JP2001074602A - Apparatus and method for evaluating image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an apparatus and a method for evaluating an image capable of objectively evaluating a resolution representing a subjective sharpness of an image. SOLUTION: The apparatus 10 for evaluating an image comprises a data holding means 11 for holding image data to output a pattern image having different dark and light density at random in one-dimensional direction, a data acquiring means 12 for outputting the image data to an image output unit 20 to and reading the pattern image after the outputting, an MTF analyzing means 14 for calculating MTF characteristics from the image data output to the unit 20 and the image data read by the means 12, a pixel size estimating means 15 for estimating the pixel size of the output image from the calculated MTF characteristics, and an index calculating means 16 for calculating an index representing resolution capability of the unit 20 from the estimated pixel size.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image evaluation apparatus and an image evaluation apparatus for evaluating the image quality of an output result from an image output apparatus such as a printer or a copying machine.

[0002]

2. Description of the Related Art Conventionally, as a method for evaluating image quality, visual comparison evaluation using a limit sample has been widely used.
This method is called so-called subjective evaluation, and evaluates the quality of an image, the degree of damage, and the like by subjective judgment through human vision. However, such a subjective evaluation has the drawback that the evaluation results vary from one evaluator to another, and that enormous effort is required to create a limit sample.

In order to compensate for these drawbacks, in recent years, an attempt has been made to shift to a so-called objective evaluation that does not depend on human subjective judgment by using an image evaluation device or the like that analyzes and evaluates an output image. Has been carried out and the results have been increasing. As an objective evaluation method, for example, an MTF (Modulation Transfe
r Function) to analyze the characteristics to evaluate the resolution of the output image, and to calculate the CT from the output (print) result of the rectangular wave pattern.
It is common practice to analyze F (Contrast Transfer Function) characteristics and evaluate the resolution of the output image.

As other objective evaluation methods, for example, JP-A-7-193708 and JP-A-7-19370
Japanese Patent Application Laid-Open No. 9-199095 and Japanese Patent Application Laid-Open No. H7-190952 each disclose techniques for analyzing printed rectangular wave patterns to obtain projection data, and evaluating the resolution of a printed image based on the projection data. According to this technique, it is possible to perform an objective evaluation that is very accurate for an analog copying machine or the like and that matches the subjective sharpness of a so-called “blur condition” or “sharpness” of an image.

[0005] Further, as another objective evaluation method, for example, Japanese Patent Application Laid-Open Nos. Hei 7-325922 and Hei 9-81
No. 742 discloses a technique for analyzing a printed line image to obtain an index representing the resolution of the printed image. According to this technique, even when an image obtained by a digital copying machine is evaluated, an objective evaluation matching a subjective evaluation value can be performed regardless of the image structure.

[0006]

However, in the above-mentioned prior art, there is a possibility that the following problems may occur. For example, when a printer outputs a periodic pattern such as a sine wave pattern or a rectangular wave pattern and evaluates the resolution of the printer, the periodic pattern interferes with the cycle of the screen of the printer. In this case, the periodic pattern is not correctly output, and as a result, accurate evaluation cannot be performed. Here, the screen of the printer refers to characteristics unique to the image output of the printer, such as the output pitch and arrangement of dots (pixels). In other words, if the pattern has a period that is an integral multiple of the screen period, the periodic pattern is correctly output, but a periodic pattern with any other frequency may not be correctly output. Therefore, printers having different screen periods cannot be evaluated using the same periodic pattern. From this, for example, 200
It becomes very difficult to compare the resolution capabilities of a printer of dpi (dot par inch) and a printer of 175 dpi.

Further, Japanese Patent Application Laid-Open No. 7-193708,
JP-A-7-193709 and JP-A-7-190
The technique disclosed in Japanese Patent Application Laid-Open No. 952 also performs objective evaluation using a periodic rectangular wave pattern, and thus it is clear that the technique includes exactly the same problem as the above-described case.

In the case of calculating an index for evaluating a resolution by analyzing a line-shaped image as in the technique disclosed in JP-A-7-325922 and JP-A-9-81742, the above-described technique is used. Although it is not affected by the screen cycle, a fatal problem arises in that it is not possible to compare printers having different numbers of gradations that can be drawn on one pixel. For example, considering a density gradation type printer capable of drawing 256 gradations and an area gradation type printer capable of drawing only two gradations, if an ideal line-shaped image can be drawn in each of them, the resolution capability of both of them can be considered. Will be evaluated as the same. However, when a subjective evaluation is performed on an actually drawn image, it can be easily analogized that a density gradation type printer having a large number of gradations is evaluated to have higher resolution.

