Platform system for creating vehicle performance checklists based on AI security cameras and disclosing information to the public

KR103047266B1Active Publication Date: 2026-11-11KOREA PUBLIC CLOUD RESEARCHER CO LTD +3
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
KR1020260031431
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2026-02-19
Publication Date
2026-11-11
Estimated Expiration
2046-02-19

Smart Images

  • Figure 112026020725424-PAT00001_ABST
    Figure 112026020725424-PAT00001_ABST
Patent Text Reader

Abstract

The present invention relates to an AI security camera-based vehicle performance checklist generation and public information disclosure platform system, characterized by comprising a user terminal equipped with a display unit and a user input unit.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to an AI security camera-based automobile performance checklist generation and public information disclosure platform system. Background Technology

[0002] Generally, vehicle performance inspection checklists are utilized for used car transactions, maintenance history management, and performance status verification. These checklists are typically generated by mechanics or inspectors who visually inspect the vehicle's exterior, major structural parts, and accident history, either by handwriting them or by electronically entering only selected items. Additionally, vehicle accident history and registration information are separately retrieved through history information systems provided by public institutions such as the Ministry of Land, Infrastructure and Transport and the Korea Insurance Development Institute.

[0003] However, conventional methods for compiling vehicle performance inspection reports rely heavily on the inspector's subjective judgment, and since the inspection process is often not sufficiently documented through filming or records, there is a problem in that it is difficult to verify the actual environment and the person who performed the inspection. In particular, when inspection results are circulated without clearly linking the identity and qualifications of the inspector, disputes regarding the reliability of the results may frequently arise.

[0004] In addition, conventional technology has a problem in that images or videos generated during the vehicle shooting process are managed separately from inspection results or are not systematically combined with information on the time, location, and inspection subject, making it difficult to verify whether the images have been falsified or altered after the fact. Furthermore, even if discrepancies occur between vehicle history information held by public institutions and actual inspection results, there is no structure in place to automatically detect or reflect these discrepancies during the performance inspection checklist creation stage, causing consumers to experience the inconvenience of having to compare different information sources individually.

[0005] As such, conventional methods of automobile performance inspection and information provision have technical limitations in terms of ensuring the accountability of the inspection entity, guaranteeing the integrity of inspection data, and providing consistent public history information with the actual vehicle condition. The problem to be solved

[0006] The present invention is intended to solve the aforementioned problems. Since a performance inspection checklist is generated with the identity and qualifications of the entity performing the inspection clearly verified during the vehicle performance inspection process, the responsibility for the inspection results is clarified and the reliability of the inspection results is improved. Consequently, the problem of unclear inspection entities or difficulty in verifying the source of inspection results can be resolved.

[0007] In addition, since information regarding the time of shooting, location of shooting, and inspection entity is managed in a combined form with the captured image or video during the vehicle shooting process, it is possible to prevent the shooting data used for inspection from being falsified or replaced with data from other vehicles after the fact, and thereby objectively support the fact that the visual data included in the performance inspection checklist was generated at the actual inspection site.

[0008] In addition, since the shooting sequence, shooting areas, and shooting conditions are managed according to certain standards during the shooting process, problems such as the omission of essential parts required for performance inspection or the creation of checklists with significantly degraded shooting quality can be reduced, and accordingly, the effect of maintaining more uniform quality in the creation of performance checklists can be expected.

[0009] Furthermore, since the inspection results regarding the actual vehicle condition and vehicle history information held by public institutions are provided together according to the same standards, users can make a comprehensive judgment through a single performance checklist without the need to individually compare different information regarding accident history or vehicle condition; in particular, since discrepancies between public history information and actual inspection results are reflected in the performance checklist, this contributes to mitigating information asymmetry that may occur during the vehicle transaction process.

[0010] Furthermore, by providing vehicle performance information in a format easily accessible to general users, this project offers an AI security camera-based platform system for generating vehicle performance checklists and disclosing information to the public, which is expected to enhance transparency in used car transactions and vehicle information verification processes, and reduce the likelihood of disputes surrounding performance checklists. means of solving the problem

[0011] A system for generating an AI security camera-based vehicle performance checklist and disclosing information to the public, according to one embodiment of the present invention, is characterized by comprising a user terminal equipped with a display unit and a user input unit.

[0012] It includes a server connected to the above-mentioned user terminal via a wired or wireless network, wherein the server comprises a communication unit; a storage unit; and a control unit; wherein the control unit, when a mechanic or performance inspector accesses through the user terminal, performs an authentication procedure using biometric information to identify the inspector, generates identity information including a license number, user identifier, access time, and access location information corresponding to the identified inspector, and stores it in a storage unit; during the process of photographing a vehicle through an AI security camera linked to the user terminal, generates shooting information by combining the vehicle license plate recognition result, GPS coordinates of the shooting location, the shooting time, and the identity information, and creates a security processing image file by inserting the shooting information into the pixel data area of ​​the captured image or video using a digital watermarking method, and after the start of shooting, sequentially identifies the shooting sequence, shooting angle, shooting area, and shooting conditions according to predefined inspection procedure data, and at each stage, judges the shooting quality by comparing the illumination, focus, shake, and field of view of the captured image with reference data, outputs shooting guidance information to the user terminal to induce re-shooting if the judgment result regarding the shooting quality does not meet a pre-set standard, and includes the vehicle body exterior, suspected accident area, sheet metal or paint traces, and undercarriage condition from the captured security processing image file. It is characterized by being configured to identify vehicle status information using an AI image recognition method, and simultaneously receive supplementary opinions through input from the inspection entity to generate inspection data merged with the vehicle status information.

[0013] The control unit receives vehicle history information including vehicle registration information, accident history, and inspection history by linking with at least one of the Ministry of Land, Infrastructure and Transport’s automobile information system, the Korea Insurance Development Institute’s accident history system, and the vehicle registration information system based on the vehicle license plate recognition result; corresponds the vehicle status information and the vehicle history information by item, calculates a status value for each item by matching it with each standard item code of the automobile performance inspection checklist notified by the Ministry of Land, Infrastructure and Transport corresponding to the vehicle body exterior condition, accident traces, and the condition of major structural parts; if the calculated status value differs from the information in the public history, identifies the item as a discrepancy item and classifies it as a peculiarity, automatically enters the classification result into the performance inspection checklist item, automatically generates an automobile performance inspection checklist by combining the real name and qualification information of the inspection entity and a shooting result with digital watermarking applied, stores the generated performance inspection checklist in the storage unit, and when a general user searches by entering a vehicle number, generates a vehicle performance information screen for public disclosure that includes the item-by-item results of the performance inspection checklist, the real name information of the inspection entity, and the shooting-based fact verification result, wherein the vehicle owner’s personal information It is characterized by controlling the output to be de-identified.

