JP2019101137A - Sight line guiding system - Google Patents

Abstract

To provide natural visual guidance without giving discomfort or stress to a user.SOLUTION: A control unit 1 of a sight line guiding system S1, which divides a visual observation object 10 that lines of sight 4a and 4b of a user 3 turn towards into a visual guidance region 11 and a non-visual guidance region 12, comprises a filter processing unit 9 for guiding the line of sight 4a of the user 3 to the visual guidance region 11 form the non-visual guidance region 12. A laterally-long monitor 2 displays an image obtained based on generation data which has been subjected to filter processing in a filter processing unit 9. The filter processing unit 9 performs bleeding processing accompanied by color shift on at least the non-visual guidance region 12.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a gaze guidance system.

  Conventionally, in the field of image processing, an image processing system using blurring such as filter processing has been studied (see, for example, Patent Documents 1 to 3).

JP, 2007-88781, A JP, 2008-77406, A JP 2002-318561 A

In recent years, WEB and digital signage have been widely spread, and many ideas have been made to effectively impress contents displayed on the medium. As one of the devices, there is an attempt to guide the user's gaze to an intended part in an image.
However, explicit line-of-sight guidance, such as screen blinking or partial animation, gives the user an unpleasant impression. In addition, if the emphasis is on natural visual guidance so as not to be uncomfortable, the effect of visual guidance becomes thin.
For example, in conventional research, gaze guidance is performed by the blurring effect of a picture (picture).
When the gaze guidance method by the blurring effect is applied to gaze guidance for a display with a large amount of edge information such as character information in text, the user recognizes that the surroundings are blurred. In addition, if the degree of blurring is increased, the surroundings are blurred and the readability of characters is impaired. For this reason, there was a problem that it was difficult to make eye gaze guidance and readability compatible.

  Then, this invention makes it a subject to provide the gaze guidance system which can perform natural gaze guidance, without giving a user a feeling of discomfort and stress.

  A gaze guidance system according to the present invention divides a visual target into a gaze guidance region and a gaze non-guidance region in which the user directs the gaze, and guides the gaze of the user from the gaze non-guidance region to the gaze guidance region. , And the filter processing unit is characterized in that bleeding processing with color misregistration is performed at least in the sight non-induction region.

  According to the present invention, natural gaze guidance can be performed without giving a sense of discomfort or stress to the user.

BRIEF DESCRIPTION OF THE DRAWINGS It is a perspective view of an example of the gaze guidance system of this embodiment, and an experimental apparatus which tests the effect. (A)-(c) is a schematic diagram which shows the example of reading of the character information using the gaze guidance system of this embodiment. (A)-(d) is a typical top view which compares the character which used the blur processing used for the gaze guidance system of embodiment, and the character which performed the blurring process. It is a figure explaining the principle of the blur filter of the gaze guidance system of an embodiment. It is a figure which shows the flow of the experiment which tests the effect of a gaze guidance system. Among the effects of the gaze guidance system, (a) is a graph that summarizes positive responses, (b) is a negative response that summarizes the results of the questionnaire conducted to subjects of the experiment of CMY bleeding. It is a graph. Among the effects of the gaze guidance system, (a) is a graph that summarizes positive responses, (b) is a negative response that is collected from the results of questionnaires conducted to subjects of B & W blur experiments. It is a graph. Among the effects of the gaze guidance system, CMY blur by blur filter, B & W blur, blur blur by blur filter, a graph summarizing the results of the questionnaire conducted to the subject of the experiment by the original image, (a) shows a positive response The collected graph, (b) is the graph which collected the negative reaction. It is a gaze guidance system, and it is a flow chart explaining the order of experiment. (A) is an original drawing before processing, (b) is a drawing of the proposed drawing, and (c) is a comparison process of blurring processing. It is the figure which went. (A) is an original drawing before processing, (b) is a drawing of the proposed drawing subjected to the blurring processing, and (c) is a drawing of the blurring processing to be compared. . It is a schematic diagram which shows an example which applies an gaze guidance system of this invention, and shows a mode that it applies to an electronic market. It is a schematic diagram which shows an example which applies an gaze guidance system of this invention, and shows a mode that it applies to digital signage. It is a schematic diagram which shows an example to which the gaze guidance system of this invention is applied, and shows a mode that it applies to information visualization. It is a schematic diagram which shows an example which applies an gaze guidance system of this invention, and shows a mode that it applies to an electronic book. It is a schematic diagram which shows an example which applies an gaze guidance system of this invention, and shows a mode that it applies to an actual real show window. It is a block diagram explaining the structure at the time of applying the gaze guidance system of this invention to an actual real show window.