Therefore, the present invention provides a solution for an image output device without depending on the inherent characteristics of the image output device such as a screen cycle of a printer, or the image drawing method such as a density gradation method or an area gradation method. It is an object of the present invention to provide an image evaluation device and an image evaluation method capable of objectively evaluating the image ability while reflecting the subjective sharpness of an image.

[0010]

SUMMARY OF THE INVENTION The present invention is an image evaluation device devised to achieve the above object, in which image data is output to an image output device and the output result is evaluated. . A data holding unit for holding image data for outputting a pattern image having a different shading at random in a one-dimensional direction; and a data transmission for outputting the image data held in the data holding unit to the image output device. Means, data acquisition means for reading a pattern image obtained by output from the image output device and acquiring image data from the pattern image, and image data output by the output instruction means and read by the data acquisition means. Image position specifying means for obtaining a correspondence relationship with image data; MTF analysis means for calculating an MTF characteristic from the output image data and the read image data based on the correspondence obtained by the image position specification means; The pixel size of an image output to the image output device from the MTF characteristics calculated by the MTF analysis means. A pixel size estimation means for estimating a's, characterized in that the pixel size the pixel size estimation means has estimated and a index calculating means for calculating an index indicating the resolution capability of the image output device.

Further, in the image evaluation method devised to achieve the above object, the present invention is a method of outputting image data to an image output device and evaluating the output result. Then, the image output device outputs image data for outputting a pattern image having different shades at random in the one-dimensional direction, reads the pattern image obtained by the output from the image output device, and reads image data from the pattern image. Is obtained, the correspondence between the read image data and the image data output to the image output device is obtained, and based on the correspondence, the read image data, the output image data, and the read image are obtained. MT from data
Calculating an F characteristic, estimating a pixel size of an image output to the image output device from the calculated MTF characteristic, and calculating an index representing a resolution capability of the image output device from the estimated pixel size. I do.

According to the image evaluation device and the image evaluation method having the above-described configuration, the image output device outputs a pattern image having a different shading at random in the one-dimensional direction, and reads and reads the output pattern image. By comparing the image data with the image data on which it is based (the image data output to the image output device), the pixel size of the image output to the image output device can be estimated. It is possible to calculate a resolution index corresponding to the above. In addition, in calculating the resolution index, a pattern image having a different shading at random in the one-dimensional direction, that is, a pattern image that does not include a specific periodic component in the one-dimensional direction is used. The resolution index can be calculated without being affected by the image and the image drawing method.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an image evaluation apparatus and an image evaluation method according to the present invention will be described with reference to the drawings. FIG. 1 is a block diagram illustrating a schematic configuration of an example of an embodiment of an image evaluation device according to the present invention.

As shown in FIG. 1, an image evaluation apparatus 10 of the present embodiment analyzes an output image 30 output from a printer 20 and determines the resolution of the output image 30 based on the analysis result, that is, the image that the printer 20 has. This is to evaluate the resolution ability. Therefore, the image evaluation device 1
0 is a data holding unit 11, a data acquisition unit 12,
The image processing apparatus includes an image position specifying unit 13, an MTF analyzing unit 14, a pixel size estimating unit 15, and a resolution index calculating unit 16.

The data holding means 11 is, for example, a RAM (Ra
Semiconductor memory such as ndom Access Memory) or HDD
(Hard Disk Drive) or the like, which previously stores image data to be output to the printer 20 at the time of the resolution evaluation. Then, in order to output the image data to the printer 20, the data holding unit 1
1 is connected to the printer 20 via a communication line or the like (not shown), and transmits image data to the printer 20 via the communication line or the like.

However, the data holding means 11 holds image data for outputting a one-dimensional random pattern as image data to be output to the printer 20.
A one-dimensional random pattern is a pattern in which the shading (gradation value) changes randomly without regularity in the one-dimensional direction, and the shading does not change in the direction orthogonal to the one and has the same gradation value. Refers to a pattern image. In addition, it is assumed that the one-dimensional random pattern does not have a frequency at which the amplitude becomes “0” in a frequency range of about 10 cycles / mm or less, and has no peak at a specific frequency. Image data for outputting such a one-dimensional random pattern (hereinafter, referred to as “output image data”) may be generated by using, for example, a random number generation by an arithmetic function of a computer. In addition,
The data holding unit 11 holds the output image data in a digital data state.

