Production support system, production support method, and production support program

Abstract

To provide appropriate support for producing visually recognized products.SOLUTION: A production support system 1 disclosed herein provides support for producing a visually recognized product. The production support system 1 comprises an image acquisition unit 2, impression inference unit 3, and display unit 4. The image acquisition unit 2 acquires an input image of the product. The impression inference unit 3 infers a degree to which information contained in the input image impresses a person who views the input image. The display unit 4 displays a degree to which memory of the input image stays.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a technology for supporting the creation of visually recognized works. [Background technology]

[0002] Visually perceived works, such as web pages, images, or advertisements such as posters, are sometimes commissioned to be created by designers or other creators. The client who requests the creation of a work communicates vague details of the request, such as the design concept or similar examples, to the creator. The creator creates the work based on these details and delivers the finished work to the client. However, sometimes the intentions of either the client or the creator are not properly conveyed to the other. In this case, the client communicates the revisions to the creator, and the creator revise the work based on the revisions communicated by the client and delivers the revised work back to the client. If such revisions are required multiple times, it can take a lot of time to complete the work.

[0003] It is effective to use some objective indicators to enable the client and the creator to discuss the work and reach a common understanding.

[0004] Patent Document 1 discloses a technology for calculating saliency, an evaluation scale that indicates the degree to which a subject's attention is likely to be drawn to the image. The technology disclosed in Patent Document 1 extracts the features of each part in image data and quantifies the saliency based on the extracted features of each part.

[0005] Furthermore, Patent Document 2 discloses a technology that uses terms that express emotions to display the impression of an image to an operator, allowing the operator to use design know-how when adding expressions. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-236508 [Patent Document 2] Japanese Patent Application Publication No. 9-16797 Summary of the Invention [Problem to be solved by the invention]

[0007] Web pages, images, or advertisements such as posters are created with the purpose of visually impressing the viewer with information, so the indicators must measure how much the information impresses the viewer.

[0008] The saliency calculated by the technology disclosed in Patent Document 1 relatively indicates where within an image people's eyes are likely to be drawn, and does not relate to how much information the entire image as a creative work impresses on people.

[0009] The impression of the image displayed to the operator by the technology disclosed in Patent Document 2 does not relate to how much the image impresses people with information as a product.

[0010] One object of the present disclosure is to provide a technology that appropriately supports the creation of visually recognized works. [Means for solving the problem]

[0011] A production support system according to one embodiment of the present invention is a production support system that supports the production of visually recognized works, and includes an image acquisition unit that acquires an image of the work, a memory estimation unit that estimates the degree to which the information contained in the image will impress a person who views the image, and a display unit that displays the degree to which the information will impress a person. [Effects of the Invention]

[0012] According to the present invention, it is possible to appropriately support the creation of visually recognized works. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is an overall view of a production support system according to a first embodiment. [Figure 2] 3A and 3B are schematic diagrams for explaining an input image and a saliency map according to the first embodiment. [Figure 3] 3 is a flowchart of a production support process according to the first embodiment. [Figure 4] 4 is a flowchart of an impression estimation process according to the first embodiment. [Figure 5] FIG. 11 is an overall view of a production support system according to a second embodiment. [Figure 6] FIG. 10 is a schematic diagram for explaining an input image according to the second embodiment. [Figure 7] FIG. 10 is a schematic diagram for explaining a saliency map according to the second embodiment. [Figure 8] FIG. 10 is a schematic diagram for explaining an information plate area ratio according to the second embodiment. [Figure 9] 10 is a flowchart of an impression estimation process according to the second embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, several embodiments will be described with reference to the drawings. [Example]

[0015] FIG. 1 is an overall view of a production support system according to a first embodiment, and FIG. 2 is a schematic diagram for explaining an input image and a saliency map according to the first embodiment.

[0016] The production support system 1 of this embodiment is a system that supports the production of visually recognized works. For example, the production support system 1 is used when a client requests a creator such as a designer to create a work and the created work is evaluated by the client, the client requests the creator to revise the work, or the creator communicates the intention of the work to the client. The work may be, for example, a web page, an image, or an advertisement such as a poster.

[0017] The production support system 1 includes an image acquisition unit 2, an impression estimation unit 3, and a display unit 4. The image acquisition unit 2 acquires an input image 5 (see FIG. 2) as an example of an "image" of a production. The impression estimation unit 3 estimates the degree to which the information contained in the input image 5 will impress a person who views the input image 5. The degree to which the information contained in the input image 5 will impress may be the degree to which the person who views the input image 5 will find the input image 5 difficult to forget. For example, a human face, an image viewed for the first time, or an image with detailed images is likely to impress a person who views the image, and the degree to which the information will impress is high.