  Embodiments of the invention are illustrated in detail with reference to the drawings. In the description, the same elements will be denoted by the same reference numerals and redundant description will be omitted.

A gaze guidance system S1 shown in FIG. 1 mainly includes a control unit 1 configured by a PC 7 or the like, and a horizontally long monitor 2 as a monitor device as an output unit.
The control unit 1 is provided with a filter processing unit 9, an area division unit 109, and an image generation unit 209. Further, the control unit 1 is connected to a gaze position detection unit 5 provided at the lower edge of the horizontally long monitor 2.
The gaze position detection unit 5 includes a plurality of cameras 25 that detect the position of the gaze 4 a of the user 3. The camera 25 is connected to the control unit 1 so as to be able to output information of the detected line of sight.
The gaze guidance system S1 of the embodiment uses an artificial color shift as a method of naturally emphasizing a specific area of an object having many edge information such as characters without impairing the readability of the original information. Use.

  FIG. 2: is a schematic diagram which shows the example of interpretation of character information using gaze guidance system S1 of embodiment. The horizontal monitor 2 constitutes a multi-window system having a pair of left and right screens 2a and 2b. Alternatively, the multi-window system may be configured by being connected to a pair of left and right monitors (not shown).

That is, the area dividing unit 109 (see FIG. 1) of the control unit 1 divides the visual target 10 in which the user 3 directs the gaze 4 a into the gaze guidance region 11 and the gaze non-guidance region 12.
Here, the line of sight non-induction area 12 exists around the line of sight guidance area 11 where the user 3 wants to direct the line of sight. That is, the visual line guidance area 11 is partially provided in a part of the visual line non-induction area 12 substantially on the entire surface.

The filter processing unit 9 performs blur processing with color shift in the sight line non-induction region 12. Bleeding processing is performed on the gaze non-guiding area 12.
The image generation unit 209 generates information of the image subjected to the filter processing so as to be outputable to the screens 2 a and 2 b of the horizontally long monitor 2.
The image output from the control unit 1 is displayed on the screens 2a and 2b. In the image displayed on the screen, the characters of the gaze guidance area 11 are displayed with high contrast and are emphasized by a strong, black and white clear display of edges.
Therefore, the line of sight 4 a of the user 3 is guided from the line of sight non-induction area 12 to the line of sight guidance area 11.

  For example, as shown in FIG. 2 (a), an English sentence (hereinafter also referred to as a text) is displayed on the left screen 2a, and a Japanese translated sentence (hereinafter referred to as a translated sentence) corresponding to the right screen 2b is displayed. Do. The user 3 reads a sentence while comparing the left and right screens 2a and 2b.

The user 3 who is looking at the left screen 2a moves the line of sight 4a to the right screen 2b. At this time, the part (L in the figure) which was viewed last on the screen 2a on the left side is stored.
In (b) of FIG. 2, when the line of sight 4 b of the user 3 returns to the screen 2 a on the left side, the line of sight non-induction region 14 other than the part L stored earlier is filtered.
In (c) of FIG. 2, when it is confirmed that the line of sight has returned to L, the filtering of the line of sight non-induction region 14 may be gradually weakened.