[0014] The control unit sequentially records shooting status information, including the shooting time, shooting location coordinates, shooting angle, shooting target area, and the biometric authentication maintenance status of the shooting subject, throughout the entire shooting process in which the security processing image file is generated, at each shooting point; based on the shooting status information, constructs a shooting progress sequence by combining the trend of change in time intervals between shooting points, the cumulative value of the distance traveled between shooting location coordinates, the directionality of change in the shooting angle, and the switching order of the shooting target area; verifies step-by-step based on whether the shooting progress sequence involves continuous movement relative to the vehicle's outer perimeter, whether excessive repeated shooting has occurred at the same shooting location, and whether the switching of the shooting target area is inconsistent with a predefined inspection procedure sequence; if any of the step-by-step verification processes of the shooting progress sequence fail to meet the criteria, the entire shooting result is classified as not being data generated by normal performance inspection activities and controlled to be excluded from the vehicle status information generation stage; and only when the shooting progress sequence satisfies all of the criteria, the shooting result is confirmed as the result of an inspection activity performed by an inspection subject moving relative to the vehicle at the actual site, and the confirmed result is used for the automatic generation of the vehicle performance inspection checklist. It is characterized by being configured such that, by controlling usage, the possibility of creating a performance checklist is determined based on whether the method of proceeding with the shooting activity itself faithfully followed a predefined inspection procedure, regardless of whether the captured images were simply collected.

[0015] The control unit is characterized by generating multiple independent record paths based on the time of creation, creation order, and reference relationship of each piece of information regarding vehicle status information, vehicle history information, and shooting sequence used in the automatic generation process of the vehicle performance inspection checklist, and performing internal consistency verification for each of the record paths based on whether the combination order of the vehicle number, shooting time, inspection subject identification information, and performance inspection checklist item code is the same, and if it is determined that any of the multiple record paths has a different creation order, reference relationship, or combination result for the same item, controlling the performance inspection checklist to be classified as not being a properly confirmed inspection result and excluded from public disclosure, and only when it is determined that all of the multiple record paths have the same creation order and reference relationship, controlling the vehicle performance inspection checklist to be identified as a confirmable result and recorded in a storage unit in a final confirmed state, so that the performance inspection checklist is disclosed only when the consistency of the internal processing paths in which the inspection result is generated, combined, and confirmed is confirmed.

[0016] The control unit is characterized by defining the pattern of repeated and changing inspection activities over time as a single inspection activity pattern based on the shooting progress sequence, vehicle status information, and the frequency of the inspection subject's shooting intervention collected throughout the process of generating the vehicle performance inspection checklist; calculating a stability index for the inspection activity pattern based on the distribution change of the interval between shooting points, the temporal variation of the frequency of switching the shooting target parts, and the degree to which the timing of the inspection subject's intervention is concentrated in a specific section; if the stability index shows a pattern of rapid change in a short period or discontinuous distortion in a specific section, the control unit classifies the inspection activity pattern as different from the stable activity pattern observed in normal field inspection activities and controls the vehicle performance inspection checklist to remain in an unconfirmed state; and only when the stability index shows a pattern of gradual change or repetitive convergence throughout the entire inspection process, the control unit identifies the inspection activity pattern as the result of an inspection activity performed in the actual field and controls the transition to the performance inspection checklist confirmation stage, thereby determining whether the performance inspection checklist can be confirmed based on the stability of the temporal and behavioral patterns appearing throughout the inspection activity. Effects of the invention

[0017] Since the performance inspection checklist is generated with the identity and qualifications of the entity performing the inspection clearly verified during the vehicle performance inspection process, the responsibility for the inspection results is clarified and the reliability of the results is improved; consequently, problems where the inspection entity is unclear or it is difficult to verify the source of the inspection results can be resolved.

[0018] In addition, since information regarding the time of shooting, location of shooting, and inspection entity is managed in a combined form with the captured image or video during the vehicle shooting process, it is possible to prevent the shooting data used for inspection from being falsified or replaced with data from other vehicles after the fact, and thereby objectively support the fact that the visual data included in the performance inspection checklist was generated at the actual inspection site.

[0019] In addition, since the shooting sequence, shooting areas, and shooting conditions are managed according to certain standards during the shooting process, problems such as the omission of essential parts required for performance inspection or the creation of checklists with significantly degraded shooting quality can be reduced, and accordingly, the effect of maintaining more uniform quality in the creation of performance checklists can be expected.

[0020] Furthermore, since the inspection results regarding the actual vehicle condition and vehicle history information held by public institutions are provided together according to the same standards, users can make a comprehensive judgment through a single performance checklist without the need to individually compare different information regarding accident history or vehicle condition; in particular, since discrepancies between public history information and actual inspection results are reflected in the performance checklist, this contributes to mitigating information asymmetry that may occur during the vehicle transaction process.

[0021] In addition, as vehicle performance information is provided in a format easily accessible to general users, it is possible to provide an AI security camera-based platform system for generating vehicle performance checklists and disclosing information to the public, which is expected to enhance transparency in used car transactions and vehicle information verification processes and reduce the likelihood of disputes surrounding performance checklists. Brief explanation of the drawing

[0022] FIG. 1 is a diagram illustrating the schematic configuration of an AI security camera-based automobile performance checklist generation and public information disclosure platform system according to one embodiment of the present invention. FIG. 2 is a diagram illustrating the schematic configuration of an AI security camera-based automobile performance checklist generation and public information disclosure platform system according to one embodiment of the present invention. FIG. 3 is a schematic flowchart illustrating an AI security camera-based automobile performance checklist generation and public information disclosure platform system according to one embodiment of the present invention. FIGS. 4 and 5 are drawings illustrating an example of a process in which an AI security camera-based automobile performance checklist generation and public information disclosure platform system according to an embodiment of the present invention is implemented. Specific details for implementing the invention

[0023] Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. In the drawings, identical reference numbers or symbols refer to components that perform substantially the same function, and the size of each component in the drawings may be exaggerated for clarity and convenience of explanation. However, the technical concept of the present invention and its core components and operations are not limited only to the components or operations described in the following embodiments. In describing the present invention, if it is determined that a detailed description of known technologies or components related to the present invention may unnecessarily obscure the essence of the invention, such detailed description will be omitted.