[0018] 1, the impression estimation unit 3 includes a gaze evaluation unit 31, a white space analysis unit 32, and an index calculation unit 33. The gaze evaluation unit 31 calculates information saliency indicating the degree of saliency of information (characters, colors, figures) that each region of the input image 5 provides to human vision.

[0019] The input image 5 in Figure 2 is an image that displays a mountain range consisting of two mountains, one large and one small, connected to each other within a rectangular frame, with the sun (or moon) positioned above the lower of the two mountains. The input image 5 may also be an image that has been input as a digital image of a paper medium such as a poster.

[0020] When the input image 5 of FIG. 2 is input, the gaze evaluation unit 31 generates a saliency map 6 that visually represents the degree of information saliency of each region of the input image 5 based on the input image 5.

[0021] 2 shows two horizontally elongated elliptical high salience regions 61 and 64 as examples of "information regions." The high salience information regions 61 and 64 are regions in which information salience is higher than a predetermined threshold.

[0022] The high salience information region 61 is displayed near the center of the taller of the two large and small peaks. Outside the high salience information region 61, a medium salience information region 62 is displayed as an example of a horizontally elongated elliptical annular "information region" having lower information salience than the high salience information region 61. Outside the medium salience information region 62, a weak salience information region 63 is displayed as an example of a horizontally elongated elliptical annular "information region" having lower information salience than the medium salience information region 62. That is, the high salience information region 61, the medium salience information region 62, and the weak salience information region 63 are displayed in that order from the inside out.

[0023] The high salience information region 64 is displayed at the position of the sun (or moon) in the input image 5. A medium salience information region 65 is displayed outside the high salience information region 64 as an example of a horizontally elongated elliptical annular "information region" having lower information salience than the high salience information region 64. A strong salience information region 66 is displayed outside the medium salience information region 65 as an example of a horizontally elongated elliptical annular "information region" having lower information salience than the medium salience information region 65, with an area larger than the sun (or moon) in the input image 5. That is, the high salience information region 64, the medium salience information region 65, and the weak salience information region 66 are displayed in this order from the inside out.

[0024] The high salience regions 61 and 64, the medium salience information regions 62 and 65, and the weak salience regions 63 and 66 are regions that contain information (text, color, and graphics) in the input image 5. The strong salience information regions 61 and 64 are regions where information salience is higher than a predetermined threshold.

[0025] Based on the information saliency calculated by the gaze evaluation unit 31, the margin analysis unit 32 distinguishes between information regions (high salience regions 61 and 64, medium salience regions 62 and 65, and weak salience regions 63 and 66) that contain information in the input image 5 and a margin region 67 that does not contain information. Specifically, the margin analysis unit 32 generates an edge image of the input information 5, scans a margin determination region of a predetermined size across the entire area of ​​the generated edge image, and identifies the margin region 67 based on the proportion of the area of ​​edge portions contained in the margin determination region. Here, the edge image is a binary image in which edge portions detected by edge detection processing are displayed in white and other portions are displayed in black. The margin region 67 in FIG. 2 is the region of the input image 5 other than the high salience information regions 61 and 64, the medium salience information regions 62 and 65, and the weak salience information regions 63 and 65.

[0026] The index calculation unit 33 calculates the information strength and the information area ratio based on the information salience, the information regions (high salience information regions 61 and 64 and weak salience information regions 63 and 66) and the blank region 67.

[0027] The information intensity is an index that indicates how efficiently information is distributed in the information regions (high salience information regions 61, 64 and weak salience information regions 63, 66) of the production (input information 5). The index calculation unit 33 calculates the ratio of the area of ​​the high salience information regions 63, 66 to the area of ​​the weak salience information regions 61, 63 as the information intensity. The information intensity is an index that indicates that the larger the value, the more likely it is to attract the attention of people who view the input image 5.

[0028] The information area ratio is an index that indicates how dispersed the areas of a production (input information 5) are in the input image 5, to which the gaze of a person viewing the input image 5 is directed. The index calculation unit 33 calculates the information area ratio as the ratio of the area of ​​the information areas (high salience information areas 61, 64 and weak salience information areas 63, 66) to the area of ​​the input image 5 of the production. The information area ratio is an index in which the smaller the value, the more simply the parts that attract attention are organized.