The location of the translation in the visual target 10 on the screen 2b on the right side corresponding to the sentence displayed at the place L on the left side of the screen 2a may be set as the elliptical line-of-sight guidance area 11 . Along with this, the periphery of the line-of-sight induction area 11 in the visual target 10 may be set as the line-of-sight non-induction area 12 to blur the line-of-sight non-induction area 12 on the screen 2b.
And, at this time, by limiting the bleeding process to such an extent that the translated text can be read, it is possible to prevent the user 3 from being given stress caused by a sense of discomfort or discomfort.
The line-of-sight guidance area 11 is relatively emphasized by the line-of-sight non-induction area 12 where the bleeding is alleviated, and the line of sight 4a is naturally guided from the line-of-sight non-induction area 12 where the characters are blurred to the line-of-sight guidance area 11.

If the user's gaze is successfully guided, the blur processing of the gaze non-induction area 12 is gradually weakened. Thereby, the stress of the user 3 can be reduced as in the case of the sight line non-induction area 14.
The user naturally repeats the guidance of the gaze 4 b to the gaze guidance area 11 or the gaze guidance area 13.
Then, by pre-reading the current line of sight in the text and filtering the translation, the line of sight can be efficiently applied to the relevant part. Also, even if the line of sight is returned to the text, the line of sight is naturally guided to the place where it was read. Therefore, efficient line-of-sight movement can be performed, and natural line-of-sight guidance can be performed without giving a sense of discomfort or stress to the user.

When emphasizing a visual target such as the line-of-sight guidance area 11, a process of blurring or blurring the surrounding non-line-of-sight guidance area 12 relatively using a blurring filter or a blurring filter is performed.
However, when emphasizing the visual target, if the degree of processing of the filter is strong, the user 3 feels uncomfortable.
Therefore, it was decided to investigate the degree of strength of each filter that the user 3 does not feel unnaturally.

[Bleeding filter]
In FIG. 3, (a) an original image 20 of a character string, (b) a blurred image 21 of a Gaussian blur filter as a blurred image, (c) a CMY blur filter image, and (d) a black only blur B & W blur filter image An example is shown.
Here, “blur” refers to a single color “plate misregistration” or multiple colors “color misregistration”, and the “color misregistration” phenomenon that occurs during projection of the projector is intentionally It is obtained by generating data to be displayed and reproducing it on an output monitor.

Among them, as shown in FIG. 4, the CMY blur filter uses three colors of Cyan (hereinafter also referred to as C component), Magenta (hereinafter also referred to as M component), and Yellow (hereinafter also referred to as Y component). An image is obtained by shifting each by x (x: natural number) pixels.
Specifically, CMY bleeding is generated through the following four steps.
Step S1: An image is generated as a result of shifting the characters in the original image 20 shown in FIG. 4A in three different directions by a designated pixel.
Then, it is colored in three colors of C component, M component and Y component respectively. For this reason, as shown in FIG. 4B, an image as a result of shifting the original image in the left, right, and downward directions by a designated pixel is generated one by one.

Step S2: An alpha channel is added to the original image 20 in order to synthesize an offset image using an alpha blend. Alpha blending is to combine two images by coefficients (α value). Further, the alpha channel is to prepare an image which defines a portion to be transmitted, which is called a mask image at the time of composition, and performs transmission of the image to be transmitted based on it. The information may be given to pixels and transmitted, and the information is called an alpha channel.
Step S3: Each image is colored and integrated to each of Cyan, Magenta, and Yellow.
The RGB values of the image f are expressed as f (R), f (G), f (B), and each value takes an integer value of 0 to 255.
Further, the integration source image (white image) is Base, the image shifted in each of the three directions is cmy, the original image is original, and the result image is a result.