[0024] In embodiments of the present invention, terms including ordinal numbers, such as first, second, etc., are used solely for the purpose of distinguishing one component from another, and singular expressions include plural expressions unless the context clearly indicates otherwise. Furthermore, in embodiments of the present invention, terms such as 'composed of,' 'include,' 'have,' etc., should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof. Additionally, in embodiments of the present invention, 'module' or 'part' performs at least one function or operation and may be implemented in hardware or software, or as a combination of hardware and software, or may be integrated into at least one module and implemented as at least one processor. Furthermore, in embodiments of the present invention, 'at least one' among a plurality of elements refers not only to all of the plurality of elements but also to each individual element excluding the remainder or all combinations thereof. Additionally, "configured to" may be used interchangeably with, depending on the context, for example, "suitable for," "having the capacity to," "designed to," "adapted to," "made to," or "capable of." "Configured to" does not necessarily mean that it is "specifically designed to" in hardware. Instead, in some situations, the expression "device configured to" may mean that the device is "capable of" doing so in conjunction with other devices or components.For example, the phrase “a processor configured (or set) to perform A, B, and C” may mean a dedicated processor for performing said operations (e.g., an embedded processor), or a generic-purpose processor (e.g., a CPU or an application processor) capable of performing said operations by executing one or more software programs stored in a memory device.

[0025] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. This description is intended to be detailed enough for a person skilled in the art to easily practice the invention, and it should be noted that the technical scope and concept of the present invention are not limited thereby.

[0026] FIG. 1 is a diagram illustrating the schematic configuration of an AI security camera-based vehicle performance checklist generation and public information disclosure platform system according to an embodiment of the present invention; FIG. 2 is a diagram illustrating the schematic configuration of an AI security camera-based vehicle performance checklist generation and public information disclosure platform system according to an embodiment of the present invention; FIG. 3 is a diagram illustrating a schematic flowchart of an AI security camera-based vehicle performance checklist generation and public information disclosure platform system according to an embodiment of the present invention; and FIG. 4 and FIG. 5 are diagrams illustrating an example of a process in which an AI security camera-based vehicle performance checklist generation and public information disclosure platform system according to an embodiment of the present invention is implemented.

[0027] Referring to FIGS. 1 to 5, an AI security camera-based vehicle performance checklist generation and public information disclosure platform system according to one embodiment of the present invention comprises: a display unit; and a user terminal provided with a user input unit.

[0028] A user terminal (100) according to one embodiment of the present invention includes, for example, a personal computer, a server computer, a handheld or laptop device, a mobile device (mobile phone, PDA, media player, etc.), a multiprocessor system, a consumer electronic device, a mini computer, a mainframe computer, a distributed computing environment including any of the aforementioned systems or devices, and an edge computing environment in which data is processed at the edge where data is generated rather than at a central server, and the configuration is not limited only to what is described.

[0029] The user terminal (100) may include at least one processor and memory. Here, the processor may include, for example, a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), and may have multiple cores.

[0030] The memory may be volatile memory (e.g., RAM, etc.), non-volatile memory (e.g., ROM, flash memory, etc.), or a combination thereof. Additionally, the user terminal (100) may include additional storage. The storage includes, but is not limited to, magnetic storage, optical storage, etc. The storage may store computer-readable instructions for implementing one or more embodiments disclosed herein, and may also store other computer-readable instructions for implementing an operating system, an application program, etc. Computer-readable instructions stored in the storage may be loaded into memory to be executed by a processor.

[0031] Additionally, the user terminal (100) may include a user input unit (110) and an output device. The user input unit (110) may include, for example, a keyboard, a mouse, a pen, a voice input device, a touch input device, an infrared camera, a video input device, or any other input device. Additionally, the output device may include, for example, one or more displays, speakers, printers, or any other output devices. Additionally, the computing device may use an input device or an output device provided in another computing device as the user input unit (110) or output device. Additionally, the computing device may include a communication module that enables the computing device to communicate with another device. Here, the communication module may include a modem, a network interface card (NIC), an integrated network interface, a radio frequency transmitter / receiver, an infrared port, a USB connection, or other interfaces for connecting the computing device to another computing device. The communication module may include a wired connection or a wireless connection.

[0032] Each component of the user terminal (100) may be connected by various interconnections such as a bus (e.g., peripheral component interconnection (PCI), USB, firmware (IEEE 1394), optical bus structure, etc.) or may be interconnected by a network. Terms such as “component,” “system,” etc. used in this specification generally refer to computer-related entities that are hardware, a combination of hardware and software, software, or running software.

[0033] A user terminal (100) according to one embodiment of the present invention may include a display unit (120), and the method of implementing the display of the display unit (120) is not limited and may be implemented in various display methods such as liquid crystal, plasma, light-emitting diode, organic light-emitting diode, surface-conduction electron-emitter, carbon nanotube, nanocrystal, etc. In the case of a liquid crystal method, the display unit (120) includes a liquid crystal display panel, a backlight unit that supplies light to the liquid crystal display panel, and a panel driving unit that drives the liquid crystal display panel. Meanwhile, the display unit (120) may be implemented as an OLED panel, which is a self-emissive element, without a backlight unit.

[0034] Meanwhile, AI according to one embodiment of the present invention refers to the entirety of information processing technology configured to derive results similar to cognitive judgment, recognition, classification, prediction, or decision performed by humans, and is not limited to a specific implementation method or algorithm structure. Such AI can be implemented in various ways depending on the form of input data, processing purpose, and utilization environment, and may be a logical processing structure implemented in software or may be implemented in a form combined with hardware.

[0035] AI is generally configured to generate output values ​​by extracting or calculating features, patterns, or relationships contained in data; in this process, methods such as rule-based computation, statistical processing, probabilistic inference, or the calculation of correlations between data may be utilized individually or in combination. Here, data may include numerical data, text data, image data, voice data, video data, sensor data, or combinations thereof, and AI can be applied without being limited to the type or format of such data.

[0036] Furthermore, AI may operate by processing data according to predefined criteria or conditions, or by deriving processing results by comparing multiple data sets or referring to the changing patterns of accumulated data. In this case, the AI ​​may consist of a single processing stage or have a structure that performs multiple processing stages sequentially or in parallel, and different computational methods or judgment criteria may be applied at each stage.

