JP2017034611A - Picture evaluation device, picture evaluation system, and picture evaluation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a picture evaluation device capable of evaluating a picture by using an evaluation index including a geometrical and optical factors having high correlation with a visual psychological characteristic of a human.SOLUTION: A picture evaluation device for evaluating picture quality of a projection picture projected by a projection device includes: an imaging unit for imaging the projection picture; an extraction unit for extracting a plurality of parameters set with respect to a geometrical factor on picture quality and an optical factor on picture quality from an imaged picture imaged by the imaging unit; and a conversion unit that converts the plurality of parameters extracted by the extraction unit to an evaluation value for the picture quality of the projection picture, by using conversion information on relation between the evaluation value for the picture quality and the geometrical and optical factors based on a visual psychological characteristic of a human.SELECTED DRAWING: Figure 11

Description

  The present invention relates to an image evaluation apparatus, an image evaluation system, and an image evaluation method.

  In recent years, as a projector used as a projection apparatus, a short focus projector that projects an image onto a projection surface from a close range has been developed and is widely used. This short-focus projector has the advantage that it can save space because it projects an image from a close range, but it projects from an angled position rather than the front of the projection surface. When the surface is uneven, the projected image is easily distorted.

  Image distortion can be reduced or eliminated by performing corrections, but when determining whether corrections are to be performed or whether the corrected images are appropriate, the distortion strength and degree Indicators are needed. For this reason, it is important to index the strength and degree of image distortion.

  Conventionally, for example, a distortion amount calculation image is projected onto a projection surface, and based on a captured image obtained by capturing the image, the distortion amount at each feature point of the distortion amount calculation image is determined as a plurality of levels of distortion amounts. A technique is known (for example, Patent Document 1). In this technique, if there is a distortion at a feature point, a shift occurs between the image pickup apparatus system coordinate and the projector system coordinate. Therefore, the magnitude of the distortion amount in a plurality of stages is determined based on the size of the shift.

  As another method for quantitatively evaluating image quality, a method relating to “color breakup”, which is an image defect peculiar to the DLP (Digital Light Processing) method, is disclosed (for example, Patent Document 2, Non-Patent Document). 1).

  By the way, the technique described in Patent Document 1 only determines the amount of distortion based on the magnitude of deviation at each feature point, and whether or not the projected image is distorted by a human sense. It cannot be evaluated. For example, according to human visual psychological characteristics, the actual individual distortion is very small, but there are cases where the distortion is strongly felt because there are many distorted portions.

  In addition to distortion, optical factors such as brightness, contrast, color temperature, and color reproduction range are important from the image quality of the image actually projected by the projection apparatus. Therefore, when evaluating the projected image quality, it is possible to perform evaluation corresponding to human vision more by combining not only geometric factors such as distortion but also optical factors.

  Therefore, there is a need for an image evaluation apparatus that can evaluate an image using an evaluation index that includes a geometric factor and an optical factor that are highly correlated with human visual psychological characteristics.

  According to one aspect of the present embodiment, there is provided an image evaluation device for evaluating the image quality of a projection image projected by a projection device, the imaging unit capturing an image of the projection image, and the image captured by the imaging unit An extraction unit that extracts a plurality of parameters set with respect to a geometric factor of image quality and an optical factor of image quality from the captured image, and the geometric factor based on human visual psychological characteristics and evaluation values of the optical factor and the image quality A conversion unit that converts the plurality of parameters extracted by the extraction unit into an evaluation value of the image quality of the projection image using conversion information on the relationship between

  According to the disclosed image evaluation apparatus, an image can be evaluated using an evaluation index including a geometric factor and an optical factor that are highly correlated with human visual psychological characteristics.

Configuration diagram of a DLP projector Explanatory diagram of the reason why distortion of the projected image occurs in the short focus projector Illustration of distortion due to angle error Illustration of angle error Illustration of sensory evaluation The figure which shows an example of the image used for sensory evaluation Correlation diagram between brightness and MOS value obtained by sensory evaluation Configuration diagram of a projection system including an image evaluation apparatus according to the first embodiment 1 is a configuration diagram of an image evaluation apparatus according to a first embodiment. Flowchart (1) of the image evaluation method in the first embodiment Flowchart (2) of the image evaluation method in the first embodiment Figure (1) illustrating various evaluation patterns Illustration of color reproduction range Flowchart (3) of the image evaluation method in the first embodiment Figure (2) illustrating various evaluation patterns Flowchart of image evaluation method according to second embodiment

  The form for implementing this invention is demonstrated below. In addition, about the same member etc., the same code | symbol is attached | subjected and description is abbreviate | omitted.