[Equation 1]
result (i) = Base (i)-(α * (255-cmy (i))
(I ∈ R, G, B)
As shown in Equation 1, each cmy and original are subtracted from Base in accordance with the color for each pixel and the transparency α of the alpha blend to obtain each value of result.
Also, in order to color cmy into colors according to each of the three directions, adjustment of the RGB values of cmy is performed. For example, when coloring cmy to Cyan, the values of both cmy (G) and cmy (B) are set to 255.
As a result, since only the cmy (R) value is subtracted from the Base (R), the cyan color is reproduced.
The same adjustment is made for other Magenta and Yellow. This adjustment is for CMY bleeds. In the case of B & W blur, Equation 1 is achieved without adjustment of cmy's RGB values.
Step S4: The original character string position is reproduced using the original image. In the case of CMY bleeding, the area where all the two colors overlap becomes white. For this reason, the calculation of the following equation 2 is finally performed using original. Thereby, the area where all three colors overlap becomes black.
[Formula 2]
result (i) = Base (i)-((1-.alpha.) * (255-original (i))
(I ∈ R, G, B, original (i) = 0)
By the above operation, the result image in which cmy3 sheets are integrated and the original text image are integrated to achieve the result image of the bleeding filter.
Also, the strength of the blur filter can be controlled by the opacity β of the following equation 3.
[Equation 3]
β = 1−α

[Comparative experiment]
A comparative experiment was performed with the blur filter and the blur filter. In the comparative experiment, it was verified whether it was possible to quickly return the line of sight once removed to the original position before movement by the blur filter and the blur filter. In addition, it is preferable to prepare a plurality of images in which the strength of each filter is different according to the change of the parameter and to verify the influence of the difference. Here, a representative example is illustrated and demonstrated.

The device of the comparative experiment is equivalent to the gaze guidance system S1 of FIG. 1 plus a jaw rest 27 shown by a two-dot chain line.
The output monitor 24 has 25 inches and a resolution of 1920 × 1080 pixels. The lower edge of the output monitor 24 is provided with a plurality of cameras 25 of a gaze position detection unit 5 that detects the gaze of the subject, and is connected to the control unit 1 in the PC 7 below the desk 26.
In front of the output monitor 24, a jaw rest 27 is provided which keeps the distance between the output monitor 24 and the head of the subject seated on the chair 28 constant. At the lower part of the output monitor 24, a gaze measuring device is provided. The room in which the device was placed was in a state of lighting and there was no human in the range where the subject's line of sight could be reached.
In addition, the size of the character was 18 pt, the font used Times New Roman Regular.

  A total of 32 subjects, 28 males and 20 females in their twenties and thirties. There were no eye abnormalities in all subjects, and subjects with poor visual acuity conducted comparative experiments with glasses and contact lenses corrected.

[Experimental procedure]
FIG. 5 is a diagram showing the flow of comparative experiments.
(A) The subject 3 reads aloud the text displayed on the output monitor 24. The reading start point is clearly indicated in the text by the red circle only for the first time. Start reading from that point and press the space key when a period comes.
(B) When the space key is depressed, the screen of the output monitor 24 is darkened for 3 seconds, during which a white cross 29 is displayed at any of the four corners on the screen, and the subject looks at the white cross 29.
(C) The darkening is canceled and the sentence is displayed again.
(D) Resume reading aloud from the continuation of the text read halfway up.
The above is repeated, and when the aloud to the specified range is finished, the point when the space key is pressed last is regarded as the end, and the questionnaire answer is shifted.

The breakdown of all 10 trials changed the type and strength of the filter and made 3 trials, 1 trial of the blur filter, 3 trials of 2 types of the blur filter (CMY, B & W) and changed the blur size .
The filter effect was used to emphasize the location when the subject pressed the space key, turned back to the reading resumption point after 3 seconds of darkening.
During the experiment, the subject's line of sight was measured, and when the subject's line of sight returned to the point where the reading was resumed, the highlighting was gradually released.
The change in this case is that an image which has been processed to 20 levels of strength in advance according to each filter is displayed for 0.030 to 0.035 seconds per sheet, and a total change of 0.600 to 0.700 seconds. did. In addition, the subject is informed in advance that there is a situation in which the original original without filtering is displayed.