[0029] The display unit 4 displays the degree to which the information contained in the input image 5 impresses a person viewing the input image 5. Specifically, the display unit 4 displays a saliency map 6 that visually represents the degree of information saliency in each region of the image 5. The display unit 4 may be a screen such as a monitor.

[0030] In this way, the saliency map 6 displays the high salience regions 61, 64, the medium salience regions 62, 65, and the weak salience regions 63, 65 in the input information 5, as well as the blank region 67, in a distinguishable manner. This makes it possible to quantify and display the degree to which the information contained in the input image 5 of the product leaves an impression on a person viewing the input image 5 of the product.

[0031] FIG. 3 is a flowchart of the production support process according to the first embodiment.

[0032] In the production support process, first, the image acquisition unit 2 acquires an input image 5 of a production (S301). That is, the client or creator inputs the input image 5 of a production created by the creator into the image acquisition unit 2. Next, the impression estimation unit 3 executes an impression estimation process (described later) that estimates the degree to which information contained in the input image 5 of the production will impress a person who views the input image 5 of the production (S302). Next, the display unit 4 displays a saliency map 6 that indicates the degree to which information contained in the input image 5 will impress a person who views the input image 5 (S303).

[0033] FIG. 4 is a flowchart of the storage estimation process according to the first embodiment.

[0034] Specifically, in the impression estimation process, first, the gaze evaluation unit 31 calculates information saliency, which indicates the degree of saliency of information provided to human vision in each region of the input image 5 (S401). Next, the margin analysis unit 32 distinguishes, based on the information saliency, information regions (high salience information regions 61, 64 and weak salience information regions 63, 66) that contain information from a margin region 67 that does not contain information in the input image 5 (S402). Next, the index calculation unit 33 estimates information intensity and information coverage ratio for the production (input image 5) based on the information saliency, the information regions (high salience information regions 61, 64 and weak salience information regions 63, 66), and the margin region 67 (S403).

[0035] According to this embodiment, a production support system 1 that supports the production of a visually recognized work includes an image acquisition unit 2, an impression estimation unit 3, and a display unit 4. The image acquisition unit 2 acquires an input image 5 of the work. The memory estimation unit 3 estimates the degree to which information contained in the input image 5 will impress a person who views the input image 5. The display unit 4 displays the degree to which the information contained in the input image 5 will impress a person. This improves communication between a client who requests the production of a visually recognized work and the creator of the work, making it possible to provide appropriate support for the production of the work.

[0036] The impression estimation unit 3 includes a gaze evaluation unit 31, a margin analysis unit 32, and an index calculation unit 33. The gaze evaluation unit 31 calculates information saliency, which indicates the degree of saliency of information provided to human vision in each region of the input image 5. The margin analysis unit 32 distinguishes, based on the information saliency, information regions (high salience information regions 61, 64 and weak salience information regions 63, 66) that contain information in the input image 5 from margin regions 67 that do not contain information. The index calculation unit 33 calculates, based on the information saliency, the information regions (high salience information regions 61, 64 and weak salience information regions 63, 66), and the margin regions 67, information intensity, which indicates how efficiently information is distributed in the information regions of the product (input information 5), and an information coverage ratio, which indicates how widely the regions to which the gaze is directed are distributed in the input image 5. This makes it possible to appropriately estimate the degree to which the information contained in input image 5 impresses a person who views input image 5 of a visually perceived work, based on both the information intensity and information area ratio indicators.

[0037] The index calculation unit 31 calculates the ratio of the area of ​​the information regions (high salience regions 61, 64 and weak salience regions 63, 66) to the area of ​​the input image 5 of the product as the information area ratio. This makes it possible to appropriately calculate how much the areas to which the gaze of a person viewing the input image 5 is directed are distributed in the input image 5.

[0038] The index calculation unit 31 determines, among the information regions (high salience regions 61, 64 and weak salience regions 63, 66), regions whose information salience is higher than a predetermined threshold as high salience regions 61, 64, and regions whose information salience is lower than the predetermined threshold as weak salience regions 63, 66, and calculates the ratio of the area of ​​the high salience regions 61, 64 to the area of ​​the weak salience regions 63, 66 as information intensity. This makes it possible to appropriately calculate how efficiently information is carried in the information regions of the input image 5. [Example]

[0039] A production support system 10 according to Example 2 will be described. Example 2 corresponds to a modified example of Example 1. The production support system 10 according to Example 2 differs from the production support system 1 according to Example 1 only in the configurations of the impression estimation unit 30 and the display unit 40, and the other configurations are the same as those of the production support system 1 according to Example 1. Therefore, the differences from Example 1 will be mainly described.