[First Embodiment]
Projectors serving as projection apparatuses that are currently mainstream can be roughly classified into a liquid crystal system, a DLP system, and an LCOS (Liquid Crystal On Silicon) system. The liquid crystal method is a method in which light from a light source is transmitted through a liquid crystal panel, magnified by a lens, and projected onto a screen as a projection surface. Unlike the liquid crystal method, the LCOS method is a method in which light is reflected on a liquid crystal panel and projected onto a screen. The DLP method will be briefly described with reference to FIG. A DLP projector splits white light from a light source 10 into light for each color by passing through a color wheel 11 that rotates at high speed, and the light is DMD (Digital Micromirror) in which minute mirrors are arranged for the number of pixels. Device) 12 is irradiated. The DMD 12 performs ON / OFF control, selects reflection / non-reflection, enters the reflected light into the optical system 13, and enlarges the optical system 13 to project an image on the screen.

  As a projector, there is a short focus projector capable of projecting from a close range. In this projector, as shown in FIG. 2A, the incident angle θ of the projection light 20 for projecting the image onto the screen 21 is large. Further, the deviation δ of the projection position becomes large. This deviation δ appears as image distortion, for example, an image having distortion as shown in FIG. This projector has a great advantage of space saving by projecting from a close distance, but since the incident angle θ of the projection light 20 for projecting the image onto the screen 21 is large, the projection of the screen 21 serving as a projection surface is effective. Under the influence of subtle unevenness, the image projected on the screen 21 tends to be locally distorted. In addition, the individual difference between the projectors is large in the distortion method. Therefore, in a short focus projector, it is extremely important to index how much the degree of distortion in the projected image affects the image quality.

  Further, in such a projected image, comprehensive evaluation of image quality including not only distortion but also luminance, contrast, color temperature, color reproduction range and the like is also required. Specifically, at present, the luminance and the color reproduction range have been quantified as to which degree strongly controls the image quality, and the luminance and color temperature to what extent the image quality is strongly controlled. No. In general, it is considered that higher luminance is better or higher color temperature is considered preferable. By evaluating the image quality from multiple optical factors, it is possible to quantitatively determine whether it is better to improve the luminance or the color reproduction range, for example, an effective approach for improving the image quality. . Thereby, the image quality can be improved efficiently and effectively. Also, there is currently no evaluation index for comprehensively evaluating image quality including geometric factors such as distortion and optical factors such as luminance. Therefore, it is preferable to have such an evaluation index because, for example, to what extent distortion and luminance dominate image quality can be quantitatively understood. Furthermore, since the projection principle differs greatly between the liquid crystal projector and the DLP projector, the difference in image quality between the two can be quantitatively known.

  By the way, in the geometric factor, an index called angle error is used as one of simple indexes for evaluating distortion. The angle error indicates how much the lattice is distorted with respect to an ideal lattice (square or rectangle) without distortion when the lattice pattern is projected. It is uniquely shown to be large.

  FIG. 3 is a diagram for explaining the projection distortion due to the angle error. FIG. 3 shows an image in which a plurality of straight lines are intersected and a lattice pattern arranged in a lattice shape is projected. The lattice pattern indicated by a broken line is an ideal lattice pattern in which straight lines are orthogonal to each other without distortion, and the lattice pattern indicated by a solid line indicates how much the strain is distorted with respect to the ideal lattice pattern. Is. An angle α formed by the broken line and the solid line is an angle error.

  In addition to angle errors, distortion indices include the spatial period of distortion, the number of projection planes, and whether the distortion is distributed continuously or discretely. Since the influence on the image quality is strong, the following description will focus on the angle error.

  The angle error indicates that the larger the angle error, the greater the local distortion. As shown in FIG. 4, the angle error increases in the order of FIG. 4 (a), FIG. 4 (b), and FIG. 4 (c).

  Next, luminance, contrast ratio, color temperature, color reproduction range, and the like, which are indicators of optical factors of the image quality of the projected image will be described.

Luminance is the brightness when White is displayed, and cd / m ² is mainly used as a unit.

  The contrast ratio is the ratio of the luminance of white (white) and black (black), and is (white luminance) / (black luminance).

