Preference information collection device and preference information collection method

Abstract

To more accurately grasp taste information for which many users have the highest preference.SOLUTION: A taste information collection device 10 comprises: an image display unit 11 on which images 31A to 31D are displayed in a plurality of predefined areas 30A to 30D, respectively; an input unit 13 which inputs a selection operation for a selected image which a user selects from the plurality of images 31A to 31D displayed on the image display unit 11; a line-of-sight detection unit 12 which detects user's line of sight when the user gazes at the image display unit 11; a gaze area determination unit 14 which determines an area where user's line of sight existed for the longest time during a prescribed period of time, out of the areas 30A to 30D on the image display unit 11 and outputs the image in this area as an image of interest; a determination unit 16 which determines whether or not the selected image and the image of interest are identical; and a counting unit 17 which counts the selected image as a representative image of user's interest if it is determined that the selected image and the image of interest are identical.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a preference information collecting device and a preference information collecting method. [Background technology]

[0002] Conventionally, there is known a technology for extracting characteristic user behavior and collecting user preference information by displaying content-related lists and icons on a screen and detecting the user's selection status regarding images including these lists and icons (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2008-191702 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, while conventional technologies can collect preference information selected based on the results of a user's thinking, they cannot obtain preference information that a user intuitively selects, making it difficult to accurately grasp the preference information that most users find most desirable.

[0005] The present invention has been made in view of the above, and has an object to more accurately grasp preference information that most users find most preferable. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the object, the preference information collection device of the present invention comprises an image display unit that displays images in each of a plurality of predetermined areas, an input unit that inputs a selection operation for a selected image selected by the user from the plurality of images displayed on the image display unit, a gaze detection unit that detects the user's gaze when the user gazes at the image display unit, a gaze area determination unit that determines which area of ​​the image display unit the user's gaze was on for the longest period of time and outputs the image of this area as an image of interest, a determination unit that determines whether the selected image and the image of interest match, and a counting unit that counts the selected image as a representative image of interest to the user when it is determined that the selected image and the image of interest match.

[0007] In addition, the preference information collecting method according to the present invention includes: A preference information collecting method executed by a preference information collecting device, comprising: The method includes the steps of: displaying images in a plurality of predetermined areas; inputting a selection operation for a selected image selected by the user from the plurality of displayed images; detecting the user's gaze when the user gazes at the image display unit; determining which area among the areas in which images are displayed the user's gaze was on for the longest period of time and outputting the image of this area as a featured image; determining whether the selected image and the featured image match; and, if it is determined that the selected image and the featured image match, counting the selected image as a representative image of interest to the user. Effect of the Invention

[0008] According to the present invention, it is possible to more accurately grasp preference information that most users find most preferable. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a block diagram showing an example of the configuration of a preference information collecting device according to this embodiment. [Diagram 2] FIG. 2 is a flowchart showing the flow of processing for investigating user preferences in the preference information collecting device according to this embodiment. [Diagram 3]FIG. 3 is a diagram for explaining feature points common to a selected image and an image of interest. [Figure 4] FIG. 4 is a flowchart showing the flow of processing for identifying an image that many users find most favorable in the preference information collecting device according to this embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of a preference information collecting device and a preference information collecting method according to the present invention will be described in detail below with reference to the accompanying drawings. Note that the present invention is not limited to the following embodiments.

[0011] 1 is a block diagram showing an example of the configuration of a preference information collecting device according to the present embodiment. The preference information collecting device 10 investigates the user's preferred images (preferences) from an image (selected image) that the user actually selected from a plurality of comparison images displayed on a screen and an image (attention image) that the user focused on due to the movement of the user's line of sight, and more accurately grasps the preference information that most users find most preferable.

[0012] The preference information collecting device 10 is an information processing device such as a personal computer, and includes an arithmetic processing device (control device) configured with a CPU (Central Processing Unit) and the like. The preference information collecting device 10 loads a stored program into a memory and executes instructions included in the program. The preference information collecting device 10 includes an internal memory (not shown), which is used for temporary storage of data such as the program in the preference information collecting device 10. As shown in FIG. 1, the preference information collecting device 10 includes an image display unit 11, a gaze detection unit 12, an input unit 13, a gaze area determination unit 14, a user information unit 15, a determination unit 16, a counting unit 17, an analysis unit 18, an image control unit 19, and an image storage unit 20.

