Information processing device, method for controlling the information processing device, and program
A dual-check system using an image analysis program for pixel-level measurements and generative AI for layout and regulatory compliance enhances the accuracy of creative checks, addressing the limitations of existing generative AI systems.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2025-10-30
- Publication Date
- 2026-03-27
AI Technical Summary
Existing creative checks using generative AI lack accuracy in objective measurements and layout assessments, particularly at the pixel level and in compliance with regulations.
A dual-check system utilizing an image analysis program for initial checks based on pixel-level measurements and a generative AI for layout and regulatory compliance, integrating results for enhanced accuracy.
Improves the accuracy of creative checks by combining pixel-level measurements with contextual layout assessments, ensuring compliance with brand guidelines and regulations.
Smart Images

Figure 0007836935000001_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to an information processing apparatus, a control method for the information processing apparatus, and a program.
Background Art
[0002] Generally, when a company uses its own logo in advertisements, websites, SNS, or printed materials, etc., a check is made as to whether the logo is being used correctly in accordance with the company's brand image or guidelines. Hereinafter, the above check will be referred to as a creative check. The creative check is mainly performed for the purpose of achieving unity of the brand image and preventing trademark infringement. Patent Document 1 discloses a technique in which a generative AI performs a series of processes for the creative check.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] There is a need for a technique for improving the accuracy of creative checks using generative AI.
[0005] Therefore, an object of the present disclosure is to improve the accuracy of creative checks using generative AI.
Means for Solving the Problems
[0006] A program according to one aspect of the present disclosure is a program for executing a check of an image including a logo, which operates in an information processing apparatus, the information processing apparatus including an acquisition unit that acquires the image, and the image This process includes determining, through measurement, the area occupied by the logo and the area occupied by the text. Perform the first check , without using generative AIA first checking means that controls the image analysis program to perform the check, and the logo included in the image that is generated by the first check. The area occupied by and text occupies Information that identifies a region Intermediate data including the results of the first check, and Using the generated AI This includes checking the layout of the aforementioned image. The system is characterized by functioning as a second checking means that controls the execution of a second check, and a notification means that provides notification indicating the results of the first and second checks. [Effects of the Invention]
[0007] According to this disclosure, the accuracy of creative checks using generative AI can be improved. [Brief explanation of the drawing]
[0008] [Figure 1] This diagram illustrates the creative review process. [Figure 2] This is a hardware configuration diagram of an information processing device. [Figure 3] This is a software configuration diagram of an information processing device. [Figure 4] This is a sequence diagram illustrating the flow of creative review. [Figure 5] This diagram shows the screen for selecting the images to be checked. [Figure 6] This diagram illustrates the creative review process. [Figure 7] This diagram shows the processes performed by the image analysis program. [Figure 8] This figure shows the intermediate data output by the image analysis program. [Figure 9] This diagram shows the processes performed by the generative AI. [Figure 10] This figure shows the results of the secondary check performed by the generating AI. [Figure 11] This is a flowchart showing the creative check process. [Figure 12] This figure shows an example of guidelines.
Best Mode for Carrying Out the Invention
[0009] <Embodiment 1> Hereinafter, preferred embodiments of an information processing apparatus, an information processing method, and a program according to the present disclosure will be described in detail with reference to the drawings. Note that the information processing apparatus, the information processing method, and the program according to the present disclosure are not limited by this embodiment. Also, in the following embodiments, the same components are denoted by the same reference numerals, and redundant descriptions are omitted.
[0010] Generally, it is known that by performing creative checks on media including logos (e.g., corporate logos) using generative AI, the efficiency can be significantly improved compared to manual work, and human errors can be suppressed.
[0011] On the other hand, when the generative AI performs a series of processes for creative checks, there is a risk that precise checks cannot be performed. For example, the generative AI is known to be good at checking the layout in the image to be checked, but may not be suitable for accurate calculations at the pixel level or objective checks based on image processing. Therefore, there is still room for improvement in creative checks using generative AI.
[0012] In view of the above, in this embodiment, a form of performing creative checks using a generative AI and an image analysis program in the creative checks of images will be described. In this embodiment, the image analysis program is for extracting specific information and features from an image by digitally processing the image. Thereby, it measures and quantifies the position information in the image in pixel units, and executes checks that can be objectively performed using the numerical values.