[0037] AI can identify specific objects, states, attributes, or anomalies from data to be processed, or generate classification or judgment results by calculating similarities, differences, or trends of change between input data. Such identification or judgment can be performed through methods such as numerical comparison, threshold application, distribution analysis, pattern matching, or distance calculation in a multidimensional data space, and can be implemented regardless of whether specific mathematical formulas or algorithms are used.

[0038] Furthermore, AI can internally store or reference threshold values, weights, judgment sequences, or computation results used in the processing process to reflect them in subsequent processing; additionally, if identical or similar inputs are provided repeatedly, it can configure the processing flow differently by referencing past processing results. Such a structure can be utilized by AI to maintain consistency in processing results over time or to reflect changes in data characteristics.

[0039] Meanwhile, AI is not necessarily limited to including a learning process; it can be implemented not only in a structure where parameters are updated based on training data, but also by applying predefined operation rules or fixed sets of conditions. Furthermore, even when learning takes place, the timing, cycle, scope of training data, and method of learning are not restricted to a specific form and may include online learning, offline learning, or a combination thereof.

[0040] As such, the AI ​​mentioned in this invention is not a concept limited by a specific model structure, computation method, or implementation technology, but can be understood as a concept that comprehensively includes all forms of artificial intelligence information processing technology capable of generating meaningful judgment results through the input, processing, and output of data.

[0041] An AI security camera according to one embodiment of the present invention refers to an image processing device that includes a function for acquiring video or images of a subject to be captured and a function for performing information processing or judgment based on the acquired video or images, and is not limited to a specific installation type, shooting purpose, or field of application. Such an AI security camera may be implemented in a fixed, mobile, portable, or detachable structure, may be composed of a single device, or may be composed of a distributed structure in which multiple shooting devices are interconnected through a network.

[0042] AI security cameras are generally configured to convert optical information of a subject into electrical signals, including a lens, an image sensor, and a signal processing circuit. The captured video or image can be stored in real-time or non-real-time or transmitted to an external device. The image sensor can be implemented in various forms such as CCD, CMOS, or similar, and the resolution, frame rate, sensitivity, and field of view can be set differently depending on the application environment.

[0043] In addition, AI security cameras can perform information processing functions on captured video or images, such as preprocessing, feature extraction, object recognition, state determination, or event detection. This processing may be performed internally within the camera or through integration with an external server or other computing device. The preprocessing stage may include operations such as noise removal, brightness correction, distortion correction, or resolution conversion, while the feature extraction stage may perform processing to identify specific regions, shapes, colors, movements, or patterns within the video.

[0044] AI security cameras can acquire or generate additional information related to the time of shooting, shooting location, shooting direction, or shooting conditions, and this additional information can be combined with video or image data to be stored or transmitted. For example, location information can be determined by a satellite-based positioning system, network-based location information, or user input, and the time of shooting can be set by an internal clock or an external time synchronization signal. Such additional information can be used as metadata for the management, retrieval, verification, or utilization of video data.

[0045] Furthermore, AI security cameras can incorporate security features and apply technologies to ensure the integrity of captured video or images, as well as access control and prevention of tampering. These security features may include various methods such as data encryption, access control, watermarking, hash generation, or signature assignment, which can be applied individually or in combination. This prevents captured video or images from being arbitrarily altered or used without authorization after their creation.

[0046] Meanwhile, AI security cameras can flexibly change their shooting or information processing methods depending on changes in the subject, environmental conditions, or the purpose of use, and their operating modes can be switched based on user input, external control signals, or predefined conditions. For example, they can be configured to perform shooting only in specific situations or to perform information processing only when specific conditions are met.

[0047] As such, the AI ​​security camera mentioned in the present invention refers to a device capable of performing various information processing and security functions based on image data, going beyond a simple image acquisition device, and can be understood as a concept whose scope is not limited by specific technical configurations, processing methods, or fields of application.

[0048] An automobile performance inspection according to one embodiment of the present invention comprehensively refers to an inspection activity performed on the exterior, internal configuration, major components, and operating status of a vehicle to verify the current state, functional operation, or structural soundness of the vehicle, and is not limited by specific inspection items, inspection sequence, or inspection method. Such automobile performance inspection may be performed in various ways depending on the purpose of use, driving history, year of manufacture, or inspection environment of the vehicle, and may be carried out in the form of a single-item inspection or a comprehensive inspection including multiple items.

[0049] Vehicle performance inspection generally involves verifying the presence of exterior damage, deformation of the body structure, paint condition, and the installation or assembly status of major components; such verification can be carried out using various methods, including visual inspection, the use of measuring tools, or recording via imaging devices. Furthermore, performance inspection may be performed on parts directly related to driving safety, such as the vehicle's undercarriage, wheels and suspension systems, braking systems, and steering systems.

[0050] In addition, vehicle performance inspections may include procedures to verify the operating status of the vehicle's power transmission system, control system, or electrical and electronic system; during this process, the condition of the engine, transmission, battery, electrical components, or various sensors may be subject to inspection. Such inspections may be performed under conditions simulating actual driving, or they may be carried out by verifying the operation of specific functions while the vehicle is stationary.

[0051] Vehicle performance inspections may be performed in conjunction with the vehicle's past usage or maintenance history, and information such as accident history, maintenance history, or inspection history can be utilized as reference material for interpreting the inspection results. Such historical information is recorded alongside the inspection results or used for comparison, contributing to a more comprehensive understanding of the vehicle's condition.

[0052] In addition, vehicle performance inspections may include the process of generating documents or data to record, store, or provide inspection results, and such records may be managed in the form of paper documents, electronic documents, or databases. Inspection results may include the status, specific details, or reference information for each inspection item, and may be provided in the form of summary information or detailed information depending on the purpose of the inspection.

[0053] Meanwhile, vehicle performance inspections may be performed in accordance with laws, industry standards, or agency-specific criteria, but are not limited thereto; they may be conducted flexibly according to various standards for the purposes of vehicle management, transactions, insurance, maintenance, or ensuring safety. The entities performing the inspections are also not restricted by specific qualifications or roles and may be carried out by individuals, institutions, or systems.

[0054] As such, the automobile performance inspection mentioned in the present invention is a concept referring to the entire series of inspection activities for verifying the condition of a vehicle and utilizing the results, and it can be understood that its scope is not limited by specific inspection items, inspection methods, or methods of expressing results.