  The color temperature corresponds to the color of light emitted from a black body at a certain temperature (high heat) when White is displayed, and is the temperature (K) of the black body at that time. , And the lower the color temperature, the stronger the redness.

  The color reproduction range indicates the entire range of colors that can be expressed by the projector, and may be represented by a two-dimensional range of xy chromaticity or may be represented by a three-dimensional range of the CIELAB color space.

  In the present embodiment, an approximate expression for obtaining an image quality evaluation value from the above-described parameters is used. When generating this approximate expression, a sensory evaluation is performed, and an evaluation result in the sensory evaluation and a parameter are combined. , Increase the correlation with human visual psychological characteristics.

  There are several evaluation methods for sensory evaluation of an image. An evaluation method suitable for this embodiment will be described with reference to FIG. FIG. 5A shows a reference image, an image A, and an image B. The reference image is an ideal image without distortion or the like, and the image A and the image B are the above-described various parameters. Is an image including distortion and the like. In order to accurately reflect various parameters, it is preferable to use simulation images in which distortion and optical factors are artificially changed by image processing. As shown in FIG. 5A, after projecting the reference image in one evaluation unit, the image A is projected, the evaluation in the image A is performed, and after the reference image is projected in the next one evaluation unit, the image B is projected and the evaluation in the image B is performed.

  The evaluation result is evaluated by using the evaluation words shown in FIG. 5B to determine how much the images A and B to be evaluated with respect to the reference image are distorted and the degree of quality in optical factors. A rating word is a word representing each stage, and a MOS value corresponds to each rating word as an evaluation value. In the case shown in FIG. 5 (b), the evaluation is made in seven stages, but is not limited to this, and may be two to six stages, eight stages or more.

  The reference image may be an image quality of a standard liquid crystal projector, for example, and the images A and B serving as evaluation images may be images projected by a DLP projector. Thereby, the relative image quality of the DLP projector with respect to the liquid crystal projector can also be known.

For example, the range of distortion and optical factors that are geometric factors given to the evaluation image can be set to the following ranges, for example. Note that the resolution and graininess are omitted for convenience.
For distortion, the angle error is 0 to 6.0 degrees, and the luminance within 100 in-plane is 200 to 1000 cd / m ^2.
Contrast ratio conforms to the above luminance.
About color temperature, 5000-10000K
For color reproduction range, sRGB ratio 80-120%
In addition, said range is an example and the range different from said range may be sufficient.

  As for the optical factors to be evaluated at the same time, it is preferable to use at least two optical factors because only one monotonic tendency is understood. For example, by using the luminance and the color temperature as optical factors, it is possible to quantitatively know how much the luminance and the color temperature have a strong influence on the image quality. Further, since each optical factor is an independent factor that does not interact with each other, the use of two or more can improve the accuracy of an expression for calculating an evaluation value, which will be described later. In the sensory evaluation evaluation image, it is preferable that all values to be shaken in various parameters have three or more levels, but it is possible to reduce the number of samples somewhat by using two levels. Moreover, it is not always necessary to evaluate all optical factors.

  The degree of image quality difference between the image to be evaluated and the reference image is evaluated using the evaluation word shown in FIG. When the evaluation image is a document, content including many straight lines as shown in FIG. 6 is preferable for evaluating distortion, and in the case of a natural image or animation, a plurality of average values such as a standard image are used. It is preferable.

  Further, the number of test subjects for sensory evaluation is preferably 20 or more, preferably at least 10 or more, and various parameters and evaluation values are analyzed by multiple regression analysis or the like, and an approximate expression for calculating the evaluation values shown in the following equations 1 and 2 obtain. Note that local distortion is strongly perceived in a document with many straight lines, but in the case of a natural image or animation, local distortion is difficult to perceive, so the approximate expression in the document shown in Equation 1 and Equation 2 are used. It is preferable to derive it by dividing it into approximate expressions in natural images and animation.

FIG. 7 shows, as an example, a result of regression analysis of one parameter (luminance) and an evaluation value (MOS value). Each plot is the evaluation value of the subject under each experimental condition, and these distributions are approximated by a straight line. In the present embodiment, since there are a plurality of parameters, FIG. 7 is expanded to a plurality of parameters, a multiple regression analysis is performed, and approximate expressions shown in the following equations 1 and 2 are obtained. In addition, the coefficients a to f and g to k in the following equations 1 and 2 are values obtained by multiple regression analysis or the like, a and g are intercepts, b to f, and h to k are multiple regressions. It is a coefficient. By incorporating this approximate expression into the image evaluation apparatus according to the present embodiment, which will be described later, the projected image is captured, and the overall evaluation of the image is automatically performed based on the geometric and optical factors of the captured image. A value can be obtained. In the present application, the approximate expression shown in the following equations 1 and 2 may be described as a conversion equation for image quality evaluation values.