[preliminary survey]
In order to determine the maximum value of the strength of each filter, we performed a threshold survey of blur and blur in advance, which the user felt unnaturally.
We displayed an image in which meaningless character strings were arranged with the font of Times New Roman Regular and the character size of 60 pt, and processed so that the degree of the filter gradually became stronger with the passage of time.
The pre-survey was conducted on 20 subjects aged 18-24 years old, 18 males and 2 females.
Then, based on the result, a Gaussian blur controlled by the variable σ is obtained. The obtained result was used as the maximum value of the strength of each filter.
This eliminates unnecessary strong filters and streamlines the experiment.

In this experiment, after the text image was displayed on the screen, the subject (user) started reading aloud, completed reading a specified range, and examined the time until the space key was pressed and the gaze position in between. We also conducted a questionnaire every trial.
Table 1 is an example of the questionnaire result performed to the test subject of the experiment which tests the effect of a gaze guidance system. Four questions were prepared, and each question was divided into five grades from 1 to 5. Here, it is the contents asking whether question 1 and question 3 are good impressions, and the question is asking whether question 2 and question 4 are bad impressions.
In addition, when the distance between the subject and the output monitor 24 is 60 cm, a gaze guidance area is generated with a radius of 118 pixels or less, preferably 30 pixels.

[Experimental result]
The experimental results are discussed from the questionnaire results.
First, with regard to the blur filter, the CMY blur and B & W blur questionnaire results are compared.
a. CMY Bleeding Filter FIG. 6 shows the result of the questionnaire of the magnitude of the bleeding of each of the CMY bleeding filters. As shown in FIG. 6, due to the difference in the size of the CMY blur filter (0.5 pixel, 1.0 pixel, 1.5 pixel), the CMY blur filter is most easy to resume reading when using 1.5 pixel. It is easy to continue reading aloud. However, processing of the image is also felt strongly.
On the other hand, when the size of the bleeding was 0.5 pixel, the result that the readability of the resumption point of reading and the readability of the sentence were not impaired so much and the processing of the image was not strong. Also, among the three types of bleeding, it is not the most disturbing.
Therefore, from the viewpoint of natural gaze guidance, the CMY blur when the blur size is 0.5 pixel is treated as a representative of the entire CMY blur. That is, hereinafter, CMY bleeding is assumed to be all 0.5 pixels in size unless otherwise specified.

b. B & W Bleeding Filter FIG. 7 shows the result of the questionnaire for the B & W bleeding filter.
Because of the difference in B & W blur size (0.5 pixels, 1.0 pixels, 1.5 pixels), B & W blur filters are most easy to resume reading and continue reading when 1.5 pixels and 0.5 pixels are used.
However, the B & W blur filter resulted in the 1.5-pixel case being the hardest to read the document of the three.
Also, the processing strength gives strong impression in the order of 1.0 pixel, 1.5 pixel and 0.5 pixel.
From the above results, B & W bleeding when the size of B & W bleeding is 0.5 pixel is treated as a representative of the entire B & W bleeding. That is, hereinafter, B & W bleeding is assumed to be all 0.5 pixels in size unless otherwise specified.

[Compare each filter]
FIG. 8 shows the results of the questionnaire performed on the original image, with CMY blur by the blur filter, B & W blur, blur blur by the blur filter among the effects of the gaze guidance system S1. There are 32 subjects, and the average score corresponds to the question number on the horizontal axis.
B & W is the best result for the readability of the document and the readability of the reading point.
On the other hand, it is B & W that the image processing is most strongly felt, and it can be seen that there is a trade-off between the degree of natural feeling and the readability of sentences and the ease of finding restart points.