[0055] A public information disclosure platform according to one embodiment of the present invention refers to an overall information provision environment configured to allow a number of general users to access, view, or utilize information regardless of the creator, purpose of provision, or target of use of specific information, and may be implemented in the form of a single service or a combination of multiple services. Such a public information disclosure platform may be provided in a web-based, application-based, or mixed form, and may be accessed through a fixed terminal, a mobile terminal, or various devices capable of connecting to a network.

[0056] A public information disclosure platform may include functions for the collection, storage, management, and provision of information, wherein the information may be data generated by public institutions, private institutions, or individuals, and may include structured data, unstructured data, or a combination thereof. The platform may include periodic updating, synchronization, or verification processes to maintain the timeliness, accuracy, or consistency of the information, and these processes may be performed through automated methods or through the intervention of an administrator.

[0057] In addition, the public information disclosure platform may include search, filtering, classification, or visualization functions to enable users to efficiently search for and understand the information they desire. For example, features such as keyword-based search, classification by item, condition-based inquiry, or the provision of summary information may be included, and these functions can be combined in various ways depending on the nature of the information or the purpose of provision.

[0058] A public information disclosure platform may include functions for managing users' access rights or scope of use, and not all information is disclosed at the same level; some information may be provided restrictively or in a de-identified form depending on the sensitivity of the information, legal requirements, or provision policies. In this regard, personal information, sensitive information, or protected information may be processed in a manner such as masking, anonymization, or access restriction.

[0059] Furthermore, the public information disclosure platform may manage usage records, view history, or interaction information generated during the information provision process, and such information may be utilized for the purposes of improving service quality, analyzing usage patterns, or operational management. In addition, the platform may provide the source, time of creation, or update history of the information to enhance the reliability of the information provided.

[0060] Meanwhile, the public information disclosure platform may be linked with external systems or other information provision services, thereby enabling the integrated provision of multiple information sources or the transmission or sharing of information with external services. Such linkage can be implemented through application program interfaces, data exchange standards, or network communication methods.

[0061] As such, the public information disclosure platform mentioned in the present invention is not limited by specific technical configurations, operating entities, or types of information provided, and can be understood as a concept encompassing all forms of information provision systems or service environments for openly providing information to a large number of users.

[0062] A server (200) according to one embodiment of the present invention may include a communication unit (210), a storage unit (220), and a control unit (230).

[0063] A communication unit (210) according to one embodiment of the present invention can communicate with a user terminal (100) or other external electronic devices, etc., via wired or wireless communication methods. Therefore, in addition to a connection unit including a connector or terminal for wired connection, it can be implemented in various other communication methods. For example, it can be configured to perform one or more of Wi-Fi, Bluetooth, Zigbee, infrared communication, Radio Control, Ultra-Wide Band (UWM), Wireless USB, and Near Field Communication (NFC). The communication unit (210) may include communication modules such as Bluetooth Low Energy (BLE), Serial Port Profile (SPP), Wi-Fi Direct, infrared communication, Zigbee, and Near Field Communication (NFC). Additionally, the communication unit (210) can be implemented in the form of a device, a software module, a circuit, a chip, etc.

[0064] A communication unit (210) according to one embodiment of the present invention may include various communication modules as described above, and may include an IoT communication module having an IoT network for each carrier. An IoT communication module may refer to any IoT communication network in which a plurality of objects having separate communication units (210) are connected through a network to enable services based on various platforms. By using such an IoT communication module, a smoother communication network can be provided within a set area.

[0065] A storage unit (220) according to one embodiment of the present invention may receive information from a user terminal (100) or from an external search platform through a communication unit (210) and store it, and may receive text information and image information included in a plurality of web pages received by a control unit (230) and store it. The storage unit (220) may store various data according to the processing and control of the control unit (230) described later. The storage unit (220) may be accessed by the control unit (230) to perform reading, recording, modification, deletion, updating, etc. of data. The storage unit (220) may include non-volatile memory such as flash memory, hard disk drive, and SSD (solid-state drive) so as to preserve data regardless of whether system power is provided to the server (200). Additionally, the storage unit (220) may include volatile memory, such as a buffer or RAM, for temporarily loading data processed by the control unit (230).

[0066] A control unit (230) according to one embodiment of the present invention can perform control for the operation of various components of a server (200). The control unit (230) may include a control program (or instruction) that enables such control operation, an inactive memory in which the control program is installed, a volatile memory in which at least a portion of the installed control program is loaded, and at least one processor or CPU (Central Processing Unit) that executes the loaded control program. In addition, such a control program may be stored in an external electronic device other than the server (200).

[0067] The control program may include program(s) implemented in at least one form among BIOS, device driver, operating system, firmware, platform, and application. In one embodiment, the application may be pre-installed or stored in the server (200) at the time of manufacturing the server (200), or may be installed in the server (200) based on the received data by receiving data of the application from an external source when it is used later. The data of the application may be downloaded to the server (200) from an external server, such as an application market, for example, a platform according to the present invention, but is not limited thereto. Meanwhile, the control unit (230) may be implemented in the form of a device, S / W module, circuit, chip, etc., or a combination thereof.

[0068] A control unit according to one embodiment of the present invention performs an authentication procedure using biometric information to identify the inspector when a mechanic or performance inspector accesses the user terminal, and generates identity information including a license number, user identifier, time of access, and location of access corresponding to the identified inspector and stores it in a storage unit.

[0069] A control unit according to one embodiment of the present invention performs an authentication procedure using biometric information to verify whether the connection was made by a person qualified to perform actual inspection activities when a maintenance technician or a performance inspector connects through a user terminal. Here, biometric information refers to information capable of identifying an individual based on their physical or behavioral characteristics, such as fingerprints, facial images, iris images, voice signals, or combinations thereof, and is not limited to specific types of biometric information.

[0070] Specifically, when biometric information is collected through a sensor or input device installed on a user terminal, the collected biometric information is converted into data form and compared with reference biometric information previously registered in a storage unit. For example, in the case of fingerprint authentication, a feature vector containing the location coordinates and number of branch points is extracted from a fingerprint image, and the similarity between the feature vector and the feature vector included in the reference fingerprint data is calculated. In this case, similarity can be determined by whether the distance value calculated by summing the squared differences in distance between each branch point coordinate and applying the sqrt() operation is below a threshold value. In the case of face recognition, a vector quantifying the relative positional relationship between key feature points extracted from the face region is generated, and distance-based comparison can be performed in the same manner.