[0013] FIG. 1 is a schematic diagram showing the flow of creative check in the present embodiment. In FIG. 1, the user can execute a check on an image to be creatively checked by using the creative check tool 301 provided in the information processing apparatus 101. Specifically, the creative check tool 301 provided in the information processing apparatus 101 requests a creative check from each of the first server apparatus 102 that performs image analysis and the second server apparatus 103 that includes a generation AI via the Internet.
[0014] In S1, the creative check tool 301 provided in the information processing apparatus 101 transmits an image to be checked to the first server apparatus 102. The image to be checked is assumed to be any one of any images stored in the information processing apparatus 101. Note that the information processing apparatus 101 may obtain an image from another device or the like and make a check request.
[0015] In S2, the first server device 102 performs a primary check using an image analysis program. In the primary check, the first server device 102 processes the image to be checked. That is, it measures the positional information of logos and text in the image in pixels. Furthermore, the first server device 102 performs checks that can be performed using the measured values. Examples include checking the color variations of the logo or checking the clear space of the logo. Details of the primary check process will be described later. Also, in the primary check, in preparation for the generation AI in the second server device 103 to perform OCR (Optical Character Recognition) on the image later, the text areas contained in the image are converted into strings. These are then written to a single file along with the results of the primary check. This allows the generation AI in the second server device 103 to accurately grasp the positional relationship of the text in the image to be checked in the secondary check. In this embodiment, since it is assumed that the image analysis program is better suited to image measurement than the generation AI, the image analysis program is responsible for generating the data for the generation AI to perform OCR. The first server device 102 writes the primary check results and the data converted from the image to strings to a single file. This file is also used for creative checks performed by the generation AI in the second server device 103, and is called intermediate data. As mentioned above, the initial check involves an objective assessment of the logo included in the image being checked. Specifically, the image analysis program identifies the shape, color, and position of the logo, and the check is performed taking this information into account. The information about the logo identified here is also used in the secondary check by the generating AI, and is therefore written into the intermediate data as part of the initial check results. Details of the intermediate data will be described later using Figure 8.
[0016] In S3, the first server device 102 sends intermediate data, including the results of the initial check, to the information processing device 101. That is, the information processing device 101 receives the intermediate data from the first server device 102. In S4, the information processing device 101 checks the intermediate data. If the result of the initial check performed by the first server device 102 is "NG", the information processing device 101 proceeds to S8 without sending the intermediate data to the second server device (i.e., without requesting a creative check by the generation AI). In S8, the user is notified of the results of the creative check. If the result of the initial check performed by the first server device 102 is "OK" (no NG parts), the information processing device 101 proceeds to S5, sends the intermediate data to the second server device 103, and requests a secondary check.
[0017] In S6, the second server device 103 performs a secondary check using the generation AI. At this time, the generation AI performs OCR on the intermediate data generated by the first server device 102 to determine the positional information of the text area in the image to be checked. It also determines the positional information of the logo based on the logo information contained in the intermediate data. Then, it performs processing that the generation AI excels at in creative checks. Processing that the generation AI excels at includes, for example, checks related to layout or checks related to the Premiums and Representations Act. Details of the secondary check performed by the generation AI will be described later.
[0018] In S7, the second server device 103 transmits the results of the secondary check to the information processing device 101. That is, the information processing device 101 obtains the results of the secondary check from the second server device 103. In S8, the information processing device 101 notifies the display unit 210 of the creative check results. The notification content may be, for example, the result of integrating the primary and secondary checks, or it may be the result of the secondary check only. Methods of notification include, for example, notification by email, notification within the display screen of the creative check tool 301, etc. The above is a general overview of the creative check in this embodiment.
[0019] First, the devices of the information processing system of this embodiment will be described using Figure 1. The information processing system of this embodiment includes an information processing device 101, a first server device 102, and a second server device. These devices are connected to each other via a network, either by wired or wireless means, enabling communication between them. The network is, for example, a LAN (Local Area Network) and a WAN (Wide Area Network) such as the Internet.