[0013] The image display unit 11 is a display device provided in the preference information collecting device 10. The image display unit 11 is, for example, a display including a liquid crystal display (LCD) or an organic EL (Electro-Luminescence) display. The image display unit 11 displays a plurality of comparison images input via the image control unit 19 in each of the predetermined regions. In the example of FIG. 1, the image display unit 11 is partitioned into four regions 30A to 30D, and comparison images (images) 31A to 31D of a predetermined product (e.g., a passenger car) are displayed in each of the regions 30A to 30D. The comparison images 31A to 31D are still images, but may be moving images. The comparison images 31A to 31D show an image of the product, and each includes a feature point such as the color or shape of the whole or part of the product. The feature point is a part that the user is interested in and pays attention to when selecting the comparison images 31A to 31D. The feature point will be described later. FIG. 1 shows a passenger car as an example of a product, but the product is not limited to this as long as it expresses the characteristic points of the shape and color.

[0014] The gaze detection unit 12 is disposed adjacent to the image display unit 11, and detects the user's gaze when the user gazes at the image display unit 11. The gaze detection unit 12 has multiple (for example, two) cameras 12A, and detects the direction (movement) of the user's gaze by photographing the user's eyes with the cameras 12A. Information related to the detected gaze direction is output to the gaze area determination unit 14.

[0015] The input unit 13 is an input device for inputting various information to the preference information collecting device 10. For example, a keyboard or a mouse can be used as the input unit 13. In this embodiment, a button switch (not shown) is formed superimposed on each of the areas 30A to 30D provided on the image display unit 11. Then, the user uses the mouse to input an image that the user thinks is most preferred from the comparison images 31A to 31D, and the image (e.g., image 31B) is selected as a selected image. Information related to the selected image is output to the determination unit 16.

[0016] The gaze area determination unit 14 measures the time during which the gaze of the user was present in each of the regions 30A to 30D of the image display unit 11 based on the gaze of the user detected by the gaze detection unit 12. Then, the gaze area determination unit 14 determines the region (e.g., region 30C) in which the gaze of the user was present for the longest time among the regions 30A to 30D of the image display unit 11. The gaze area determination unit 14 acquires the image 31C displayed in the determined region 30C as an image of interest, and information regarding the image of interest is output to the determination unit 16. The gaze area determination unit 14 also outputs information regarding the time during which the gaze was present in each of the regions 30A to 30D to the analysis unit 18. Here, the predetermined time refers to, for example, the time from the time when the comparison images 31A to 31D are displayed in the regions 30A to 30D of the image display unit 11 to the time when the user performs an input operation of the selected image. The region where the user's gaze was present the longest refers to the region where the user gazed the longest in total within the above-mentioned predetermined time, and may be a discontinuous time such as alternating between images of two regions. Furthermore, the presence of the user's gaze in a region refers to the user's gaze being continuously observed in the region for a certain period of time (e.g., 0.1 seconds) or more. This makes it possible to eliminate the situation where the region just happened to catch the user's eye regardless of the user's intention, and to accurately determine the region (image) where the user gazed.

[0017] The user information unit 15 acquires information (attribute information) on the attributes of the user, such as the user's gender, age, and preferences (areas of interest). The information on the user's attributes is input, for example, via the input unit 13 when performing a series of operations to select a selected image from the above-mentioned multiple comparison images. The acquired user attribute information is output to the analysis unit 18.

[0018] The determination unit 16 determines whether or not the selected image selected via the input unit 13 matches the attention image output by the gaze area determination unit 14. The determination unit 16 outputs information on the determination result to the counting unit 17.

[0019] The counting unit 17 counts (counts) images based on the input judgment result. For example, if it is judged that the selected image and the image of interest match, the counting unit 17 counts the selected image as a representative image of interest to the user. If it is judged that the selected image and the image of interest do not match, the counting unit 17 counts the images of interest that do not match the selected image as sensory images, and counts the images of selected images that do not match the image of interest as logical images. The counting unit 17 also counts the number of representative images, sensory images, and logical images for each user, and outputs the counting result to the analysis unit 18. Here, the representative image is selected from a plurality of comparison images as an image of interest to the user, and the representative image selected most frequently by a plurality of users is treated as a recommended image, which will be described later.

[0020] The analysis unit 18 generates statistical information according to the user attributes based on information on the counted number of each image including the representative image, the sensory image, and the logical image counted by the counting unit 17 and information on the user attributes acquired from the user information unit 15. This statistical information includes at least various representative images in the user preference survey, the management numbers of these representative images, the number of representative images, the number of logical images, and the time during which the gaze was present in each of the areas 30A to 30D. The generated statistical information is output to the image control unit 19. Furthermore, the analysis unit 18 determines the representative image with the highest count number as the recommended image based on the statistical information of multiple users. This recommended image indicates an image (preference information) that many users find most preferable. In this embodiment, an image selected from multiple comparison images as an image that the user is interested in is determined to be the representative image, so that the representative image selected most by multiple users within a range categorized by a predetermined user attribute can be understood as the recommended image that these users find most preferable.

[0021] The image control unit 19 controls the display of a plurality of comparative images of the target product in the predetermined areas 30A to 30D of the image display unit 11. These comparative images are set in advance for each product. A new set of comparative images may be set depending on the state of image selection by the user. Specifically, when it is determined that the selected image and the target image do not match, the image control unit 19 extracts feature points that are set in advance for the selected image and the target image, selects new comparative images having matching feature points, and displays these newly selected images on the image display unit 11, replacing the images displayed on the image display unit 11.

[0022] The image storage unit 20 stores a plurality of comparative images for the target product, and outputs an image selected under the control of the image control unit 19. The image storage unit 20 also stores statistical information according to user attributes under the control of the image control unit 19.

[0023] Next, a description will be given of the flow of processing performed by the preference information collecting device 10. Fig. 2 is a flowchart showing the flow of processing performed by the preference information collecting device 10 according to the present embodiment to investigate a user's preferences. Fig. 3 is a diagram for explaining feature points common to a selected image and an image of interest.

[0024] The preference information collecting device 10 displays comparative images that are the subject of the preference survey in each area of ​​the image display unit 11 (step S1). Specifically, the preference information collecting device 10 reads out a plurality of comparative images related to the target product from the image storage unit 20 by the image control unit 19, and displays these comparative images in each area of ​​the image display unit 11.

[0025] Next, the preference information collecting device 10 inputs the selected image selected by the user (step S2). Specifically, the preference information collecting device 10 selects an image that the user finds most preferable from among the plurality of comparison images as the selected image based on the user's operation via the input unit 13, and inputs this selected image. Information regarding this selected image is output to the determination unit 16.

[0026] Next, the preference information collecting device 10 acquires user attribute information (step S3). The user information unit 15 of the preference information collecting device 10 acquires user attribute information such as the user's gender, age, and preferences (hobbies or areas of interest) based on the user's operation via the input unit 13. In the example of Fig. 2, the user attribute information is acquired after the selected image is input, but the timing of acquiring the user attribute information may be before the selected image is input.

[0027] Next, the preference information collecting device 10 acquires the user's attention image based on the gaze detection (step S4). Specifically, the preference information collecting device 10 uses the gaze area determining unit 14 to determine the area (e.g., area 30C) where the user's gaze was the longest during a predetermined time among the areas 30A to 30D of the image display unit 11 based on the user's gaze. Then, the image 31C displayed in the determined area 30C is acquired as the attention image. In this case, the predetermined time is the time from when the comparison images 31A to 31D are displayed in the areas 30A to 30D of the image display unit 11 to when the user performs an input operation of the selected image. In addition, information related to the acquired attention image is output to the determining unit 16.

[0028] Next, the preference information collecting device 10 judges whether or not the selected image matches the attention image (step S5). Specifically, the preference information collecting device 10 judges whether or not the selected image selected by the user matches the attention image that the user has paid attention to for the longest time, based on the information input to the judgment unit 16. If it is judged that the selected image matches the attention image (step S5; Yes), the preference information collecting device 10 proceeds to step S6.

[0029] On the other hand, if the selected image does not match the image of interest (step S5; No), the preference information collecting device 10 counts the selected images that do not match as logical images (step S9). The preference information collecting device 10 counts the number of logical images by the counting unit 17. Here, a logical image refers to an image among the selected images that does not match the image of interest. Also, an image among the image of interest that does not match the selected image may be counted as a sensory image.

[0030] Next, the preference information collecting device 10 extracts matching feature points between the selected image and the image of interest (step S10). Specifically, in FIG. 3, the image 31B displayed in the area 30B is the selected image, and the image 31C displayed in the area 30C is the image of interest. In this embodiment, the images 31B and 31C have the feature points of the product (passenger car) set in advance. For example, for a passenger car, the overall shape of the vehicle body, the color of the vehicle body, the shape of the headlights, the shape of the rear lamps, the shape of the doors, the shape of the windows (side windows), the shape of the tire wheels, and the like are set as the feature points. These feature points are only examples, and feature points may be set in other locations, or the number of feature points may be increased or decreased. The preference information collecting device 10 extracts matching feature points between the selected image and the image of interest by the image control unit 19. The image control unit 19 performs image analysis of the selected image and the image of interest to extract the window (side window) shapes 32B and 32C as matching feature points.

[0031] Next, the preference information collecting device 10 selects an image having a feature point that matches the extracted feature point (step S11). That is, the preference information collecting device 10 selects two images having windows (matching feature points) of the same shape as the window shapes 32B and 32C that are feature points of the selected image and the target image from the images related to the product (passenger car) stored in the image storage unit 20 by the image control unit 19. Then, the preference information collecting device 10 displays a new comparison image by adding the selected image, the target image, and the two selected images, replacing the image displayed on the image display unit 11, and returns the process to step S1 to repeat the process. Note that the preference information collecting device 10 may select four images having feature points that match the extracted feature points, and use these four selected images as new comparison images.

[0032] In the judgment of step S5 described above, if the selected image and the image of interest match (step S5; Yes), the preference information collecting device 10 counts the matching selected image as a representative image that is an image of interest to the user (step S6). The representative image is the image in which the user is most interested among the multiple (four) comparison images. In this embodiment, if the selected image actually selected by the user matches the image of interest that the user has focused on for the longest time, the selected image is set as the representative image, so that it is possible to eliminate cases where the selected image is selected by mistake, and it is possible to accurately count the representative images of interest to the user.

[0033] Next, the preference information collecting device 10 judges whether there is an additional survey (step S7). Here, the additional survey is conducted by using another comparative image of the same product (passenger car) or a comparative image of another product (such as a motorcycle) in the above-mentioned preference survey. If there is an additional survey (step S7; Yes), the process returns to step S1 and the process is repeated.

[0034] If there is no additional survey (step S7; No), the preference information collecting device 10 outputs statistical information according to the user attributes (step S8) and ends the process. The preference information collecting device 10 uses the analysis unit 18 to associate the user attribute information with the survey results of the user, generates statistical information according to the user attributes, and outputs this statistical information to the image control unit 19. The image control unit 19 stores this statistical information in the image storage unit 20 and ends the process. This statistical information includes at least various representative images in the user preference survey, the management numbers of the representative images, the number of representative images, the number of logical images, and the time that the gaze was present in each of the areas 30A to 30D.

[0035] Next, a process flow for grasping an image (preference information) that many users find most favorable from statistical information of a plurality of users will be described. Fig. 4 is a flowchart showing a process flow for grasping an image that many users find most favorable in the preference information collecting device according to this embodiment.

[0036] First, the preference information collecting device 10 inputs statistical information of each user (step S21). Specifically, the image control unit 19 reads out the statistical information stored in the image storage unit 20 for each user, and inputs the statistical information of each user to the analysis unit 18.

[0037] Next, the preference information collecting device 10 judges whether or not there is a user whose number of logical images exceeds a predetermined threshold based on the input statistical information of each user (step S22). In other words, it judges whether or not the number of logical images counted for the same user exceeds a predetermined threshold. This predetermined threshold is set appropriately, but is preferably set to, for example, 1 / 3 to 1 / 2 of the number of a series of surveys of users. In this judgment, if the analysis unit 18 of the preference information collecting device 10 judges that there is no user whose number of logical images exceeds the predetermined threshold (step S22; No), the process proceeds to step S25.

[0038] On the other hand, when the preference information collecting device 10 determines that there is a user whose number of logical images exceeds a predetermined threshold (step S22; Yes), the analysis unit 18 determines whether or not there is an area where the gaze of the user is not detected until the generation of statistical information about the same user is completed (step S23). The period until the generation of statistical information about the same user is completed refers to the period until all predetermined times corresponding to all images of interest output about the same user have elapsed. In this case, the analysis unit 18 determines whether or not there is an area where the gaze of the user is not detected based on the time during which the gaze was present in each of the areas 30A to 30D. In this determination, when the preference information collecting device 10 determines that there is no area where the gaze of the user is not detected until the generation of statistical information about the same user is completed (step S23; No), the process proceeds to step S25.

[0039] On the other hand, when the preference information collecting device 10 determines that there is an area where the gaze of the user is not detected until the generation of the statistical information for the same user is completed (step S23; Yes), it stops using the statistical information of the user (step S24). That is, when the number of logical images counted for the same user exceeds a predetermined threshold and there is an area where the gaze of the user is not detected until all the predetermined times corresponding to all the attention images output for the same user have elapsed, it can be determined that the user is performing these preference surveys carelessly (fraudulently). Since this type of user's statistical information does not accurately reflect the user's preferences, if this user's statistical information is used to grasp the recommended images that multiple users find most preferable, it may become noise instead. For this reason, in this embodiment, by stopping the use of the statistical information of a user who is thought to be performing the preference survey incorrectly, it is possible to accurately grasp the preference information of multiple users.

[0040] Next, the preference information collecting device 10 collects representative images for each user (step S25), and outputs the most numerous representative image corresponding to the user attribute as the recommended image (step S26). As described above, the representative image is the image in which the user is most interested among the comparison images. The preference information collecting device 10 collects the representative images of many users by user attribute (gender and age) using the analysis unit 18, and can accurately analyze (understand) the recommended image, which is the image in which many users are most interested in the categorized user attributes. The analysis unit 18 of the preference information collecting device 10 can output the recommended image information corresponding to the user attribute to, for example, an external management device (not shown), and can propose it as a design image of the optimal product (goods) in which many users are most interested.

[0041] As described above, the preference information collecting device 10 according to the present embodiment includes an image display unit 11 that displays a plurality of images 31A-31D in each of the predetermined regions 30A-30D, an input unit 13 that inputs a selection operation for a selected image selected by a user from the plurality of images 31A-31D displayed on the image display unit 11, a gaze detection unit 12 that detects the gaze of the user when the user gazes at the image display unit 11, a gaze area determination unit 14 that determines the area in which the user's gaze was present the longest during a predetermined time among the regions 30A-30D of the image display unit 11 and outputs an image of this area as an attention image, a determination unit 16 that determines whether the selected image and the attention image match, and a counting unit 17 that counts the selected image as a representative image in which the user is interested if it is determined that the selected image and the attention image match. Therefore, when the selected image actually selected by the user matches the attention image that the user has paid attention to the longest, the selected image is set as the representative image, thereby eliminating cases in which the selected image is selected by mistake, and the representative images in which the user is interested can be accurately counted. Therefore, it is possible to more accurately grasp the images that most users find most desirable (preference information).

[0042] In addition, images 31A to 31D are each set with a number of feature points, which are characteristic shapes, in advance, and if it is determined that the selected image does not match the image of interest, an image control unit 19 is provided which compares the feature points of the selected image with those of the image of interest to extract matching feature points, and selects a new image having feature points similar to the extracted feature points and displays it in place of the image displayed on image display unit 11.As a result, it is possible to provide the user with an image having the feature points that the user has focused on, and it is possible to accurately tally up representative images that interest the user.

[0043] The tallying unit 17 also counts the representative image for each user. The analysis unit 18 also determines the most numerous representative image among the representative images of the multiple users as the recommended image that the multiple users find most favorable, so that the recommended image that the multiple users are most interested in can be accurately grasped for the categorized user attributes, and the recommended image can be proposed as the design image of the optimal product that the multiple users are most interested in.