[0071] This biometric authentication method establishes thresholds by experimentally comparing the distribution of similarity between the same individual and others, utilizing sets of biometric data repeatedly measured from actual mechanics and performance inspectors. For instance, the average and variance ranges of distance values ​​are calculated by collecting the same individual's biometric data multiple times at different time points. The system is configured to determine authentication status based on the intervals that do not overlap with the distance value distributions derived when comparing the data with that of others. Through this procedure, the criteria used for comparing biometric data are not arbitrarily determined but are specifically derived based on actual collected data.

[0072] If the biometric authentication result meets the criteria, the control unit identifies the corresponding user as an inspection subject and generates identity information corresponding to the identified inspection subject. Here, identity information refers to a set of information intended to identify the inspection subject, including a license number, a user identifier, and connection time and location information. The license number is an identification number of a national or accredited qualification held by the inspection subject and can be retrieved through a correspondence with pre-registered qualification information data, while the user identifier may consist of a unique value assigned to distinguish the inspection subject within the system.

[0073] The connection time may be recorded in units of seconds or less based on the internal clock of the user terminal or server, and may be corrected using an external time synchronization signal. Connection location information may be calculated using a satellite-based location receiving module embedded in the user terminal, mobile communication base station information, or network connection point information. For example, in the case of GPS-based location information, latitude and longitude coordinates are received, and the distance and speed of movement between consecutive location coordinates are calculated to verify whether the coordinates are within a reliable range. The distance of movement can be calculated by converting the difference in latitude and longitude between two consecutive coordinates into planar coordinates and applying the sqrt() operation to the sum of the squares of the differences on each axis; if an unrealistic speed is calculated, the corresponding location data may be excluded.

[0074] The identity information generated in this manner is stored in a storage unit in a state logically combined with the inspection subject identified through biometric authentication. When stored, the license number, user identifier, access time, and access location information are bundled and recorded as a single record, and the inspection subject can be consistently tracked based on the said identity information even when an inspection activity is performed through the same user terminal or a different user terminal thereafter. Through this, the correspondence between the inspection result and the inspection subject is clearly maintained, and a person skilled in the art can easily implement this embodiment through the processes of biometric information collection, feature extraction, distance calculation, and reference value comparison as described above.

[0075] A control unit according to one embodiment of the present invention generates shooting information by combining a vehicle license plate recognition result, GPS coordinates of a shooting location, a time of shooting, and the identity information during the process of shooting a vehicle through an AI security camera linked to the user terminal, and generates a security-processed image file by inserting the shooting information into a pixel data area of ​​a captured image or video using a digital watermarking method.

[0076] A control unit according to one embodiment of the present invention generates shooting information by combining a plurality of pieces of information so that the source and generation environment of the shooting result can be clearly identified simultaneously with the shooting act, during the process of shooting a vehicle through an AI security camera linked to a user terminal. Here, shooting a vehicle refers to the act of acquiring visual data in the form of an image or video targeting a vehicle in a stationary or moving state, and the shooting method may be implemented as any one of single-frame shooting, continuous-frame shooting, or video shooting over a certain period of time.

[0077] The license plate recognition result is generated by extracting character patterns from areas containing license plates within captured image or video frames. Specifically, rectangular regions with high contrast within the captured frames are identified as candidate regions, and the horizontal and vertical projection values ​​of pixels within these regions are calculated to identify areas with a high probability of character arrangement. Subsequently, binarized pixel patterns are extracted from each character region, and the number string is derived through comparison with predefined numeric and character data. If characters with low recognition reliability are included in this process, the results recognized in adjacent frames within the same capture interval are compared, and the string with the highest frequency is confirmed as the final license plate recognition result. This license plate recognition procedure is verified by repeatedly applying license plate images captured under various lighting, angle, and distance conditions to confirm whether the recognition success rate remains stable.

[0078] The GPS coordinates of the shooting location are generated based on latitude and longitude values ​​received from a location receiving module embedded in the user terminal. While location information may consist of coordinate values ​​from a single point in time, the reliability of the location data can be verified by receiving the coordinates immediately before and at the time of shooting together and calculating the distance traveled between the coordinates. For example, the planar distance d between two consecutive coordinates can be calculated by squaring the difference between the coordinates of each axis, summing them, and applying the sqrt() operation; if the calculated distance appears abnormally large within a short period, the corresponding coordinate is determined to be an error value and excluded. This method is verified by comparing the distribution of location data repeatedly measured in various environments, such as actual parking lots, roadsides, and indoor spaces, and by configuring the system to use only coordinates that fall within a realistic range of movement.

[0079] The shooting time is generated based on the internal clock value of the user terminal and can be corrected by comparing it with an external time synchronization signal as needed. The shooting time is recorded in seconds or smaller units, allowing the exact point in time when a shot was taken on the same vehicle to be identified. The accuracy of the recorded shooting time can be verified by checking whether the time interval increases consistently during continuous shooting on the same terminal.

[0080] The aforementioned identity information is information generated in correspondence with an inspection subject already identified through a biometric authentication procedure, and includes a license number, user identifier, and connection time and location information. The control unit retrieves the most recently validly authenticated identity information at the time vehicle shooting is performed and logically combines it with the shooting result. Through this, the captured image or video is clearly linked to the fact that it was taken by a specific inspection subject at a specific time and location.

[0081] In this way, the capture information is constructed by combining the vehicle license plate recognition result, the GPS coordinates of the shooting location, the time of shooting, and identity information. The capture information is not merely a collection of character data, but can be generated as a data structure in which each item is arranged in a defined order and format. For example, a license plate string, latitude and longitude coordinate values, time information, and user identification value can be composed of a single data block along with a separator or length information, and the same capture information can be repeatedly referenced in subsequent processing steps.

[0082] The control unit generates a security-processed image file by inserting the above-mentioned shooting information into the pixel data area of ​​the captured image or video using a digital watermarking method. Digital watermarking can be performed by distributing and inserting the shooting information into pixel values ​​within a range that does not degrade the visual quality of the image. Specifically, a low-frequency or mid-frequency range that is difficult for humans to perceive among the brightness or color components of the image is selected, and the shooting information is converted into binary data form and inserted into the pixel values ​​of that range. For example, the shooting information can be recorded by adjusting the least significant bit of the pixel value or minute differences in value between adjacent pixels.

[0083] This watermarking method is verified by confirming whether the watermark data is successfully extracted even after undergoing repeated saving, transmission, and compression processes for images with the same captured information inserted. By comparing the success rates of watermark data extraction on actual image files with various resolutions and compression rates, an insertion method that stably maintains the captured information is selected.