[0020] The information processing device 101 incorporates an OS (Operating System), and each component within the information processing device 101 is managed by the OS. Furthermore, in this embodiment, the information processing device 101 can run a creative check tool. The hardware configuration of the information processing device 101 and the software configuration including the creative check tool will be described later.
[0021] The first server device 102 is a device capable of implementing an image analysis program. In this embodiment, the image analysis program is used to extract specific information or features from an image by digitally processing that image. By using the image analysis program, it is possible to efficiently analyze minute changes that cannot be perceived by the human eye, or large amounts of image data. In this embodiment, the first server device 102 is described as having an image analysis program, but in reality, this is not the only option, and for example, the information processing device 101 may have an image analysis program.
[0022] The second server device 103 is equipped with a generation AI specifically designed for performing creative checks on images. In this embodiment, a secondary check is performed by the generation AI based on intermediate data acquired from the information processing device 101 via the network and predetermined prompts. In this embodiment, there is no particular limitation on whether or not the generation AI undergoes retraining, which is a process of additional learning to adapt to new information and data.
[0023] Figure 2 is a block diagram showing an example of the hardware configuration of the information processing device 101. The information processing device 101 includes a CPU 201, ROM 202, RAM 203, storage device 204, input / output interface 205, and network interface 206. Each component is connected to the others by a system bus.
[0024] The CPU 201 reads the program stored in the ROM 202 into the RAM 203 and executes the read program to perform various control functions. The CPU 201 may be an MPU (Micro Processing Unit) or an ASIC (Application Specific Integrated Circuit). The CPU 201 may also be an FPGA (Field Programmable Gate Array), etc.
[0025] ROM 202 stores control programs such as the startup program for the information processing device 101. RAM 203 temporarily stores control programs or data when the CPU 201 performs various controls. RAM 203 is used as work memory. Storage device 204 is memory for storing applications, software, and various data. Examples of storage device 204 include HDD (Hard Disk Drive), SSD (Solid State Drive), flash memory, and storage devices. Intermediate data obtained from the first server device 102 and secondary check results obtained from the second server device 103 can be stored in ROM 202 or storage device 204.
[0026] The input / output interface 205 is an interface for transmitting output information to an output device such as a display unit 210. The display unit 210 can be a display or a projector. Furthermore, as shown in Figure 2, the display unit 210 may be located outside the information processing device 101 or located within the information processing device 101.
[0027] The input / output interface 205 is an interface for receiving input information from input devices such as the input unit 211. Examples of input devices 211 include a mouse, keyboard, buttons, or touch panel. Furthermore, the input unit 211 may be located outside the information processing device 101, as shown in Figure 2, or it may be located within the information processing device 101.
[0028] Furthermore, the input / output interface 205 may be connected to the display unit 210 and the input unit 211 wirelessly or via a wired connection. The input / output interface 205 may be configured as an input interface and an output interface. The network interface 206 is an interface for receiving data from other devices via a network and transmitting data generated or stored by the information processing device 101 to other devices via the network.
[0029] Note that the information processing device 101 is not limited to the configuration shown in Figure 2. For example, it may be provided with a camera unit for implementing camera functionality, a positioning unit for implementing GPS functionality, and a microphone unit and speaker unit for implementing call functionality.
[0030] The information processing device 101 may also be equipped with a media interface (not shown). The media interface reads programs and data stored in a storage device (not shown) connected by wire or wireless, and provides them to the CPU 201 via RAM 203. The CPU 201 can read the provided program from the storage device onto RAM 203 via the media interface and execute the read program. The storage device is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase Change Rewritable Disk), or a magneto-optical recording medium such as an MO (Magneto-Optical disk). Alternatively, the storage device may be a tape medium, a magnetic recording medium, or a semiconductor memory.
[0031] The CPU 201 of the information processing device 101 realizes various functions of the information processing device 101 by executing programs read from the RAM 203. The CPU 201 reads and executes these programs from the storage device 204, but as an alternative, these programs may be obtained from other devices via a predetermined communication network. Furthermore, it goes without saying that the information processing device 101 is provided with the necessary configuration to realize the information processing in this embodiment. Note that the block diagram shown in Figure 2 may also be applied to the hardware configuration of the first server device 102 and the second server device 103, in addition to the information processing device 101.