(Image evaluation apparatus and image evaluation system)
Based on the above contents, the image evaluation apparatus and the image evaluation system in the present embodiment will be described. The image evaluation apparatus and the image evaluation system in the present embodiment can obtain an evaluation value of image quality based on the approximate expression shown in the above equations 1 and 2 from the parameters of the geometric factor and the optical factor. It is.

  FIG. 8 shows a projection system including the image evaluation apparatus according to the present embodiment, and FIG. 9 shows the image evaluation apparatus according to the present embodiment. The projection system in the present embodiment includes an image evaluation device 30, a projection device 31, a screen 32, and the like. The projection device 31 is a short focus projector having a projection distance of about 0.1 to 1.5 m.

  In addition, the image evaluation apparatus 30 in the present embodiment includes an imaging unit 51, an extraction unit 52, a conversion unit 53, an output unit 54, a storage unit 55, a transmission unit 56, and the like.

  The imaging unit 51 captures a projection image projected on the screen 32 by the projection device 31. The projection image projected by the projection device 31 is an image such as a grid pattern, white solid, black solid, or a color bar, and the imaging unit 51 captures these projection images.

  The extraction unit 52 extracts each parameter set with respect to an angle error, an optical factor, and the like from the captured image captured by the imaging unit 51. For example, the angle error is extracted from the grid pattern, the luminance and the color temperature are extracted from the white solid image, the contrast ratio is extracted from the white solid image and the black solid image, and the color reproduction range is extracted from the color bar.

  As shown in FIG. 3, the angle error is extracted from the difference between the ideal state in the lattice pattern image and the projection image projected by the projection device 31. If there are a plurality of angle errors in the projection plane, the peak is extracted. Use the value. Or it is good also as an average value of a plurality of angle errors beyond a threshold (for example, 3 degrees). The luminance is extracted from the average luminance in the projection plane of the projected white solid image. The color temperature is extracted by measuring the color temperature in the projection plane of the projected solid image. The contrast ratio is extracted by dividing the average luminance of the projected white solid image by the average luminance of the projected black solid image. The color reproduction range is extracted by measuring the xy chromaticities of R, G, and B from the image of the color bar including the projected RGB primary colors, and extracting it from the area of the triangle formed by each x value and y value. .

  Note that the image projected for evaluation does not necessarily have to be a solid image, a solid image, a color bar, or the like, as long as it is an image that can detect the optical factors described above. For example, solid images of R, G, and B may be used in place of the color bar image including RGB primary colors. In addition, when the resolution of the projection device 31 is 1280 × 800 pixels, the lattice pattern has an interval between the straight lines of 20 pixels if the vertical straight lines are 64 lines and the horizontal straight lines are about 40 lines. A wide range of distortion can be detected.

  As will be described later, when extracting the resolution of an image, it is preferable to use vertical and horizontal line patterns, and when extracting the granularity of an image, it is preferable to use a gray solid image.

  The resolution (sharpness) is sharper as the resolution is higher, and is often expressed in ppi (pixel per inch). The resolution can be extracted from the contrast ratio composed of white lines and black lines of the line pattern. The higher the contrast ratio, the higher the resolution, and the lower the contrast, the lower the resolution. The line pattern has a vertical direction and a horizontal direction, the resolution can be detected in each direction, and it is preferable to use both line patterns, but either one may be used.

Graininess is unevenness in brightness and color within the projection surface, and is a measure of how much unevenness the gray, which is ideally uniform. The granularity is generally known as RMS granularity (Root Mean Square granularity) using gray luminance dispersion (σ ² ) in the projection plane. In the case of RMS granularity, the deviation from the average luminance is known. Of the root mean square.

  The converting unit 53 stores the approximate expressions shown in the above formulas 1 and 2, and substitutes the parameters extracted in the extractor 52 into the approximate formulas shown in the above formulas 1 and 2 for visual psychology. Convert to an image quality evaluation value that is highly correlated with the characteristics.