When the printer 20 outputs the output image data, the data acquisition means 12 optically reads a one-dimensional random pattern obtained by the output, and converts the one-dimensional random pattern into image data (hereinafter referred to as "input image data"). ). This reading is performed by the data acquisition unit 12 using, for example, a scanning densitometer.

The scanning densitometer is, for example, 10 μm × 500
μm slit-shaped reading surface (hereinafter, referred to as “aperture”), and the direction of the aperture is configured to be substantially perpendicular to the scanning direction.
It can measure the concentration profile for each μm. By using such a scanning densitometer, the data acquisition unit 12 can measure the density of the entire one-dimensional random pattern every 10 μm along the direction in which the gradation of the one-dimensional random pattern changes randomly. A value is obtained, and this is set as an input image data value.

The data acquisition means 12 is not a scanning densitometer, but a high-resolution scanner or CCD (Charge C).
The input image data may be obtained using an oupled device) camera or the like.

The image position specifying means 13 obtains the correspondence between the output image data output by the printer 20 and the input image data obtained by the data obtaining means 12, as will be described in detail later. That is, the image position specifying unit 13 attempts to match the profile of the output image data with the profile of the input image data. MT
The F-analyzing unit 14 calculates the MTF characteristic from the output image data and the input image data based on the correspondence obtained by the image position specifying unit 13 as described later in detail. The pixel size estimating unit 15 estimates the pixel size on the output image 30 output by the printer 20, based on the MTF characteristics calculated by the MTF analyzing unit 14, as will be described later in detail. The resolution index calculating unit 16 calculates an index indicating the resolution capability of the printer 20 based on the pixel size estimated by the pixel size estimating unit 15 as described later in detail.

Each of these means, that is, the image position specifying means 13, the MTF analyzing means 14, the pixel size estimating means 15, and the resolution index calculating means 16 is provided, for example, by a calculation provided with a scanning densitometer. It can be realized by using functions. However, instead of the calculation function of the scanning densitometer, it may be realized by a computer or the like that executes a predetermined program.

Next, an example of a processing operation when the image evaluation apparatus 10 configured as described above evaluates the resolution, that is, an image evaluation method in the present embodiment, will be described with reference to FIGS.
This will be described with reference to FIG. FIG. 2 is an explanatory diagram showing an example of a one-dimensional random pattern, FIG. 3 is an explanatory diagram showing one measurement example of a density profile, FIG. 4 is an explanatory diagram showing an example of an MTF analysis result, and FIG. FIG. 6 is an explanatory diagram showing an analysis example of the line spread function, and FIG. 6 is an explanatory diagram showing a specific example of pixel size estimation.

In the image evaluation apparatus 10, a printer 20
To evaluate the resolution of the printer, first, the output image data held in the data holding unit 11 is transmitted to the printer 20 through a communication line or the like, and
Output the output image data. Thus, the printer 20 outputs a one-dimensional random pattern as shown in FIG. 2, for example, on the recording paper.

Since the output one-dimensional random pattern is a visualized image of the output image data, the density changes randomly in the one-dimensional direction (A direction in the figure), as is clear from the example of FIG. However, in the direction orthogonal to this (direction B in the figure), the shading is constant, and there is no frequency at which the amplitude becomes “0” in a frequency range of approximately 10 cycles / mm or less, and a peak exists at a specific frequency. do not do.

After the printer 20 outputs the one-dimensional random pattern, in the image evaluation apparatus 10, the data obtaining means 12 reads the one-dimensional random pattern and obtains input image data from the one-dimensional random pattern. As a result, in the data acquisition unit 12, for example, “Output (solid line portion in FIG. 3)” shown in FIG. 3, the entire region along the direction in which the gradation of the one-dimensional random pattern changes is 10 μm A density profile for each is obtained.

The image evaluation apparatus 10 performs the following processing operations in accordance with the transmission of the output image data to the printer 20 and the acquisition of the input image data from the one-dimensional random pattern output by the printer 20. It is supposed to do.

First, the image evaluation apparatus 10 includes a printer 20
When the output image data is transmitted to the printer 20, the patch data corresponding to the gradation value of the output image data is transmitted to the printer 20. The patch data is, for example, “0” to “25” if the output image data is digital data of 256 gradations.
6 "is image data for outputting a patch image corresponding to each data value.