[0032] Figure 3 shows an example of the software configuration of the information processing device 101 in this embodiment. Specifically, it illustrates the applications stored in the ROM 202 or storage device 204 of the information processing device 101, and the program modules of the applications.
[0033] Figure 3(a) shows the application data stored in the storage device 204 of the information processing device 101. As shown in Figure 3(a), the storage device 204 of the information processing device 101 stores the creative check tool 301 (also called tool 301). The creative check tool 301 is a tool that transmits the image or various data to be checked, received from the user, to the first server device 102 and the second server device 103, and controls each device to request a creative check. The creative check tool 301 also plays a role in integrating the check results obtained from each server device and notifying the user.
[0034] Figure 3(b) shows the configuration of the program module of the creative check tool 301. As shown in Figure 3(b), the tool 301 of this embodiment comprises a reception unit 302, a display control unit 303, a transmission / reception unit 304, and a generation unit 305 as program modules.
[0035] The reception unit 302 receives a selection from the user regarding which image data stored in the storage device 204 or ROM 202 of the information processing device 101 to use for the creative check. The display screen for the user to select an image is displayed when the creative check tool 301 is launched (see Figure 5). The display control unit 303 controls the display of various screens in the creative check tool 301 on the display unit 210.
[0036] The transmitting / receiving unit 304 transmits and receives data with the first server device 102 and the second server device 103. Specifically, the transmitting / receiving unit 304 sends images to be checked to the first server device 102 and sends intermediate data or prompts to the second server device. The transmitting / receiving unit 304 also receives intermediate data from the first server device and receives the results of the secondary check from the second server device 103.
[0037] The generation unit 305 compiles the check results from the creative check obtained from the first server device 102 or the second server device 103 and generates a final report for notification to the user. Specifically, it compiles the intermediate data obtained from the first server device 102 and the secondary check results obtained from the second server device 103 and generates a final report for notification to the user. The generation unit 305 may also generate a prompt for the generation AI provided in the second server device 103 to perform the creative check.
[0038] Figure 4 is a sequence diagram of the creative check process in this embodiment. This process is executed by the CPUs of the information processing device 101, the first server device 102, and the second server device 103 loading the program stored in ROM or storage device into RAM. In reality, the CPU of each device issues the processing instructions, but for the sake of clarity, the names of each device are used to indicate the main entity of each process in the following sequence diagrams. This process is started by activating the creative check tool 301 installed on the information processing device 101. In the following descriptions of each process, "S" indicates the step number.
[0039] In S401, the information processing device 101 accepts the selection of an image to be creative checked.
[0040] Figure 5 shows the display screen for selecting an image to be checked, as displayed by the creative check tool 301. On the display screen 501, the user can use a device such as a mouse to select any one image from the group of images stored in the information processing device 101. In this embodiment, image 502 is selected. After selecting an image, the user presses the send button 503, and the creative check tool 301 executes the following process.
[0041] In S402, the information processing device 101 transmits the image to be checked, which was selected in S401, to the first server device 102.
[0042] Figure 6 is a diagram showing an overview of the processing performed by each device on images subject to creative checks in this embodiment. Images selected by the information processing device 101 are transmitted to the first server device 102 in S402. The first server device 102 uses an image analysis program to perform mainly the following processes: "identification of logo objects," "initial check based on guidelines or regulations," and "extraction of text areas." After processing in the first server device 102, the second server device 103 performs a secondary check using the intermediate data output by the first server device 102. In the secondary check, the generation AI is used to perform mainly "checks related to layout" and "checks related to the Premiums and Representations Act (checks to automatically detect the presence or absence of misleading representations under the Premiums and Representations Act)." Details of the processing performed by the first server device 102 and the second server device 103 will be described later with reference to Figures 7 and 9. In S403, the first server device performs an initial check of the image using an image analysis program.