  The evaluation value obtained in the conversion unit 53 is not limited to the case where it is obtained by the approximate expression shown in Equation 1 and Equation 2 above, and can be obtained by conversion using a table or the like. The evaluation value obtained by the conversion unit 53 is sent to the output unit 54, and the output unit 54 displays and outputs the evaluation value or the rating word corresponding to the evaluation value. In the present application, the approximate expression shown in the above equations 1 and 2 and the above-described table are the conversion information. In addition to display, the output unit 54 may output sound or print it out. The storage unit 55 stores image data for image evaluation, specifically, image data such as a grid pattern, white solid, black solid, and color bar.

  In order to appropriately extract parameters such as angle error and spatial period, it is preferable that the resolution in the imaging unit 51 is higher than the resolution in the projection device 31. Further, the image picked up by the image pickup unit 51 may be subjected to a process of correcting the influence of lens distortion or the like of the image pickup unit 51, and these factors may be excluded.

  The imaging unit 51 includes an imaging element that converts incident light into an electrical signal, a shutter that allows external light to enter the imaging element only during imaging, a diaphragm that adjusts the amount of incident light, and the electrical signal into digital data. It has an A / D converter for conversion. The imaging unit 51 also includes an image processing circuit that performs lens distortion correction, gamma correction, shading correction, and the like. As the output unit 54, a liquid crystal display or the like can be used.

  The transmission unit 56 transmits to the projection device 31 a projection image to be projected on the screen 32 from the projection device 31, that is, image data such as an image to be evaluated.

(Image evaluation method)
Next, an image evaluation method according to the present embodiment will be described with reference to FIGS. The image evaluation method in the present embodiment is performed using an image evaluation system including the image evaluation apparatus 30 in the present embodiment described above. The image evaluation method in the present embodiment is shown in the steps of obtaining the approximate expression shown in the above-described formulas 1 and 2 shown in FIG. 10 and the obtained formulas 1 and 2 shown in FIG. Based on the approximate expression, an evaluation value is obtained from an image to be evaluated and output. First, a method for obtaining the approximate expressions shown in the above equations 1 and 2 shown in FIG. 10 will be described.

  First, as shown in step 102 (S102), an image for obtaining the approximate expression shown in the above equations 1 and 2 is projected from the projection device 31 onto the screen 32. Specifically, image data for obtaining an approximate expression is transmitted from the image evaluation apparatus 30 to the projection apparatus 31 via the transmission unit 56, and the projection apparatus 31 obtains an image for obtaining an approximate expression based on the image data. Project onto the screen 32. The image projected at this time is, for example, the image shown in FIG.

  Next, as shown in step 104 (S104), sensory evaluation of the projected image projected on the screen 32 is performed. Specifically, each image projected on the screen 32 is evaluated by a person with a score corresponding to the rating word corresponding to the motto shown in FIG.

  Next, as shown in step 106 (S106), it is determined whether or not the sensory evaluation of the projection image has been performed a predetermined number of times. If the sensory evaluation of the projection image has been performed a predetermined number of times, the process proceeds to step 108. If the sensory evaluation of the projection image has not reached the predetermined number of times, the process proceeds to step 102. Therefore, step 102 and step 104 are repeated a plurality of times. For example, projection of an image on the screen 32 and evaluation by a person as shown in FIG. 5A are repeated a plurality of times while changing the person or the like.

  Next, as shown in step 108 (S108), from the image projected on the screen 32 and the score of the image, the approximate expression shown in the above equations 1 and 2 is calculated by multiple regression analysis or the like.

  By the above, the approximate expression shown in the above equations 1 and 2 can be obtained. The obtained approximate expression is stored in the conversion unit 53 or the storage unit 55 of the image evaluation apparatus 30 shown in FIG.

  Next, a method for obtaining and outputting an evaluation value from the image to be evaluated shown in FIG. 11 will be described.

  First, as shown in step 202 (S202), the image of the lattice pattern as shown in FIG. 12A is projected from the projection device 31 onto the screen 32 and captured by the imaging unit 51 of the image evaluation device 30. The image data of the lattice pattern is stored in the storage unit 55 of the image evaluation device 30, transmitted from the storage unit 55 to the projection device 31 via the transmission unit 56, and projected from the projection device 31 onto the screen 32. The image of the projected lattice pattern is captured by the imaging unit 51.

  Next, as shown in step 204 (S204), the extraction unit 52 of the image evaluation apparatus 30 extracts an angle error based on the lattice pattern image captured by the imaging unit 51.