[0084] As a result, the security-processed image file appears outwardly identical to a standard captured image or video, while maintaining a state in which shooting information is combined within the internal pixel data area. Through this, it is possible to verify from the shooting file itself when, where, and by whom the shooting result was generated, and a person skilled in the art can sufficiently implement this embodiment through the processes of license plate recognition, location and time collection, data combination, and pixel-based watermarking as described above.

[0085] A control unit according to one embodiment of the present invention identifies the shooting sequence, shooting angle, shooting area, and shooting conditions step by step according to predefined inspection procedure data after the shooting begins, determines the shooting quality by comparing the illumination, focus, shake, and field of view of the captured image at each step with reference data, and outputs shooting guidance information to the user terminal to induce re-shooting if the result of the judgment on the shooting quality does not satisfy a preset standard.

[0086] A control unit according to one embodiment of the present invention identifies the shooting sequence, shooting angle, shooting area, and shooting conditions in stages after vehicle shooting has commenced, in order to determine whether the shooting activity is proceeding normally according to a predefined inspection procedure. Here, inspection procedure data refers to a set of predetermined shooting steps for performing vehicle performance inspection, and refers to information in which the vehicle part to be shot, the recommended shooting angle range, the shooting distance range, and the shooting environment conditions are stored in the form of data for each step.

[0087] The identification of the shooting order can be based on the shooting time information generated whenever a shot is performed. Specifically, a shooting time is assigned to each captured image or video frame, and the control unit sorts this time information in chronological order to verify whether the current shooting stage matches the order defined in the inspection procedure data. For example, if a procedure is defined requiring a side shot to follow a front shot, the identification is performed by comparing whether the chronological order of the actual captured images follows that sequence. This sequence comparison is verified by confirming whether the general inspection progress sequence is stably maintained, based on actual shooting logs collected from multiple inspection cases.

[0088] The identification of the shooting angle can be performed by analyzing the orientation information of the AI ​​security camera at the time of shooting or the inclination of the vehicle contour within the captured image. For example, after detecting the major straight-line components of the vehicle's outline, the inclination value of the corresponding line can be calculated to determine whether the shooting angle falls within the horizontal, upward, or downward range. The inclination calculation can be performed using the difference in coordinates between the two endpoints of the line, and the angle range can be calculated based on the ratio of inclination to length, which is obtained by squaring the difference between the vertical and horizontal components, summing them, and applying the sqrt() operation. This angle determination method is verified by repeatedly analyzing images of actual vehicles taken from various angles to confirm whether an angle range suitable for inspection is clearly distinguishable.

[0089] Identification of the captured area can be performed based on whether specific structural features of the vehicle are included within the captured image. For example, the captured area can be distinguished by checking whether the shape of the headlights, radiator grille, or bumper is included for the front area, and whether the taillights or trunk area are included for the rear area. This identification of areas can be achieved by determining whether predefined morphological features are included in the captured image above a certain proportion, and is iteratively verified for each area using an actual inspection image dataset.

[0090] Identification of shooting conditions is intended to determine whether the shooting environment is suitable for inspection, and is performed based on image illumination, focus status, degree of shake, and field of view. Illumination can be determined by calculating the average brightness value of the entire image or a specific area, and can be configured to exclude cases that are excessively dark or bright. Focus status can be determined by measuring the sharpness of edge components within the image, and shake can be determined based on whether the variance of pixel position changes between consecutive frames exceeds a certain range. For example, if the value obtained by applying the sqrt() operation to the sum of the squared positional differences of identical feature points between consecutive frames exceeds a threshold value, it can be determined that the shake is significant. The field of view can be determined by calculating the proportion that the entire vehicle or the inspection target area occupies within the image and checking whether it is greater than or equal to a predefined minimum ratio.

[0091] The image quality values ​​calculated at each stage are compared with reference data included in the inspection procedure data. The reference data may consist of statistically derived values ​​for illumination ranges, focus clarity ranges, shake tolerance ranges, and field of view ranges that allow for normal inspection, obtained by analyzing inspection images repeatedly captured in various actual environments. These standards are experimentally verified by establishing them based on ranges consistently satisfied across multiple inspection cases, rather than relying on a single shooting result.

[0092] If the judgment result regarding the shooting quality does not meet the criteria, the control unit determines that the corresponding shooting step is not completed and outputs shooting guidance information to the user terminal. The shooting guidance information may include areas requiring re-shooting, insufficient shooting conditions, and matters requiring improvement; for example, it may be provided in the form of guidance to supplement lighting if illumination is insufficient, or guidance to stabilize the shooting posture if shaking is excessive. Through this, the inspector can immediately supplement the shooting status and perform re-shooting, and a person of ordinary skill in the art can sufficiently implement this embodiment through the process of identifying the shooting sequence, judging angles, areas, and conditions, and comparing reference data as described above.

[0093] A control unit according to one embodiment of the present invention is configured to identify vehicle condition information, including the exterior of the vehicle body, suspected accident areas, traces of sheet metal or paint, and the condition of the undercarriage, from a captured security processing image file using an AI image recognition method, and simultaneously receive supplementary opinions through input from the inspection subject to generate inspection data merged with the vehicle condition information.

[0094] A control unit according to one embodiment of the present invention identifies the state of a vehicle by component using a captured security-processed image file as input data. Here, a security-processed image file refers to a file in which the time of capture, location of capture, identity information, etc., are combined with an image or video generated at the time of capture and inserted into a pixel data area; it can be treated in the same format as a general image file, but is utilized as data in which the capture environment and the subject are logically connected.

[0095] The identification of the vehicle body exterior is performed based on the vehicle's outer contours and surface patterns included in the security-processed image file. Specifically, after extracting the vehicle's entire contour from the image, the area within the contour is divided into a grid, and the color distribution, contrast variation, and surface reflection characteristics of each grid area are calculated. Based on the grid-specific distribution ranges of images repeatedly captured of a vehicle with a normal exterior, areas where color deviation or contrast variation exceeds the specified range can be identified as potential areas of exterior anomaly. These criteria for judging the exterior are verified by analyzing a set of repeatedly captured images of vehicles of various models and colors to derive the distribution ranges that are commonly maintained in a normal state.