[0043] Figure 7 shows the details of the processing performed by the first server device 102. After acquiring the image to be checked, the first server device 102 uses an image analysis program to first perform binarization of the image, as shown in Figure 7(b). Then, as shown in Figure 7(c), the first server device 102 reads a comparison template image stored in the first server device 102 and performs template matching between the image to be checked and the comparison template image. Template matching is an image processing technique that searches for areas in the target image that have a pattern similar to the template image. Similar areas are identified by calculating the similarity to the template image while gradually shifting the template image (one pixel at a time) on a larger image. At this time, the similarity (confidence level) may be displayed as a percentage, for example. In this embodiment, since a creative check of a company logo is assumed, the template image here refers to the company logo. In other words, it identifies which part of the image to be checked contains the company logo, which is the template image. Also, a single company may have multiple company logos. Therefore, for example, multiple template images of company logos used for template matching may be retrieved depending on the company, and template matching may be performed with the target image for each template image. Then, the system can identify which company logo is used in the target image, and display a numerical result indicating how similar the identified shape is to the template image.
[0044] Next, the first server device 102 creates a mask based on the shape and position information of the logo identified within the image to be checked, as shown in Figure 7(e). Then, it checks the color tone defined by the color map for the masked area containing the logo. This allows the color tone used in the logo to be identified in hexadecimal, as shown in Figure 7(f). Next, as shown in Figure 7(g), the first server device 102 checks the identified logo for any violations based on the logo usage terms predetermined by the company. This check, for example, verifies whether the color tone of the logo identified in Figure 7(g) matches the colors defined in the usage terms. It also checks whether the minimum and maximum sizes of the identified company logo are within the range defined in the usage terms, and whether a clear space of the length defined in the usage terms is secured around the logo.
[0045] Next, as shown in Figure 7(h), the first server device 102 extracts the OCR area from the image to be checked, converts it to base-64, and stores it in a file along with the results of the initial check. In this embodiment, in the secondary check, the generating AI of the second server device 103 does not directly read the image to be checked, but rather the image analysis program of the first server device 102 grasps the image information by performing OCR on the base-64 converted character information. This process is performed to avoid risks such as leakage if the image to be checked contains information that has not yet been made public. The first server device 102 outputs a file containing the initial check results and the base-64 converted string of the OCR area as intermediate data. Thus, the first server device 102 performs three main processes: logo identification, objective check of the identified logo (initial check), and extraction of the OCR area.
[0046] Let's return to the explanation of the sequence diagram in Figure 4. In S404, the information processing device 101 acquires intermediate data from the first server device 102. In S405, the information processing device 101 checks the intermediate data. Specifically, it checks the results of the primary check included in the intermediate data.
[0047] Figure 8 shows the intermediate data generated by the first server device 102. In the intermediate data shown in Figure 8, the logo identification result display area 801 displays information indicating the position and size of the logo identified by the image analysis program. The AI generated by the second server device 103 can understand the positional relationship of the logo based on the logo identification result display area 801. In addition, the primary judgment result display area 802 in the intermediate data displays the judgment result of whether the identified logo conforms to the specifications. In this embodiment, it is indicated whether the shape and color of the identified logo are "OK" or "NG" according to the specifications.
[0048] The detailed check result display area 803 displays a description of the items checked based on the information identified by the image analysis program, along with the check results. In Figure 8, as an example, a description of the check results for the identified logo's clear space, minimum size, and background suitability is displayed. The overall judgment result display area 804 displays the overall judgment result, which summarizes the results of each check item performed by the first server device 102. The overall judgment result display area 804 is an item that shows the results of the primary check. If all of the check items are "OK", "OK" is displayed, and if even one of the check items is "NG", "NG" is displayed. In this embodiment, the check items include the logo's shape, logo's color variations, ensuring clear space, the logo's minimum size, and background suitability. In addition to the above, the primary check may also include checks related to logo rotation and deformation. Furthermore, the rules for describing the overall result are not limited to the above. For example, if a predetermined number of check items are "OK", the overall result may be described as "OK".
[0049] Furthermore, in the intermediate data, the OCR extraction area 805 contains information that represents the text area of the image to be checked, extracted by the image analysis program, in base-64 format. Based on this information, the generating AI of the second server device 103 can perform OCR to accurately understand the positional relationships of the text placed in the image to be checked.