[0040] FIG. 5 is an overall view of a production support system according to the second embodiment, and FIG. 6 is a schematic diagram for explaining an input image according to the second embodiment.

[0041] The impression estimation unit 30 includes a gaze evaluation unit 31, a white space analysis unit 32, an index calculation unit 33, and an impression word identification unit .

[0042] The line-of-sight evaluation unit 31 calculates information saliency, which indicates the degree of saliency of information (characters, colors, figures) given to human vision in each region of the input image 7 (see FIG. 6).

[0043] The input image 71 in Figure 6 has a white background. The input image 71 displays the letters "DE," "C," "R," "E," "o," "G," "I," and "a" arranged clockwise from the top of the figure in a circular pattern.

[0044] Each of the letters "DE," "C," "R," "E," "o," "G," "I," and "a" is surrounded by a blue, pink, red, orange, yellow, light-green, green, or light blue circle, respectively.

[0045] The letter "DE" is followed by the letter "sign," the letter "C" is followed by the letter "oncept," the letter "R" is followed by the letters "ecursive" and "e:gi_dec" above and below, the letter "E" is followed by the letter "valuation," the letter "o" is followed by the letter "f," the letter "G" is followed by the letter "raphical," the letter "I" is followed by the letter "mpression," and the letter "a" is followed by the letter "nd" on the right side of each figure.

[0046] The size of the letters "Re:gi_dec" is larger than the size of the letters "DEsign", "Concept", "Evaluation", "of", "Graphical", "Impression", and "and".

[0047] Below the "Re:gi_dec" text, the words "TENSOR CONSULTING" are displayed. The "TENSOR CONSULTING" text is displayed at a size between the "Re:gi_dec" text and the other text.

[0048] When the input image 7 of FIG. 6 is input, the gaze evaluation unit 31 generates a saliency map 8 that visually represents the degree of information saliency of each region of the input image 7 based on the input image 7.

[0049] The gaze evaluation unit 31 may generate a saliency grayscale image in which the higher the information saliency, the lighter the color. The display unit 40 then displays a saliency map 8 (see FIG. 7 ) generated by overlaying the saliency grayscale image generated by the gaze evaluation unit 31 on the input image 7 of the product.

[0050] The saliency grayscale image is an image obtained by combining an edge image of the product with a variation image that represents the difference value relative to the image value that occupies the largest area within the product.

[0051] Furthermore, the display unit 40 may display a saliency map 8 generated by overlaying the saliency grayscale image generated by the gaze evaluation unit 31 onto the input image 7 of the production so that the input image 7 can be seen through it.

[0052] FIG. 7 is a schematic diagram illustrating the saliency map according to the second embodiment.

[0053] The saliency map 8 in FIG. 7 is generated by overlaying a saliency grayscale image generated from the input image 7 in FIG. 6 onto the input image 7 so that the input image 7 of the work is visible through the image. Specifically, in the saliency map 8 in FIG. 7, the letter "DE" surrounded by a blue circle, the letter "C" surrounded by a pink circle, the letter "R" surrounded by a red circle, the letter "E" surrounded by an orange circle, and the letter "I" surrounded by a green circle have high information salience values ​​and are therefore displayed in light colors. On the other hand, the letter "o" surrounded by a yellow circle, the letter "G" surrounded by a yellow-green circle, and the letter "a" surrounded by a light blue circle have low information salience values ​​and are therefore displayed in dark colors. In other words, the letter "o," the letter "G," and the letter "a" have information salience values ​​closer to the margin area. Thus, in the saliency map 8, the closer the color of information (characters, figures) in the input image 7 is to the margin area, the lower the information salience.

[0054] 7, the latter letter "e" in "Re:gi_dec" has the highest information salience value. That is, in the salience map 8, the larger the size of the information (characters, figures), the higher the information salience value.

[0055] The impression word specifying unit 34 specifies impression words that indicate the impression that the input image 7 gives to people, based on the colors contained in the input image 7 as a whole.

[0056] Specifically, the color analysis unit 33 assigns one or more words that represent the impression given to human vision to each hue and tone of the PCCS (Practical Color Coordinate System: Japan Color Research Institute Color Coordination System) in advance as impression allocation information, and identifies words that represent the impression given to humans by the input image 7 based on the hue and tone of each area used in the input image 7 and the impression allocation information.