When the printer 20 outputs a patch image corresponding to each gradation value by transmitting the patch data,
Subsequently, in the image evaluation device 10, the data acquisition unit 12 measures the average density of each patch image in accordance with the acquisition of the input image data from the one-dimensional random pattern. Then, based on the measurement result, a conversion table is created that indicates the correspondence between the tone values of the output image data and the density data obtained by the data acquisition unit 12.

With this conversion table, the correspondence relationship between output image data, which is digital data, and density data, which is analog data, is clarified. That is, the image evaluation device 10 creates a conversion table based on the above-described patch image so that the output image data can be handled as density data.

As a result, when the output image data is converted into a density value based on the created conversion table, for example, FIG.
It has a density profile indicated by “Input (broken line portion in the figure)” in FIG. 2, and further, after the printer 20 outputs the output image data as a one-dimensional random pattern, the data acquisition unit 12 reads the output image data. When you do
It can be seen that the density profile shown in “Output (solid line portion in FIG. 3)” in FIG. 3 is obtained as the input image data.

Thereafter, in the image evaluation apparatus 10, the image position specifying means 13 obtains the correspondence between the density profile of the output image data and the density profile of the input image data. At this time, the image position specifying means 13 divides the output image data and the input image data into a plurality of regions, respectively, and performs the processing. That is, the image position specifying means 13
First, with respect to output image data converted to a set of approximately 10,000 density data, for example, one point from a reference point (for example, a reading start point corresponding to an end of a one-dimensional random pattern).
024 pieces of data are extracted. Then, this is shifted one data at a time with respect to the input image data composed of a set of approximately 10,000 density data in the same manner as the output image data,
The correlation coefficient of the density value between the two is calculated, and the position where the correlation coefficient becomes maximum is stored as a corresponding position in, for example, a predetermined area in the data holding unit 11. Next, the image position specifying means 13 re-starts from the point shifted from the reference point of the output image data by, for example, 512 data points.
024 pieces of data are extracted, and the corresponding relationship is obtained in the same manner as in the above-described case.

By repeating this for the entire data range, the image position specifying means 13 converts the output image data and the input image data into, for example, 1
While the image data is divided into eight regions, the correspondence between them is obtained for each region. It goes without saying that the size of each area and the number of divisions thereof can be set arbitrarily.

After specifying the correspondence by the image position specifying means 13, the MTF analyzing means 14 calculates the MTF characteristics for each area based on the correspondence. MTF
The calculation of the characteristics may be performed by using a known technique. As an example, the calculation may be performed as follows. That is, an orthogonal transformation such as a one-dimensional Fourier transform is performed on the output image data and the input image data in the mutually corresponding areas, and after the transformation, M
Find TF (λ).

Further, the MTF analysis means 14 calculates the MTF
When (λ) is obtained, the MTF (λ) is converted into a complex format, and then the inverse Fourier transform is performed. This corresponds to a line spread function (LSF) for the output image data. These processes are referred to as MT
The F analyzing means 14 performs the processing on all of the 18 areas specified by the image position specifying means 13 and averages the processing results for each area. As a result, the MTF analysis unit 14 can obtain an MTF analysis result as shown in FIG. 4 and an LSF analysis result as shown in FIG. 5, for example.

Thereafter, the pixel size estimating means 15 outputs
The pixel size on the output image 30 output from the printer 20 is estimated based on the analysis result by the F analysis unit 14. The pixel size here corresponds to the minimum required pixel size for expressing a certain gradation (for example, 256 gradations).

More specifically, the pixel size estimating means 15 compares the LSF analysis result obtained by the MTF analyzing means 14 with Gaussian (Gaussi
an) Approximate by a function, and the width of a portion corresponding to approximately 10% of the maximum value in the Gaussian function is defined as the pixel size. Then, when the pixel size estimating unit 15 estimates the pixel size using the Gaussian function, the resolution index calculating unit 16 determines the pixel size by the pixel size estimating unit 15 when the printer 20 outputs the output image 30. This is an index indicating the resolution capability.

Therefore, the resolution index calculating means 16
According to the index according to the resolution capability of the printer 20,
That is, the quality of the image output result by the printer 20 can be objectively evaluated. Moreover, since the index reflects the estimation result of the pixel size, the index matches the subjective sharpness of the image, such as "blur condition" and "sharpness".