With regard to CMY bleeding, the readability and reusability of finding the reopening point are lower than those of the other filters, but the score is better than the trial of the original image without processing. In addition, CMY bleeding did not strongly sense the image processing in all the filters, and did not feel in the way of image processing when reading a sentence.
Therefore, if priority is given to readability in a situation that the user may notice, B & W bleeding is suitable, and CMY bleeding is suitable if aiming to achieve natural gaze guidance.

  Blur blur also tends to be similar to B & W blur. Blur blur is not felt stronger than B & W smearing, but it gives a strong impression of processing compared to CMY blur. The degree of feeling in the way of image processing was the same as B & W bleeding.

  In addition, it was an overall low score result to the question whether it felt that image processing was disturbed. It can be seen that the mechanism of gradually canceling the image processing when the line of sight is returned after the end of the line of sight guidance contributes to the discomfort and the stress alleviation of the subject.

FIG. 9 is a flowchart for explaining the order of experiments in the gaze guidance system S1 of the embodiment.
When the processing is started (started) by the control unit 1 of the gaze guidance system S1, in step S1, the filter processing unit 9 of the gaze guidance system S1 is an ROI (abbreviation of Region of Interests: region of interest; Apply filtering other than just the ROI).
In step S <b> 2, the gaze position detection unit 5 detects the gaze of the user 3.
In step S3, the control unit 1 determines whether the line of sight of the user 3 has entered the ROI.
If the controller 1 determines that the line of sight of the user 3 has entered the ROI (YES in step S3), the process proceeds to the next step S4, and if it is determined that the line of sight of the user 3 does not enter the ROI (step S4) (NO at S3) returns the process to step S2.
In step S4, the filter processing unit 9 gradually cancels the filtering (filtering process).
In step S5, the gaze position detection unit 5 detects the gaze of the user 3.
In step S6, it is determined by the gaze position detection unit 5 whether or not the gaze of the user 3 has moved to another process (window). If it is determined that the line of sight of user 3 has moved to another process (YES in step S6), the process proceeds to the next step S7, and the line of sight of user 3 has not moved to another process (NO in step S6) If it is determined that the condition is, the process returns to step S5.

In step S7, the area around the last fixation point is set to the ROI of this process, and the process proceeds to another process. It is assumed that, for each window, the ROI at the time when the line of sight left the window last time is held. That is, when the line of sight leaves, the last fixation point is left as the ROI of the window of the line-of-sight guidance area.
Thus, when the line of sight returns to the window, it is possible to start blurring (blurring) other than the previously recorded ROI.
Therefore, when performing work or the like while reciprocating a plurality of screens 2a and 2b and work areas on a plurality of screens such as a multi-window system, natural visual line guidance can be performed continuously.

  In step S8, it is determined whether the process has ended. If the process is not completed (NO in step S8), the process returns to step S1 to repeat the process. When the process ends (YES in step S8), the control unit 1 ends the series of processes (END).

FIG. 10 illustrates an image of a document used in the gaze guidance system S1 of the embodiment. In addition, in order to facilitate the explanation, it is schematically illustrated by emphasizing in part.
FIG. 10 (a) is an original original drawing 101 before processing. For example, it is a text composed of characters having a large edge (border) contrast and strong edges as compared with a pictorial diagram (see FIG. 11) in which the contrast is lowered due to a neutral color or gradation.
FIG. 10B is a diagram in which the bleeding process is performed by the CMY blur filter according to the embodiment. In the line-of-sight non-induction area 112 at the periphery of the line-of-sight guidance area 111, the black characters are subjected to blurring processing with color deviation.
FIG. 10C is a diagram in which the blurring process by the Gaussian blur filter of the comparative example is performed. Referring to (b) and (c), it appears that the gaze guidance effect is in the gaze guidance area 211. However, in the case of (c) of the comparative example, if it is attempted to obtain a sufficient gaze guidance effect, the filter of the gaze non-guiding region 212 obfuscates characters.
On the other hand, in (b) of the embodiment, since the blur processing by the CMY blur filter is performed, the visual line non-induction region 112 is not conscious of the feeling of incongruity and not conscious of the filter being applied. You can keep reading the letters in