[0057] For example, input image 7 in FIG. 6 displays black and red on a white background. White is a color that gives a clean and simple impression, black is a color that gives an impression of "chic and cool" as an example of an "impression word," and red is a color that gives an impression of "cheerful and strong" as an example of an "impression word." However, black and red are also colors that give an impression of "danger and fear" as an example of an "impression word." Note that in input image 7 in FIG. 6, the ratio of the area of ​​the black and red colors to the area of ​​the white background is small, and they are not enough to draw the gaze of a person viewing input image 5.

[0058] Furthermore, input image 7 in Figure 6 displays pure colors (vivid tones, bright tones) around the characters. Pure colors are colors that give a "bright and vivid" impression, which is an example of an "impression word." However, when there are many different types of pure colors or when they are displayed over a wide area, they can also give a "gaudy and unsettling" impression, which is an example of an "impression word."

[0059] The display section 40 may display the impression words identified by the impression word identification section 34.

[0060] FIG. 8 is a schematic diagram for explaining the information plate area ratio according to the second embodiment.

[0061] Specifically, in the saliency map 9 of Fig. 8, the string of small letters is identified as a single figure. Furthermore, the space between the outer edge of the red circle surrounding the letter "R" and the letter "R" itself is identified as a blank space 91. Specifically, the information area ratio of the saliency map 9 is 0.291.

[0062] FIG. 9 is a flowchart of the impression estimation process according to the second embodiment.

[0063] Specifically, in the impression estimation process, first, the gaze evaluation unit 31 calculates information saliency, which indicates the degree of saliency of information given to human vision in each region of the input image 7 (S401). Next, the margin analysis unit 32 identifies information regions containing information and margin regions not containing information in the input image 7 based on the information saliency (S402). Next, the index calculation unit 33 estimates the information intensity and information coverage ratio for the product (input image 5) based on the information saliency and the information regions and margin regions (S403). Next, the impression word identification unit 34 identifies impression words that indicate the impression given to humans by the input image 7 based on the colors contained in the input image 7 (S901).

[0064] According to this embodiment, the impression estimation unit 30 calculates information saliency indicating the degree of saliency of information given to humans in each region of the input image 7, and the display unit 40 displays a saliency map 8 generated by superimposing a saliency grayscale image, in which the higher the information saliency, the lighter the color, on the input image 7 of the product. This makes it easier for a client requesting the production of a product and the creator of the product to visually recognize the information saliency indicating the degree of saliency of information given to human vision in each region of the input image 7 of the product.

[0065] The impression estimation unit 30 further includes an impression word identification unit 34 that identifies impression words that indicate the impression that the input image 7 gives to people based on the colors contained in the input image 7, and the display unit 40 displays the impression words. This makes it possible to appropriately estimate, based on the impression words, the degree to which the information contained in the production (input image 5) makes an impression on people who view the input image 5 of the production.

[0066] The color analysis unit 33 assigns in advance, as impression allocation information, one or more words that represent the impression given to human vision to each hue and each tone of the PCCS, and identifies words that represent the impression given to humans by the input image 7 of the product based on the hue and tone of each region used in the input image 7 of the product and the impression allocation information. This makes it possible to appropriately identify the impression given to humans by the input image 7 of the product.

[0067] The above-described embodiments of the present invention are merely examples for explaining the present invention, and are not intended to limit the scope of the present invention to these embodiments. Those skilled in the art can implement the present invention in various other forms without departing from the gist of the present invention. [Explanation of symbols]

[0068] 1,10...production support system, 2...image acquisition unit, 3,30...impression estimation unit, 4,40...display unit, 5,7...input image, 6,8,9...saliency map, 31...gaze evaluation unit, 32...margin analysis unit, 33...index calculation unit, 34...impression word identification unit, 61,64...high salience information region, 62,65...medium salience information region, 63,66...weak salience information region

Claims

1. A production support system that supports the production of visually recognized works, an image acquisition unit that acquires an image of the work; an impression estimation unit that estimates the degree to which information included in the image will impress a person who views the image; a display unit that displays the degree to which the information is impressive; Equipped with The impression estimation unit a gaze evaluation unit that calculates information saliency indicating the degree of saliency of information given to human vision in each region of the image; a margin analysis unit that identifies an information area containing information and a margin area not containing information in the image based on the information saliency; an index calculation unit that calculates, based on the information saliency, the information area, and the blank area, an information intensity that indicates how efficiently information is placed in the information area of ​​the product, and an information area ratio that indicates how areas to which gazes are directed are dispersed in the image; A production support system with

2. the index calculation unit calculates the ratio of the area of ​​the information region to the area of ​​the image of the production as the information area ratio; The production support system according to claim 1 .

3. the index calculation unit determines, among the information regions, a region in which the information saliency is higher than a predetermined threshold as a strong salience information region, and a region in which the information saliency is lower than the predetermined threshold as a weak salience information region, and calculates, as the information intensity, a ratio of an area of ​​the strong salience information region to an area of ​​the weak salience information region. The production support system according to claim 1 .

4. A production support system that supports the production of visually recognized works, an image acquisition unit that acquires an image of the work; an impression estimation unit that estimates the degree to which information included in the image will impress a person who views the image; a display unit that displays the degree to which the information is impressive; Equipped with the impression estimation unit calculates information saliency indicating a degree of saliency of information given to a person in each region of the image; the display unit displays a saliency map generated by superimposing a saliency grayscale image, in which the higher the information saliency, the lighter the color, on the image of the product; A production support system in which the saliency grayscale image is an image obtained by combining an edge image of the work and a change amount image that represents a difference value relative to the image value that occupies the largest area within the work.

5. the impression estimation unit further identifies impression words that indicate the impression that the image gives to people based on the colors included in the image; the display unit displays the impression words.

5. The production support system according to claim 1.

6. the impression estimation unit assigns one or more words or phrases representing the impression given to human vision to each hue and each tone of the PCCS in advance as impression assignment information, and identifies words or phrases representing the impression given to humans by the production based on the hue and tone of each region used in the production and the impression assignment information; 5. The production support system according to claim 1.

7. the margin analysis unit generates an edge image of the production, scans a margin determination area of ​​a predetermined size across the entire area of ​​the generated edge image, and determines whether the margin determination area is included in the margin area or the information area based on the area ratio of the edge portion included in the margin determination area. The production support system according to claim 1 .

8. A production support method using a production support system that supports the production of visually recognized works, The production support system includes: an image acquisition step of acquiring an image of the work; an impression estimation step of estimating the degree to which information included in the image makes an impression on a person who views the image; a display step of displaying the degree to which the information is impressive; The impression estimation step includes: a gaze evaluation step of calculating information saliency indicating the degree of saliency of information given to human vision in each region of the image; a margin analysis step of identifying information areas containing information and margin areas not containing information in the image based on the information saliency; an index calculation step of calculating, based on the information saliency, the information area, and the blank area, an information intensity indicating how efficiently information is placed in the information area of ​​the product, and an information area ratio indicating how much areas to which the gaze is directed are dispersed in the image; A production support method for carrying out the above.

9. A production support method using a production support system that supports the production of visually recognized works, The production support system includes: an image acquisition step of acquiring an image of the work; an impression estimation step of estimating the degree to which information included in the image makes an impression on a person who views the image; a display step of displaying the degree to which the information is impressive; the impression estimation step calculates information saliency indicating a degree of saliency of information given to a person in each region of the image; The display step displays a saliency map generated by superimposing a saliency grayscale image, in which the higher the information saliency, the lighter the color, on the image of the product; A production support method in which the saliency grayscale image is an image obtained by combining an edge image of the production and a change amount image that represents a difference value relative to the image value that occupies the largest area within the production.

10. an image acquisition unit provided in the computer acquires an image of the work that is visually recognized; causing a gaze evaluation unit included in the impression estimation unit of the computer to calculate information saliency indicating a degree of saliency of information given to human vision in each region of the image; causing a margin analysis unit included in the impression estimation unit to distinguish, based on the information saliency, an information region containing information and a margin region not containing information in the image; an index calculation unit included in the impression estimation unit is caused to calculate, based on the information saliency, the information area, and the blank area, an information intensity indicating how efficiently information is placed in the information area of ​​the product, and an information area ratio indicating how much areas to which gazes are directed are dispersed in the image; causing the impression estimation unit to estimate a degree to which information included in the image will impress a person who views the image; A production support program that displays the degree to which the information is impressive on a display unit provided in the computer.

11. an image acquisition unit provided in the computer acquires an image of the work that is visually recognized; an impression estimation unit included in the computer calculates information saliency indicating the degree of saliency of information given to a person in each area of ​​the image, and estimates the degree to which information included in the image will impress a person who views the image; a display unit provided in the computer displays a saliency map generated by superimposing a saliency grayscale image, in which the higher the information saliency, the lighter the color; and A production support program, wherein the saliency grayscale image is an image obtained by combining an edge image of the work and a change amount image that represents a difference value relative to the image value that occupies the largest area within the work.

Images