  Next, as shown in step 212 (S212), the solid image as shown in FIG. 12B is projected from the projection device 31 onto the screen 32, and is captured by the imaging unit 51 of the image evaluation device 30. The white solid image data is stored in the storage unit 55 of the image evaluation device 30, transmitted from the storage unit 55 to the projection device 31 via the transmission unit 56, and projected from the projection device 31 onto the screen 32. The projected solid image is picked up by the image pickup unit 51.

  Next, as shown in step 214 (S214), the extraction unit 52 of the image evaluation apparatus 30 extracts the luminance and color temperature based on the solid white image captured by the imaging unit 51.

  Next, as shown in step 216 (S216), the solid image as shown in FIG. 12C is projected from the projection device 31 onto the screen 32 and captured by the imaging unit 51 of the image evaluation device 30. The solid black image data is stored in the storage unit 55 of the image evaluation device 30, transmitted from the storage unit 55 to the projection device 31 via the transmission unit 56, and projected from the projection device 31 onto the screen 32. The projected solid black image is captured by the imaging unit 51.

  Next, as shown in step 218 (S218), the extraction unit 52 of the image evaluation apparatus 30 extracts contrast based on the solid black image captured by the imaging unit 51 and the solid white image captured in step 214. To do.

  Next, as shown in step 222 (S222), the image of the color bar as shown in FIG. 12D is projected from the projection device 31 onto the screen 32 and captured by the imaging unit 51 of the image evaluation device 30. The color bar image is originally a color image, but in FIG. 12D, it is shown in black and white for convenience. The color bar image data is stored in the storage unit 55 of the image evaluation device 30, transmitted from the storage unit 55 to the projection device 31 via the transmission unit 56, and projected from the projection device 31 onto the screen 32. The image of the projected color bar is captured by the imaging unit 51.

  Next, as shown in step 224 (S224), the extraction unit 52 of the image evaluation apparatus 30 extracts a color reproduction range as shown in FIG. 13 based on the color bar image captured by the imaging unit 51. The image shown in FIG. 13 is originally a color image, but is shown in black and white for convenience.

  Next, as shown in step 242 (S242), the conversion unit 53 obtains an evaluation value from the extracted angle error, luminance, color temperature, contrast, and color reproduction range. Specifically, from the angular error extracted in step 204, the luminance and color temperature extracted in step 214, the contrast extracted in step 218, and the color reproduction range extracted in step 224, An evaluation value is obtained from the following approximate expression. This approximate expression is an approximate expression shown in the above formula 1 or 2.

  Next, as shown in step 244 (S244), the evaluation value obtained in step 242 is output from the output unit 54.

  Through the above steps, image evaluation in the image evaluation method in the present embodiment can be performed. In the above description, the order of the images projected by the projection device 31 may be changed.

Next, in the evaluation method according to the present embodiment, for the sake of convenience, a case will be described in which the evaluation image shown in FIG. 6 is the target and evaluation items are evaluated by focusing on distortion, luminance, and color temperature of the DLP projector. The reference image is a liquid crystal projector, the angle error indicating distortion is 0 (°), the luminance is 520 (cd / m ² ), and the color temperature is 6500 (K).

As a result of the sensory evaluation, the multiple regression equation shown in the following equation 3 can be obtained between the MOS value as the evaluation value, and the correlation coefficient between the MOS value and various characteristics is 0.8 or more, which is high. Correlation was observed.

For example, when the luminance is 300 (cd / m ² ), the color temperature is 8000 (K), and the average value of the angle error is 4.5 (°), the MOS value as the evaluation value is −1.5. This means that the image quality is inferior to that of the reference image. On the other hand, when the luminance is 400 (cd / m ² ), the color temperature is 9300 (K), and the average angle error is 3.0 (°), the MOS value is 0.38, which is higher than that of the reference image. It means that the image quality is slightly better. The above is an example focusing on distortion, brightness, and color temperature, but other parameters may be added.

  Further, in the present embodiment, an evaluation value may be obtained by further adding terms relating to resolution and graininess to the approximate expression shown in Equation 1 and Equation 2 above. A flowchart of the image evaluation method in this case is shown in FIG. The flowchart of the image evaluation method shown in FIG. 14 is the same as the flowchart of the image evaluation method shown in FIG. 11, in which a line pattern or a gray solid image is projected, and the projected line pattern or a gray solid image is captured. Contains steps to do.