Next, another embodiment according to the present invention will be described. Here, a case will be described in which the estimation of the pixel size and the calculation of the index indicating the resolution capability are different from those in the above-described embodiment. FIG. 7 is an explanatory diagram showing another specific example of pixel size estimation.

Also in this embodiment, the process until the MTF analysis means 14 finds MTF (λ) is exactly the same as in the above-described embodiment. MTF analysis means 14
When (λ) is obtained, in the present embodiment, the pixel size estimating unit 15 estimates the pixel size as follows.

In the pixel size estimating means 15, for example, FIG.
As shown in FIG.
(Λ) is approximated by a synchronization (Sync) function.
The frequency λ at which TF = 0.5 is calculated. Then, the resolution index calculating unit 16 calculates the index R representing the resolution using the following equation (1), assuming that the frequency λ obtained by the MTF analyzing unit 14 is the Nyquist frequency.

[0041]

(Equation 1)

Also with the index R calculated in this manner, an objective evaluation reflecting the subjective sharpness can be made, as in the case of the above-described embodiment.

That is, in any of the above-described embodiments, in evaluating the resolution, first, a one-dimensional random pattern is output to the printer 20 to be evaluated, and input image data is read from the output one-dimensional random pattern. Then, the pixel size of the output image is estimated by comparing the output image data with the output image data based thereon, and an index indicating the resolution capability of the printer 20 is obtained from the pixel size. A resolution index corresponding to the subjective sharpness can be calculated.

In order to confirm this, for some printers having different drawing methods, the resolution was evaluated using a one-dimensional random pattern, and a portrait photographic image sample for subjective evaluation was collected. Subjective evaluation was performed by the paired comparison method using the image sample. In the subjective evaluation, 10 technicians handling images were used as subjects, and all the image samples were evaluated for the sharpness and sharpness of the images, and the scores were obtained.

FIG. 8 is an explanatory diagram showing a comparative example of the evaluation result when the resolution index is obtained from the width of less than 10% of the LSF maximum value and the evaluation result by the subjective evaluation. As is clear from the example of the figure, when the respective evaluation results are compared, the correlation coefficient between the two is about 0.9865, and it can be seen that the psychological sharpness is greatly reflected in the result of the objective evaluation.

FIG. 9 is an explanatory diagram showing a comparison example between the evaluation result when the resolution index is calculated using the frequency at which MTF = 0.5 as the Nyquist frequency and the evaluation result by the subjective evaluation. Also in this case, the correlation coefficient between the two is about 0.9452, and it is understood that the psychological sharpness is highly reflected in the result of the objective evaluation.

Furthermore, according to the image evaluation device 10 and the image evaluation method performed by the image evaluation device 10 in each of the above-described embodiments (hereinafter, these are simply referred to as “image evaluation devices and the like”), the subjective sharpness is reduced. Not only can we calculate a compatible resolution index, but in calculating the resolution index, we use a one-dimensional random pattern, that is, a pattern image that does not include a specific periodic component in the one-dimensional direction. The resolution index can be calculated without being affected by the cycle of the screen 20 or the image drawing method of the printer 20. This makes it possible to easily compare the resolution capabilities of printers having different screen periods, such as a 200 dpi printer and a 175 dpi printer. Furthermore, if the printers having different numbers of gradations that can be drawn in one pixel, such as the density gradation method and the area gradation method, are used, the resolution of the density gradation method printer having a large number of gradations is higher. The objective evaluation result is obtained.

That is, according to the image evaluation apparatus of each embodiment, the index representing the resolution ability of the printer 20 is obtained by using the one-dimensional random pattern as in the first or eighth aspect of the present invention. For example, without depending on the intrinsic characteristics of the printer 20 or the image drawing method, the printer 2
The resolution ability at 0 can be objectively evaluated while reflecting the subjective sharpness of the image.

Further, in the image evaluation apparatus of each embodiment,
According to the second or ninth aspect of the present invention, in the process of performing the objective evaluation, the MTF characteristic is calculated based on the ratio between the output image data and the input image data. Thus, the calculation result of the MTF characteristic is more accurate than that of the case described above, and as a result, the accuracy of the evaluation result can be improved.

Further, in the image evaluation device or the like, in the process of performing the objective evaluation, the MT evaluation is performed as in the third or tenth aspect of the present invention.
The pixel size is estimated using a Gaussian function obtained from the result of applying the inverse Fourier transform to the F characteristic, or the frequency at which the value of the MTF characteristic is approximately 0.5 as in the invention according to claim 4 or 11 is calculated. The pixel size is estimated based on this. That is, by using one of the two most representative methods among the methods for reflecting the psychological sharpness, the subjective sharpness can be surely reflected on the result of the objective evaluation.

Further, in the image evaluation device of each embodiment,
In the process of performing the objective evaluation, the correspondence between the output image data and the input image data is determined while changing the relative position between the output image data and the input image data. Therefore, it is possible to accurately determine the correspondence between the two, and for example, based on the ratio between the two, MT
Even when calculating the F characteristic, its accuracy increases,
As a result, an improvement in the accuracy of the evaluation result can be expected.

Further, in the image evaluation apparatus and the like of each embodiment,
In the process of performing the objective evaluation, the data to be analyzed is divided into a plurality of regions and analyzed, and the analysis results for each region are averaged, as in the invention according to claim 6 or 13. I have. Therefore, even if there is an uncertain element such as a noise component included in the data or expansion and contraction of the recording paper, the uncertain element can be eliminated by averaging, so that the processing result becomes accurate, and the result becomes accurate. As a result, the accuracy of the evaluation result can be improved.

Further, in the image evaluation device and the like of each embodiment,
As in the invention according to claim 7, the amplitude of the one-dimensional random pattern is "0" in a range of about 10 cycles / mm or less.
Is not included. About 10
The range of cycle / mm or less is a range that can be visually recognized by humans. Therefore, at least within the range that can be seen by human eyes, the frequency whose amplitude is “0” is not included. Thereby, for example, even when the ratio between the output image data and the input image data is required in the process of performing the objective evaluation, the ratio cannot be calculated in a range of approximately 10 cycles / mm or less, so that the ratio cannot be calculated. Matching with human subjective evaluation can be achieved reliably.

In the above embodiment, the printer 2
Although the case where the resolution capability of the image possessed by 0 is evaluated has been described as an example, the evaluation target is not limited to the printer 20, and if the image data is to be output as a visible image, for example, copy Needless to say, the present invention can be similarly applied to other image output devices such as a printer.

[0055]

As described above, according to the image evaluation apparatus and the image evaluation method of the present invention, a pattern image whose shading is randomly different in one-dimensional direction, that is, a pattern image which does not include a specific periodic component in one-dimensional direction. Since images are used, in addition to being able to calculate the resolution index corresponding to the subjective sharpness, the calculation of the resolution index depends on the influence of the inherent characteristics of the image output device, the image drawing method, and the like. It can be done without receiving it. From this, for example, 200 dpi
And 175 dpi printers, and image output devices having different image rendering methods such as density gradation type printers and area gradation type printers such as an area gradation type printer. It is possible to calculate a resolution index corresponding to the above. That is,
According to the image evaluation device and the image evaluation method, the resolving power of the image output device can be improved without depending on the intrinsic characteristics and the image drawing method of the image output device to be evaluated. Can be evaluated objectively while reflecting

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of an example of an embodiment of an image evaluation device according to the present invention.

FIG. 2 is an explanatory diagram illustrating an example of a one-dimensional random pattern.

FIG. 3 is an explanatory diagram showing one measurement example of a density profile.

FIG. 4 is an explanatory diagram showing an example of an MTF analysis result.

FIG. 5 is an explanatory diagram showing an example of analysis of a line spread function.

FIG. 6 is an explanatory diagram showing a specific example of pixel size estimation.

FIG. 7 is an explanatory diagram showing another specific example of pixel size estimation.

FIG. 8 is an explanatory diagram (part 1) illustrating a relationship between an estimated pixel size and psychological sharpness of an image.

FIG. 9 is an explanatory diagram (part 2) illustrating a relationship between an estimated pixel size and a psychological sharpness of an image.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 ... Image evaluation apparatus, 11 ... Data holding means, 12 ... Data acquisition means, 13 ... Image position specifying means, 14 ... MTF
Analysis means, 15: Pixel size estimation means, 16: Resolution index calculation means, 20: Printer, 30: Output image

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Tsutomu Oyama 430 Sakai Nakaicho, Ashigarashimo-gun, Kanagawa Prefecture Green Tech Nakai Inside Fuji Xerox Co., Ltd. F term (reference) 2C061 AP04 KK04 KK18 KK25 KK26 KK28 5B057 BA02 CA02 CA08 CA12 CA16 CC03 CD20 DA03 DB02 DB05 DB09