FIG. 11 exemplifies an image of a picture (photograph) used for the gaze guidance system S1 of the embodiment. In addition, in order to facilitate the explanation, it is schematically illustrated by emphasizing in part.
FIG. 11A is an original original drawing 201 before processing. For example, the edge is weak due to a neutral color or gradation. For this reason, the picture is hard to catch at the border (boundary).
FIG. 11B is a diagram in which the bleeding process is performed by the CMY blur filter according to the embodiment. The peripheral line of sight non-induction region 312 is subjected to bleeding processing with color shift. However, it is difficult to distinguish between the gaze guidance region 311 and the gaze non-guide region 312, and it is difficult to obtain the gaze guidance effect.
FIG. 11C is a diagram in which the blurring process by the Gaussian blur filter of the comparative example is performed. In (c), the visual line guidance area 311 and the visual line non-induction area 312 can be easily distinguished, and the visual line guidance effect can be obtained.
As described above, in an original drawing having weak edges such as a picture or a picture, the blurring process using the Gaussian blur filter is likely to obtain the line-of-sight guidance effect.

  FIG. 12 is a schematic view showing an example of application of the gaze guidance system S1 of the present invention, which is applied to an electronic market. In this embodiment, the screen of the electronic advertisement medium 500 is divided into a gaze guidance area 511 and a gaze non-guiding area 512.

  FIG. 13 shows an example of application of the visual line guidance system S1 of the present invention, and is a schematic view showing application to digital signage. In this embodiment, the screen of the digital signage 600 is divided into a gaze guidance area 611 and a gaze non-guidance area 612.

  FIG. 14 is a schematic view showing an example of application of the visual line guidance system S1 of the present invention, which is applied to information visualization. In this embodiment, the screen of the 3D display 700 on the three-dimensionalized screen is divided into a gaze guidance area 711 and a gaze non-guidance area 712.

  FIG. 15 is a schematic view showing an example of application of the visual line guidance system S1 of the present invention, which is applied to an electronic book. In this embodiment, the screen of the electronic book 800 is divided into a gaze guidance area 811 and a gaze non-guidance area 812.

FIG. 16 and FIG. 17 show an example of applying the gaze guidance system S2 of the modification, and are a schematic diagram and a block diagram showing application to an actual show window 900. The same or equivalent parts as those of the visual line guidance system S1 of the above embodiment will be described with the same reference numerals.
As shown in FIG. 16, this gaze guidance system S2 is replaced by the screens 2a and 2b configured by the horizontally long monitor 2 of the gaze guidance system S1 of the embodiment, and is output to the output unit 102 that projects data generated by the PC7. An external projector 8 is provided as a projector device. The portion where the output unit 102 is provided may be installed anywhere inside or outside the actual show window 900, and may be reflected or transmitted by a mirror, glass or the like.
The external projector 8 distinguishes and projects the gaze guidance area 911 and the gaze non-guiding area 912 when projecting the shadow image on the articles 901 to 902 displayed in the actual show window 900.

As shown in FIG. 17, in the visual line guidance system S2 of the modified example configured in this way, in addition to the effects of the embodiment, the external projector 8 is used to add subtleties such as blur to actual products. You can project the filtered image to.
For example, on the article 901 and the article 902, a finely filtered image such as blur is projected. The article 901 and the article 902 change the appearance of the real object according to the projected image. As a result, relatively, the unprojected article 903 in the line of sight guidance area 911 floats, and can naturally guide the line of sight so that the line of sight of a person is poured on the article 903.
In the modified examples shown in FIGS. 12 to 17, the other configurations and effects than those described above are the same as or equivalent to those of the embodiment, and thus the description thereof will be omitted.