  Specifically, as shown in step 252 (S252), the image of the line pattern shown in FIG. 15A is projected from the projection device 31 onto the screen 32, and is captured by the imaging unit 51 of the image evaluation device. Also good. The line pattern image data is stored in the storage unit 55 of the image evaluation device 30, transmitted from the storage unit 55 to the projection device 31 via the transmission unit 56, and projected from the projection device 31 onto the screen 32. . The projected image is captured by the imaging unit 51.

  Next, as shown in step 254 (S254), the extraction unit 52 of the image evaluation apparatus 30 may extract the resolution based on the image captured by the imaging unit 51.

  Next, as shown in step 256 (S256), the gray solid image shown in FIG. 15B may be projected from the projection device 31 onto the screen 32 and captured by the imaging unit 51 of the image evaluation device. The gray solid image data is stored in the storage unit 55 of the image evaluation device 30, transmitted from the storage unit 55 to the projection device 31 via the transmission unit 56, and projected from the projection device 31 onto the screen 32. The projected image is captured by the imaging unit 51.

  Next, as shown in step 258 (S258), the extraction unit 52 of the image evaluation device 30 may extract the graininess based on the image captured by the imaging unit 51.

  Next, in step 242 (S242), the conversion unit 53 obtains an evaluation value from the extracted angle error, luminance, color temperature, contrast, color reproduction range, resolution, graininess, and the like. The approximate expression that becomes the conversion formula of the evaluation value is an approximate expression obtained by adding terms relating to resolution and graininess to the approximate expression shown in Equation 1 or 2.

  Next, as shown in step 244 (S244), the evaluation value obtained in step 242 is output from the output unit 54.

  As described above, the image evaluation device, the image evaluation system, and the image evaluation method according to the present embodiment can derive an evaluation value of an image having a high correlation with human visual psychological characteristics. Accordingly, it is possible to appropriately determine whether or not correction is necessary in the image in accordance with human senses, and it is possible to perform appropriate correction in accordance with human senses even when image correction is performed.

  In addition, by using the image evaluation method according to the present embodiment for the projector inspection process, individual differences between projectors close to human visual psychological characteristics can be easily known.

[Second Embodiment]
Next, a second embodiment will be described. In the present embodiment, a process of outputting an evaluation value for each content is added to the image evaluation method in the first embodiment. By setting the evaluation value for each content, for example, when the images are D1, D2, and D3, the evaluation values for the images D1 to D3 are output. The average brightness, the average saturation, etc. can be considered as characteristics that differ depending on the content. The content with lower average luminance has a lower influence of luminance, and the content with lower average saturation has a lower influence of color reproduction range.

For example, an L * a * b * colorimeter may be used to calculate the average luminance and the average saturation. The average luminance is the average value of L * of each pixel of the content, and the average saturation is the average of each pixel of the content. An average value of C * = (a ² + b ² ) ^1/2 may be used. Therefore, the highest luminance (white solid) and the highest saturation (color bar) are set to 1, respectively, and the average luminance (0 to 1) and average saturation (0 to 1) of the content are normalized. Thereafter, the luminance coefficient (c, h) in the approximate expression shown in the above equations 1 and 2 is multiplied by the coefficient (s, t), and the color reproduction range coefficient (f, k) is multiplied by (u, y). ). The approximate expressions obtained in this way are shown in the following equations 4 and 5. Note that the approximate expression shown in the following equations 4 and 5 may be described as a conversion equation for the evaluation value of image quality.

The image evaluation method in the present embodiment will be described with reference to FIG. The evaluation method in the present embodiment is obtained by adding a step of calculating average luminance and average saturation for each content to the image evaluation method in the first embodiment.

  Specifically, as shown in step 362 (S362), an image as evaluation content is projected from the projection device 31 onto the screen 32, and is captured by the imaging unit 51 of the image evaluation device. The image data serving as the evaluation content is stored in the storage unit 55 of the image evaluation device, transmitted from the storage unit 55 to the projection device 31 via the transmission unit 56, and projected from the projection device 31 onto the screen 32. An image serving as the projected evaluation content is captured by the imaging unit 51.

  Next, as shown in step 364 (S364), the extraction unit 52 of the image evaluation device 30 calculates the average luminance based on the image that is the evaluation content captured by the imaging unit 51.

  Next, as shown in step 366 (S366), the extraction unit 52 of the image evaluation device 30 calculates the average saturation based on the image that is the evaluation content captured by the imaging unit 51.