Claims (14)

[Claims] An image evaluation device for outputting image data to an image output device and evaluating an output result thereof, wherein the image evaluation device holds image data for outputting a pattern image having a different shading at random in a one-dimensional direction. A data holding unit, a data transmitting unit for outputting the image data held in the data holding unit to the image output device, and reading a pattern image obtained by output from the image output device to read the pattern image from the pattern image. Data acquisition means for acquiring image data, image position identification means for associating image data output by the output instruction means with image data read by the data acquisition means, and image position identification means An MT for calculating an MTF characteristic from the output image data and the read image data based on the correspondence relationship; Analysis means; pixel size estimation means for estimating the pixel size of an image output to the image output device from the MTF characteristics calculated by the MTF analysis means; and the image output device from the pixel size estimated by the pixel size estimation means And an index calculating means for calculating an index indicating the resolution capability of the image. 2. The method according to claim 1, wherein the MTF analysis unit performs an orthogonal transformation on the output image data and the read image data, and calculates an MTF characteristic from a ratio between the orthogonally transformed image data and the read image data. The image evaluation device according to claim 1, wherein: 3. The image processing apparatus according to claim 2, wherein the pixel size estimating means includes an MTF.
3. The image evaluation device according to claim 1, wherein the pixel size is estimated by using a function obtained from a result obtained by performing an inverse orthogonal transform on the MTF characteristic calculated by the analysis unit. 4. The method according to claim 1, wherein the pixel size estimating means includes an MTF.
3. The image evaluation device according to claim 1, wherein the pixel size is estimated based on a frequency at which the value of the MTF characteristic calculated by the analysis unit is approximately 0.5. 5. The image position specifying means calculates a correlation coefficient while changing a relative position between the output image data and the read image data, and calculates a correlation coefficient based on a position where the correlation coefficient becomes maximum. 5. The image evaluation apparatus according to claim 1, wherein the correspondence relation is determined. 6. The MTF analysis means, when calculating MTF characteristics, divides and analyzes image data to be analyzed into a plurality of pieces, calculates an average of analysis results for each of the divided image data, and calculates 6. The calculation result of the MTF characteristic is obtained.
An image evaluation device as described in the above. 7. The image data for outputting the pattern image does not include a frequency having an amplitude of 0 within a range of about 10 cycles / mm or less.
7. The image evaluation device according to 4, 5, or 6. 8. An image evaluation method for outputting image data to an image output device and evaluating the output result, wherein the image data for outputting a pattern image having a different shading at random in a one-dimensional direction is obtained. An image output device, reads a pattern image obtained by the output from the image output device, acquires image data from the pattern image, and outputs the read image data and the image data output to the image output device. A correspondence is obtained, and based on the correspondence, an MTF characteristic is calculated from the read image data, the output image data, and the read image data, and the MTF characteristic is output to the image output device from the calculated MTF characteristic. It is characterized in that the pixel size of an image is estimated, and an index representing the resolution capability of the image output device is calculated from the estimated pixel size. The image evaluation method to be used. 9. When calculating the MTF characteristic,
9. An MTF characteristic is calculated based on a ratio between the output image data and the read image data, and a ratio between the orthogonally converted image data and the read image data.
The described image evaluation method. 10. After calculating the MTF characteristic, the MTF
9. The pixel size is estimated using a function obtained from a result obtained by performing an inverse orthogonal transform on a characteristic.
Or the image evaluation method according to 9. 11. After calculating the MTF characteristic, the MTF
10. The image evaluation method according to claim 8, wherein the pixel size is estimated based on a frequency at which the characteristic value is approximately 0.5. 12. When obtaining the correspondence, a correlation coefficient is calculated while changing a relative position between the output image data and the read image data, and the correlation coefficient is calculated based on a position where the correlation coefficient becomes maximum. 12. The image evaluation method according to claim 8, wherein a correspondence between the two is obtained. 13. When calculating the MTF characteristic, image data to be analyzed is divided into a plurality of parts and analyzed, and an average of analysis results for each divided image data is obtained.
13. The image evaluation method according to claim 8, wherein the result is a calculation result of the MTF characteristic. 14. The image data for outputting the pattern image does not include a frequency having an amplitude of 0 within a range of about 10 cycles / mm or less.
The image evaluation method according to 0, 11, 12 or 13.