[0044] Furthermore, the counting unit 17 counts as logical images those images among the selected images for each user that are determined not to match the image of interest, and if the number of logical images counted for the same user exceeds a threshold value and there is an area among multiple areas in which the user's gaze is not detected before all of the specified times corresponding to all of the images of interest output for the same user have elapsed, the analysis unit 18 does not use the statistical information (representative image) of that user to determine the recommended image. This makes it possible to eliminate information that does not accurately reflect the user's preferences, and to accurately determine the preference information of many users.

[0045] Although the preference information collecting device 10 according to the present embodiment has been described above, it may be implemented in various different forms other than the above-described embodiment. Each component of the illustrated preference information collecting device 10 is a functional concept, and does not necessarily have to be physically configured as illustrated. In other words, the specific form of each device is not limited to that illustrated, and all or part of the devices may be functionally or physically distributed or integrated in any unit depending on the processing load and usage status of each device.

[0046] The configuration of the preference information collecting device 10 is realized, for example, as software by a program loaded into a memory. In the above embodiment, the configuration has been described as a functional block realized by cooperation of these hardware and software. In other words, these functional blocks can be realized in various forms by only hardware, only software, or a combination of both.

[0047] The above-mentioned components include those that a person skilled in the art can easily imagine and those that are substantially the same. Furthermore, the above-mentioned configurations can be appropriately combined. Furthermore, various omissions, substitutions, or modifications of the configurations are possible without departing from the scope of the present invention. [Explanation of symbols]

[0048] 10. Preference information collection device 11 Image display section 12 Gaze detection unit 12A Camera 13 Input section 14 Gaze area determination unit 15 User Information Section 16 Judgment section 17 Counting Department 18 Analysis Department 19 Image control section 20 Image storage section 30A, 30B, 30C, 30D area 31A, 31B, 31C, 31D Images 32B, 32C Window shape (feature points)

Claims

1. an image display unit that displays an image in each of a plurality of predetermined regions; an input unit for inputting a selection operation for a selected image selected by a user from the plurality of images displayed on the image display unit; a gaze detection unit that detects a gaze of the user when the user gazes at the image display unit; a gaze area determination unit that determines an area on the image display unit where the user's gaze is the longest for a predetermined period of time, and outputs an image of this area as an image of interest; a determination unit that determines whether the selected image and the image of interest match; a counting unit that, when it is determined that the selected image matches the image of interest, counts the selected image as a representative image in which the user is interested; A preference information collecting device comprising:

2. A plurality of feature points each having a characteristic shape is set in advance in the image, If it is determined that the selected image does not match the image of interest, The preference information collecting device according to claim 1, further comprising an image control unit that compares feature points of the selected image with feature points of the target image to extract matching feature points, selects a new image having feature points similar to the extracted feature points, and displays the image in place of the image displayed on the image display unit.

3. The counting unit counts the representative images for each user, The preference information gathering device according to claim 1 , further comprising an analysis unit that sets the most numerous representative image among the representative images of the plurality of users as a recommended image that the plurality of users find most favorable.

4. the counting unit counts, for each of the users, images among the selected images that are determined not to match the image of interest as logical images; The preference information collecting device of claim 3, wherein the analysis unit does not use the representative image of the user to determine the recommended images when there is an area among the multiple areas in which the user's gaze is not detected before the number of logical images counted for the same user exceeds a threshold value and all of the specified times corresponding to all of the target images output for the same user have elapsed.

5. A preference information collecting method executed by a preference information collecting device, comprising: Displaying an image in each of a plurality of predetermined regions; inputting a selection operation for a selected image selected by a user from the displayed plurality of images; detecting a line of sight of the user when the user gazes at the image display unit; determining an area where the user's gaze was the longest within a predetermined period of time among the areas where the image is displayed, and outputting the image of this area as an image of interest; determining whether the selected image and the image of interest match; counting the selected image as a representative image of interest to the user when it is determined that the selected image and the image of interest match; The method for collecting preference information comprises:

Images