  Next, as shown in step 242 (S242), the conversion unit 53 obtains an evaluation value from an approximate expression that is an evaluation value conversion expression based on the extracted angle error, luminance, color temperature, contrast, color reproduction range, and the like. obtain. The approximate expression that is the conversion formula of the evaluation value is the approximate expression shown in the above formula 4 or 5.

  Next, as shown in step 244 (S244), the evaluation value obtained in step 242 is output from the output unit 54.

  Although the embodiment has been described in detail above, it is not limited to the specific embodiment, and various modifications and changes can be made within the scope described in the claims. Accordingly, it is possible to provide an image evaluation method performed by the image evaluation apparatus and the image evaluation system, a recording medium on which the program is recorded, an external device that provides the program, a projection system including a projection surface and a projection apparatus, and the like.

10 Light source 11 Color wheel 12 DMD
13 Optical system 20 Projection light 21 Screen 30 Image evaluation device 31 Projection device 32 Screen 51 Imaging unit 52 Extraction unit 53 Conversion unit 54 Output unit 55 Storage unit 56 Transmission unit

JP 2010-028411 A Japanese Patent No. 3719944 JP 2006-042376 A

Y Y. Huang et al., "Human Vision Model for Color Break-Up (CBU) and A CBU-less 5.6" Field Sequential Color Display ", ASID'07 (2007).

Claims (13)

An image evaluation device for evaluating the image quality of a projection image projected by a projection device,
An imaging unit for imaging the projected image;
An extraction unit that extracts a plurality of parameters set with respect to a geometric factor of image quality and an optical factor of image quality from a captured image captured by the imaging unit;
Using the conversion information of the relationship between the geometric factor and the optical factor based on human visual psychological characteristics and the evaluation value of the image quality, the image quality of the projection image is determined from the plurality of parameters extracted by the extraction unit. A conversion unit for converting to an evaluation value;
An image evaluation apparatus comprising:   The image evaluation apparatus according to claim 1, wherein the projected image includes an image of a lattice pattern in which a plurality of straight lines intersect.   The parameter set as the geometric factor includes an angle error obtained from an angle formed by a straight line of a reference lattice pattern and a straight line of the lattice pattern projected as the projection image. Item 3. The image evaluation apparatus according to Item 2.   4. The image evaluation apparatus according to claim 1, wherein the projected image includes an image including white, black, RGB primary colors, gray, and a line pattern.   The image evaluation apparatus according to claim 4, wherein the parameter set as the optical factor includes two or more of brightness, contrast ratio, color temperature, resolution, and granularity.   The image evaluation apparatus according to claim 4, wherein the imaging unit measures a distribution of luminance and chromaticity in a projection plane of the projection image.   The image evaluation apparatus according to claim 1, further comprising: a transmission unit that transmits image data of the projection image projected by the projection apparatus to the projection apparatus.   The image evaluation apparatus according to claim 1, wherein a resolution of the imaging unit is higher than a resolution of the projection apparatus.   The image evaluation apparatus according to claim 1, wherein the conversion information is a conversion formula of a relationship between the geometric factor and the optical factor and an evaluation value of image quality. The conversion formula is a sensory evaluation of a projection image projected by the projection device,
The image evaluation apparatus according to claim 9, wherein the image evaluation apparatus is generated by analyzing a relationship between the sensory-evaluated projection image and the sensory evaluation. An output unit that outputs the evaluation value converted by the conversion unit;
The image output apparatus according to claim 1, wherein the output unit displays or outputs a sensory evaluation rating word corresponding to the evaluation value. An image evaluation device for evaluating the image quality of a projection image projected by a projection device,
An imaging unit for imaging the projected image;
An extraction unit that extracts a plurality of parameters set with respect to a geometric factor of image quality and an optical factor of image quality from a captured image captured by the imaging unit;
Using the conversion information of the relationship between the geometric factor and the optical factor based on human visual psychological characteristics and the evaluation value of the image quality, the image quality of the projection image is determined from the plurality of parameters extracted by the extraction unit. A conversion unit for converting to an evaluation value;
An image evaluation system. An image evaluation method for evaluating the image quality of a projection image projected by a projection device,
Capturing the projected image;
Extracting a plurality of parameters set with respect to an image quality geometric factor and an image quality optical factor from the captured image,
Using the conversion information of the relationship between the geometric factor and the optical factor based on human visual psychological characteristics and the evaluation value of image quality, the image quality evaluation value of the projection image is obtained from the extracted parameters. Converting, and
An image evaluation method characterized by comprising: