image evaluation value and the comprehensive evaluation value and the psychological evaluation. Here, when the image evaluation value calculated by the evaluation value calculation unit 12 is used as it is, when the image evaluation value corrected for the lightness by the lightness characteristic correction unit 15 is used, and further, the color characteristic correction unit 16 performs color correction on the color. Relationship between the case of using the corrected image evaluation value and the total evaluation value obtained from the image evaluation value in each case with the psychological evaluation value quantified by the quality that the observer feels Is shown. Each point indicates a result for a different image to be evaluated.

7A, 7B, and 7C respectively show the relationship between the image evaluation value calculated by the evaluation value calculation unit 12 and the value of psychological evaluation, and FIG. 7A shows the L ^* component. FIG. 7B shows the relationship for the a ^* component, and FIG. 7C shows the relationship for the b ^* component. FIG. 7D shows the relationship between the overall evaluation value calculated by the comprehensive evaluation value calculation unit 17 without correction from the image evaluation value calculated by the evaluation value calculation unit 12 and the value of psychological evaluation. . It can be seen that there is variation in the relationship between the calculated comprehensive evaluation value and the psychological evaluation value.

The relationship between the image evaluation value calculated by the evaluation value calculation unit 12 and the value of the psychological evaluation after the lightness characteristic correction unit 15 corrects the lightness is shown in FIGS. G). FIG. 7E shows the relationship regarding the L ^* component, FIG. 7F shows the relationship regarding the a ^* component, and FIG. 7G shows the relationship regarding the b ^* component. FIG. 7H shows the relationship between the overall evaluation value calculated by the comprehensive evaluation value calculation unit 17 and the value of psychological evaluation from the image evaluation value corrected by the lightness characteristic correction unit 15 without performing color correction. ing. The relationship between the calculated comprehensive evaluation value and the value of the psychological evaluation is less varied as compared to the case where the correction for the brightness shown in FIG. 7D is not performed, but like the points surrounded by ○ , You may get results that are distant from other points. Therefore, there are cases where the correction for the lightness alone does not match the psychological evaluation.

FIG. 7 (I), (J), (K) shows the relationship between the image evaluation value and the psychological evaluation value obtained by correcting the lightness by the lightness characteristic correcting unit 15 and further correcting the color by the color characteristic correcting unit 16. It shows. FIG. 7 (I) shows the relationship for the L ^* component, FIG. 7 (J) shows the relationship for the a ^* component, and FIG. 7 (K) shows the relationship for the b ^* component. The overall evaluation value calculated by the comprehensive evaluation value calculation unit 17 from the image evaluation value of the L ^* component corrected by the lightness characteristic correction unit 15 and the image evaluation value of the a ^* and b ^* components corrected by the color characteristic correction unit 16 The relationship with the value of psychological evaluation is shown in FIG. The calculated comprehensive evaluation value matches the value of the psychological evaluation, and the points indicated by circles in FIG. Therefore, it can be considered that the obtained comprehensive evaluation value indicates psychological evaluation, and the evaluation value of the image quality suitable for human sense is obtained.

  FIG. 8 is an explanatory diagram of an example of a computer program, a storage medium storing the computer program, and a computer when the functions described in the embodiment of the present invention are realized by a computer program. In the figure, 21 is a program, 22 is a computer, 31 is a magneto-optical disk, 32 is an optical disk, 33 is a magnetic disk, 34 is a memory, 41 is a CPU, 42 is an internal memory, 43 is a reading unit, 44 is a hard disk, 45 is An interface 46 is a communication unit.

  You may implement | achieve the whole or a part of function of each part demonstrated as one Embodiment of the above-mentioned this invention with the program 21 which a computer runs. In this case, the program 21 and data used by the program may be stored in a storage medium that can be read by a computer. The storage medium refers to the reading unit 43 provided in the hardware resource of the computer, causing a change state of energy such as magnetism, light, electricity, etc. according to the description content of the program, and corresponding signals. In this format, the description content of the program is transmitted to the reading unit 43. For example, there are a magneto-optical disk 31, an optical disk 32 (including a CD and a DVD), a magnetic disk 33, a memory 34 (including an IC card, a memory card, a flash memory, and the like). Of course, these storage media are not limited to portable types.

  The program 21 is stored in these storage media. For example, the storage medium 43 is mounted on the reading unit 43 or the interface 45 of the computer 22 to read the program 21 from the computer, and the internal memory 42 or the hard disk 44 (magnetic disk or the like). The functions described in the embodiment of the present invention described above are realized in whole or in part. Alternatively, the program 21 is transferred to the computer 22 via the communication path, and the computer 22 receives the program 21 by the communication unit 46 and stores it in the internal memory 42 or the hard disk 44, and the program 21 is executed by the CPU 41. Also good.

  In addition, the computer 22 may be connected to various devices via an interface 45. For example, an image reading device may be connected, and an image to be evaluated may be read by the image reading device to obtain an evaluation value of the image to be evaluated. The image to be evaluated is stored in a storage medium and read from the storage medium by the interface 45, stored in the internal memory 42 or the hard disk 44, or sent through a communication path. Good.

  The interface 45 may be connected to an output device that outputs information, and the evaluation result may be output. Of course, other devices may be connected to the interface 45. Note that each component does not have to operate on one computer, and processing may be executed by another computer depending on the stage of processing.

  DESCRIPTION OF SYMBOLS 11 ... Spatial frequency distribution generation part, 12 ... Evaluation value calculation part, 13 ... Luminous spatial frequency characteristic correction part, 14 ... Integration part, 15 ... Lightness characteristic correction part, 16 ... Color characteristic correction part, 17 ... Comprehensive evaluation value calculation Part, 21 ... program, 22 ... computer, 31 ... magneto-optical disk, 32 ... optical disk, 33 ... magnetic disk, 34 ... memory, 41 ... CPU, 42 ... internal memory, 43 ... reading part, 44 ... hard disk, 45 ... interface 46 Communication unit.

Claims (3)

Means for generating a spatial frequency distribution for the L ^* component and the a ^* and b ^* components of the image to be evaluated, and an integrated value of a frequency band that is predetermined based on the spatial frequency distribution and sensitive to human vision And an image of the a ^* component using a function that corrects the image evaluation value of the L ^* component according to the average brightness of the image to be evaluated and monotonically increases with respect to the average brightness. Lightness correction means for correcting the image evaluation value of the b ^* component using a function that corrects the evaluation value and monotonously decreases with respect to the average lightness, and a ^* and b ^* according to the average color value of the image to be evaluated An image evaluation apparatus comprising color correction means for correcting the image evaluation value of a component. Said color correction means, a ^*, b ^* a ^* using a function that increases monotonically with respect to the absolute value of the average of the components, according to claim 1, characterized in that to correct the image evaluation value of the b ^* component Image evaluation device.   An image evaluation program for causing a computer to execute the function of the image evaluation apparatus according to claim 1 or 2.

Images