[0050] In S405, the information processing device 101 confirms the results of the initial check and determines the subsequent processing based on the content of the initial check. If it determines that the overall judgment result of the initial check (the result displayed in the overall judgment result display area 804) is "NG", the information processing device 101 proceeds to S409. On the other hand, if it determines that the overall judgment result of the initial check is anything other than "NG", it proceeds to S406. In this embodiment, for the purpose of improving work efficiency and reducing costs such as communication expenses, images that are judged as "NG" at the time of the initial check by the first server device are not transmitted to the second server device 103. However, this is not the only option, and there may be a configuration in which the generating AI performs a secondary check even if the overall result of the initial check is "NG".
[0051] In S406, the information processing device 101 transmits intermediate data to the second server device 103 based on the overall judgment result of the primary check being "OK". In S407, the second server device 103 performs a secondary check based on the intermediate data.
[0052] Figure 9 illustrates the secondary check performed in the second server device 103 equipped with a generation AI. First, the generation AI in the second server device 103 reads the text in the image to be checked using OCR based on the OCR extraction area 805 of the acquired intermediate data. Figure 9(b) shows the result of OCR on the OCR extraction area of the intermediate data. By performing OCR, the generation AI can grasp the text and text position information in the image to be checked. In addition to this process, the generation AI also grasps the positional relationship of the logo based on the information described in the logo identification result display area 801 of the intermediate data shown in Figure 8. Subsequently, as shown in Figure 9(c), the generation AI performs a secondary check based on the OCR information. The secondary check mainly involves layout judgment and judgment under the Act against Unjustifiable Premiums and Misleading Representations. In the layout judgment, a process is performed to check the overall validity of the layout based on context, such as whether the logo is used as part of the text (logo in text use) or whether it is placed in an unnatural location in the design. In addition, checks are performed to ensure the independence of the logo, check for prohibited words, or check the regular expression of the text company name. In determining compliance with the Premiums and Representations Act, the system evaluates the risk under the Act based on the entire content of the displayed text. Specifically, it assesses whether, for example, there are any misleading representations or excessive premiums being offered. Furthermore, as shown in Figure 9(c), if the determination is "NG," the generating AI will present the reason for the NG and suggested corrections.
[0053] Returning to the explanation of the sequence diagram in Figure 4, in S408, the information processing device 101 obtains the results of the secondary check from the second server device 103. The results of the secondary check obtained from the second server device 103 are assumed to be a file consisting of text data similar to the intermediate data described above. In S409, the information processing device compiles the results of the primary and secondary checks and notifies the user of the final report.
[0054] Figure 10 shows the final report. The final report is generated by the generation unit 305 of the creative check tool 301 provided in the information processing device 101, based on intermediate data including the primary check results obtained from the first server device 102 and the secondary check results obtained from the second server device 103. As shown in Figure 10, in the final report, the overall judgment result display area 1001 displays the final judgment result based on the results of the primary and secondary checks. In this embodiment, when the judgment result is "OK", it is assumed that there are no NG items in either the primary or secondary check. However, the conditions for the overall result to be "OK" may be changed for each user (or company).
[0055] The primary check result display area 1002 contains the same information as the primary check results performed by the image analysis program of the first server device 102. In the final report, the layout judgment result display area 1003 and the Premiums and Representations Act judgment result display area 1004 contain the results of the secondary check performed by the generation AI of the second server device 103. If these results are "NG", the proposed corrections presented by the generation AI are also included, as shown in Figure 10.
[0056] Furthermore, assuming that the overall result of the primary check for images undergoing a secondary check is "OK," the final report may omit the results of the primary check. Also, for images where the overall result of the primary check is "NG" (for example, if the result of the judgment in S405 in Figure 4 leads to S409), a secondary check will not be performed, and therefore only the results of the primary check will be included in the final report.
[0057] The notification method for the final report shown in Figure 10 may be, for example, displaying it on the display screen within tool 301, generating a file such as a sheet with the check results inserted, or sending the results as an email to any email address.
[0058] Based on the final report, users can modify the image content (for example, correcting the size of the company logo or correcting its use within the text). They can then re-check the modified image using the creative check tool.
[0059] Figure 11 is a flowchart showing the processing performed by the information processing device 101 equipped with the creative check tool 301 in this embodiment. This processing is executed by the CPU 201 of the information processing device 101 loading the program of the creative check tool 301 stored in the ROM 202 or storage device 204 into the RAM 203. This processing is initiated when the creative check tool 301 is started in the information processing device 101.