  The gaze guidance systems S1 and S2 and the image generation program according to the present embodiment have been described above in detail, but the present invention is not limited to these embodiments, and appropriate changes can be made without departing from the spirit of the present invention It goes without saying that it is possible.

  For example, in the present embodiment, as the blur filter using an artificial color shift, the use of a CMY blur filter has been shown and described, but the present invention is not particularly limited thereto. For example, an RGB filter or a black and white B & W blur filter may be used, and the number, color scheme, and combination of blurs may be any as long as the blur processing with color shift is performed. .

  Further, the directions of shifting the respective colors and the shifting amounts are not particularly limited. For example, it is not limited to the ones which are shifted at an equal angle radially, but may be ones which are not at equal angles but in other directions, and the x pixels which are shifted may be different for each color.

Furthermore, on a screen in which a picture and text information are mixed, a blur image of a blur filter and a CMY blur filter image may be mixed in one screen.
Further, the transparency of the color may be changed for each color element and displayed with a blur according to the application.

  And although the control part 1 is divided and described in the filter processing part 9, the area division part 109, and the image generation part 209 in the embodiment, the present invention is not limited to this, and the processing is integrated in the control part 1. It may be done. Furthermore, the control unit 1 is not limited to a single unit, and a plurality of processing units such as four or eight units may be combined, and the processing capability, quantity, type, etc. of the control unit 1 and the internal processing units Is not particularly limited.

  Furthermore, a CMY blur filter, a black and white blur B & W blur filter, or a Gaussian blur filter may be made to correspond to part or all of the character part and the picture part, respectively. That is, a plurality of filter processing units 9 for performing different types of filter processing may be provided, and different filter processing may be performed for each area divided by the area dividing unit 109, and the type and number of filters are particularly limited. It is not a thing.

  In the modification, the filter image is projected on the articles 901 to 902 displayed in the actual show window 900 using the external projector 8. However, the present invention is not limited to this, and it may be a real object. There are no particular limitations on the shape, number, and material of what is projected.

1 control unit 2 horizontal monitor (output unit)
2a, 2b Screen 3 Users 4a to 4c Line of sight 5 Line of sight position detection unit 8 External projector (output unit)
9 filter processing unit 10 visual object 11, 111 to 911 gaze guidance region 12, 112 to 912 gaze non induction region 900 real show window 901 to 903 articles S1, S2 gaze guidance system

Claims (8)

The image processing apparatus further comprises a filter processing unit that divides the visual target into a gaze guidance area where the user looks at the gaze and a gaze non-guidance area, and guides the gaze of the user from the gaze non-guidance area to the gaze guidance area.
The line-of-sight guidance system, wherein the filter processing unit performs blur processing with color shift in at least the line-of-sight non-induction area.   The gaze guidance system according to claim 1, further comprising an output unit for displaying the generated data filtered by the filter processing unit.   The gaze guidance system according to claim 2, wherein the output unit is a monitor device.   The gaze guidance system according to claim 2, wherein the output unit is a projector device.   When the filtering processing unit performs bleeding processing in the non-visual-direction area, it uses a bleeding image obtained by shifting at least one of C, M, and Y components of the original image of the edge portion with high contrast. The gaze guidance system according to any one of claims 1 to 4, characterized by the above.   The line-of-sight guidance system according to any one of claims 1 to 5, wherein the filter processing unit gradually cancels the blurring process by the image processing of the line-of-sight non-induction region after the end of line-of-sight guidance.   The said filter process part leaves the last gaze point as ROI of the window of the said gaze guidance area | region, when eyes gaze leaves the said gaze guidance area | region, It is characterized by the above-mentioned. Gaze guidance system. A filter processing unit that divides a visual target into a gaze guidance area and a gaze non-guidance area where the user directs the gaze, and guides the gaze of the user from the gaze non-guidance area to the gaze guidance area; A projector device for projecting the generated data filtered at step
The visual line guidance system, wherein the filter processing unit generates a shadow image projected from the projector device at least in the visual line non-induction region.