[0060] In S1101, the CPU 201 of the information processing device 101 selects an image to be subjected to a creative check. As explained above, the image selected in this process is one of the images stored in the ROM 202 or storage device 204 of the information processing device 101, and one is selected by the user's selection operation.
[0061] In S1102, the CPU 201 transmits the image to be checked to the first server device 102. In S1103, the CPU 201 obtains intermediate data from the first server device 102, including the results of the initial check performed in the first server device 102.
[0062] In S1104, the CPU 201 checks the results of the initial check included in the intermediate data and determines whether the check result is "OK" or not. Specifically, the determination is made based on the result displayed in the overall judgment result display area 804 of the intermediate data shown in Figure 8. If the overall judgment result is determined to be "OK", the CPU 201 proceeds to S1105, and if it is determined to be "NG", it proceeds to S1107.
[0063] In S1105, the CPU 201 transmits intermediate data to the second server device 103. In S1106, the CPU 201 obtains the results of the secondary check from the second server device 103. As explained above, the secondary check is performed by the generation AI provided in the second server device 103, and the main check items are the layout determination of the image to be checked and the determination of compliance with the Act against Unjustifiable Premiums and Misleading Representations. In S1107, CPU201 notifies the user of the results of the secondary check. The notification method and content are as described above. After that, CPU201 terminates the processing of this flowchart.
[0064] The guidelines that serve as the criteria for the primary and secondary checks described in the embodiments above are determined by each company or user. An example of such guidelines is described in this embodiment.
[0065] Figure 12 is a diagram illustrating guidelines regarding the clear space of a corporate logo. As shown in Figure 12, the size of the area reserved for clear space differs for each corporate logo. For example, in the logo 1201 shown in Figure 12(a), when the length of one side of the part showing the corporate mark is A, the clear space 1204 of the square of the logo is defined as 1 / 3A on each side. In this embodiment, determining whether clear space is secured is included in the primary check and is performed by the image analysis program. Therefore, the image analysis program provided in the first server device 102 will determine whether any objects such as text overlap with the area defined in the guidelines as the clear space of the logo. Even for logos different from Figure 12(a), as shown in Figures 12(b) and 12(c), it is necessary to check whether the clear space defined in the guidelines is secured.
[0066] Other examples of rules stipulated in the guidelines include prohibited terms in text, regular expressions for company names in text, co-distribution with third-party logos, backgrounds that impair the visibility of the logo or mark (background suitability), and the use of logos within text.
[0067] As explained above, this embodiment makes it possible to improve the accuracy of creative checks. Specifically, creative checks are requested from an image analysis program that excels at objective judgments based on calculations and image processing, and a generative AI that excels at overall layout validity based on the context of the text contained in the image, as well as checks for compliance with the Act against Unjustifiable Premiums and Misleading Representations. The results of each check are then integrated and notified to the user. Furthermore, the secondary check by the generative AI is performed using accurate image processing information measured at the pixel level by the image analysis program, thus improving the accuracy of the check. As described above, by reducing the effort required from users in creative checks, work efficiency can be improved, while the image analysis program and generation AI can enhance the accuracy of checks on aspects that are difficult for the human eye to judge.
[0068] <Other Embodiments> In the above-described embodiment, an information processing system was explained as an example comprising an information processing device 101 equipped with a tool 301, a first server device 102 capable of executing an image analysis program, and a second server device 103 equipped with a generation AI. However, the system is not limited to this, and for example, it may consist of only one server device. That is, the information processing system may consist of one server device equipped with an image analysis program and a generation AI, and an information processing device. In other words, the primary check using the image analysis program and the secondary check using the generation AI may be performed on the same device. In this case, the information processing device only needs to request a creative check from the server device and then obtain the results of both checks.
[0069] Furthermore, the processing of data transmission and reception with each server device, requests for checks, and compilation of check results may be performed by an external device separate from the information processing device 101 and each server device. In this case, the external device acts as an intermediary between the information processing device 101 and each server device, receives instructions from the tool 301, communicates with each server, generates a final report, and notifies the information processing device 101. This reduces the processing load on the information processing device 101.
[0070] Furthermore, while the above-described embodiment explained an example where the image to be checked contained one logo and multiple texts, in reality, it is not limited to this, and for example, multiple logos may be placed within the image. In this case, a primary or secondary check can be performed for each of the multiple logos, and the check results for each logo can be recorded in the intermediate data or the final report.
[0071] In addition to the embodiments described above, when the receiving unit 302 of the tool 301 receives an image to be checked, it may also receive information such as what type of logo is included in the image and what medium the image to be checked will be output to. By receiving the above information in addition, the content of the creative check may be changed or added depending on the type of logo and the output medium.
Claims
1. A program for performing an image check including a logo, which operates on an information processing device, the information processing device The acquisition means for acquiring the aforementioned image, A first checking means controls an image analysis program that does not use generating AI to perform a first check, which includes a process of measuring and identifying the areas occupied by the logo and the areas occupied by the text in the aforementioned image. A second checking means controls the generating AI to perform a second check, which includes a check on the layout of the image, using intermediate data that includes information identifying the area occupied by the logo and the area occupied by the text in the image, generated by the first check, and the result of the first check. Notification means for providing notification of the results of the first check and the second check, A program characterized by being designed to function as such.
2. The program according to claim 1, characterized in that the information used to identify the area occupied by the logo and the area occupied by the text in the aforementioned image includes the size of the logo, the position information of the logo, the size of the text, and the position information of the text.
3. The program according to claim 1, wherein, if multiple logos or texts are placed in the aforementioned image, the first checking means outputs information that identifies the area occupied by the number of logos and texts that are placed.
4. The program according to claim 1, wherein the second checking means reads the area occupied by the text in the image by performing OCR (Optical Character Recognition) on the information that identifies the area occupied by the text contained in the intermediate data, and performs the second check.
5. The program according to claim 1, characterized in that the first check includes at least one of the following items: checking the minimum size of the logo included in the image, checking for clear space, checking for rotation or deformation, or checking for color variations.
6. The program according to claim 1, characterized in that if the check results for all items included in the first check indicate OK, the first check means determines the overall result of the first check to be OK.
7. The program according to claim 1, characterized in that the second check further includes a check relating to the Premiums and Representations Act for the image.
8. The program according to claim 1, characterized in that the layout check includes at least one of the following: checking for prohibited words in the text included in the image, checking for the use of the logo object within text, checking for the independence of the logo object, or checking for rules regarding the layout.
9. If the overall result of the first check indicates OK, the second checking means controls the generating AI to perform the second check. The program according to claim 6, characterized in that, if the overall result of the first check indicates NG, the second checking means controls the generating AI not to perform the second check, and the notification means notifies the result of the first check.
10. The program according to claim 1, characterized in that the notification method of the notification means includes displaying the check results on the display screen of the information processing device, outputting the sheet with the check results inserted as a file, or notifying via email.
11. Means for acquiring images, A first checking means controls an image analysis program that does not use generating AI to perform a first check, which includes a process of measuring and identifying the areas occupied by the logo and the areas occupied by the text in the aforementioned image. A second checking means controls the generating AI to perform a second check, which includes a check on the layout of the image, using intermediate data that includes information identifying the area occupied by the logo and the area occupied by the text in the image, generated by the first check, and the result of the first check. Notification means for providing notification of the results of the first check and the second check, An information processing device characterized by comprising:
12. Image acquisition step, A first check step involves controlling an image analysis program that does not use generating AI to perform a first check, which includes a process of measuring and identifying the areas occupied by the logo and the areas occupied by the text in the aforementioned image. A second check step controls the generating AI to perform a second check using intermediate data that includes information identifying the area occupied by the logo and the area occupied by the text in the image, generated by the first check, and the result of the first check. A notification step that provides notification indicating the results of the first check and the second check, A control method for an information processing device, characterized by comprising the following:
Citation Information
Patent Citations
Information processor and its control method
JP2008146604A
Information processing device, information processing method, and information processing program
JP2024027687A
Information processing device, information processing method, and information processing program
JP2024027690A
Examination supporting apparatus, examination supporting method, and examination supporting program
JP2024143739A
System
JP2025046435A