[0096] Identification of suspected accident sites is performed by focusing on areas on the vehicle body exterior where continuity is interrupted or asymmetrical shapes appear. For example, in areas such as bumpers, doors, or fenders that have a left-right symmetrical structure, if the change in curvature on one side differs significantly from the other, that area can be classified as a suspected accident site. Changes in curvature can be determined by calculating the change in slope between consecutive points on the contour line; if the trend of change in the value obtained by applying the sqrt() operation to the sum of the squared coordinate differences between adjacent points exceeds the normal range, it is determined to be an abnormal deformation. This method is verified by comparing and analyzing image data collected from actual accident vehicles and accident-free vehicles to determine whether deformation patterns that repeatedly appear during an accident can be identified.

[0097] The identification of traces of sheet metal repair or paintwork can be performed based on subtle texture differences and changes in color uniformity on the vehicle body surface. By comparing color histograms between adjacent areas of the same body within security-processed image files, regions where the center value or variance of the color distribution changes drastically can be classified as areas with potential paint traces. Additionally, surface texture can be determined by calculating the distribution of high-frequency components; if energy values ​​in specific frequency bands are significantly higher or lower compared to a normal painted surface, they can be identified as traces of sheet metal repair or repainting. These criteria are derived by collecting and repeatedly analyzing pre- and post-painting image data from vehicles with verified actual maintenance histories.

[0098] Identification of the undercarriage condition is performed on image or video frames capturing the vehicle's undercarriage. By analyzing the contours and surface conditions of the body frame, exhaust system, or major connection points within the undercarriage images, the presence of corrosion traces, deformation, or foreign matter adhesion can be determined. For example, the potential for corrosion can be assessed by detecting irregular dot patterns or color changes on the surface, or deformation can be identified based on whether the linearity of structural members is maintained. These judgment criteria are established by comparing image data repeatedly captured on vehicles in various conditions within an actual undercarriage shooting environment.

[0099] In this way, the vehicle body exterior, suspected accident areas, traces of sheet metal or paintwork, and the condition of the undercarriage are each identified based on independent image features and constitute vehicle condition information. The vehicle condition information may include the results identified for each item and information on the image area from which the results were derived; results derived from multiple images of the same vehicle can be comprehensively compared to determine the final state.

[0100] Meanwhile, the control unit receives supplementary comments through input from the inspector simultaneously with the image recognition results. Supplementary comments refer to information entered by the inspector in the form of text, selection items, or simple indicators, including specific details perceived by visual inspection or experience, conditions that are difficult to clearly distinguish in the image, or matters requiring further explanation. These supplementary comments can be stored linked to specific vehicle status information items and have a corresponding relationship with the image recognition results at the same item level.

[0101] The control unit generates inspection data by merging the vehicle status information and supplementary opinions. During the merging process, the system is configured so that image recognition results and the inspection subject's opinions are recorded together for the same item, and the data structure is defined to ensure that neither piece of information is omitted. For example, if image recognition results exist for a specific part and supplementary opinions are entered, the inspection data for that item is configured to include both pieces of information. Through this, the inspection data is generated in a form that combines automatically identified image-based information with human judgment, and a person skilled in the art can sufficiently implement this embodiment through the processes of image feature analysis, iterative data verification, and input information merging as described above.

[0102] A control unit according to one embodiment of the present invention receives vehicle history information including vehicle registration information, accident history, and inspection history by linking with at least one of the Ministry of Land, Infrastructure and Transport’s automobile information system, the Korea Insurance Development Institute’s accident history system, and the vehicle registration information system based on the vehicle license plate recognition result.

[0103] A control unit according to one embodiment of the present invention receives vehicle history information by linking with an external vehicle-related information system, using a vehicle license plate recognition result derived during the shooting process as reference information for identifying a vehicle. Here, the vehicle license plate recognition result refers to a vehicle number string derived from a captured image or video through a character recognition procedure, and said string is utilized as a unique identifier capable of legally identifying a vehicle.

[0104] When a vehicle license plate recognition result is generated, the control unit uses the corresponding number string as a query value to access at least one of the Ministry of Land, Infrastructure and Transport’s Automobile Information System, the Korea Insurance Development Institute’s Accident History System, and the Vehicle Registration Information System. This integration can be performed via a data request method over a network, and a request message including the vehicle number is transmitted according to the public or authorized interface specifications provided by each system. The request message may include the time of the request, a code value to distinguish the purpose of the request, and other elements along with the vehicle number string.

[0105] Vehicle registration information can be received through integration with the Ministry of Land, Infrastructure and Transport's vehicle information system. This information may include items such as vehicle type, model name, model year, initial registration date, usage classification, or deregistration status; this information is retrieved from the database based on the vehicle number and returned. The returned data has a structure organized by item, and the control unit can store it as is or convert it into a normalized format for subsequent processing.

[0106] Through integration with the Korea Insurance Development Institute's accident history system, accident records with a history of insurance claims for the relevant vehicle can be received. The accident history may include the time of the accident, the type of accident, whether insurance was processed, or information on the damaged parts, and past accident records are retrieved based on the vehicle registration number. This accident history data is provided through managed data based on repeatedly accumulated insurance records, so that it can be utilized to distinguish between vehicles that have actually been involved in accidents and those that are accident-free.

[0107] Inspection history related to periodic or temporary inspections can be received through integration with the vehicle registration information system. The inspection history may include information such as the time of inspection, the type of inspection, and whether the vehicle passed or failed, and inspection results for a certain period can be retrieved based on the vehicle number. This inspection history information can be used to determine whether the vehicle has met legal inspection requirements.

[0108] The control unit integrates the vehicle registration information, accident history, and inspection history received from each system as described above into vehicle history information. During the integration process, each piece of information is combined based on the same vehicle number to correspond to a single vehicle, and the information can be managed while maintaining its source, time of receipt, and item classification. For example, even if registration information exists for the same vehicle but accident history does not, the corresponding item can be stored clearly distinguished as an empty value or non-existent state without being omitted.

[0109] This procedure for receiving vehicle history information can be verified by repeatedly performing linkage requests on actual vehicle numbers to confirm whether appropriate data is returned for each of the following: normally registered vehicles, vehicles with accident histories, and vehicles with updated inspection histories. Additionally, it is possible to verify whether the linkage structure based on vehicle numbers operates correctly by checking whether no search results are returned or an error response is received when an incorrect license plate string is entered.

[0110] Consequently, the control unit can receive vehicle history information, including vehicle registration information, accident history, and inspection history, by linking with multiple external vehicle-related systems based on a single identification information, which is the result of vehicle license plate recognition; and a person skilled in the art can sufficiently implement this embodiment through the process of number-based querying, system linkage, and data reception as described above.

[0111] A control unit according to one embodiment of the present invention corresponds the vehicle status information and the vehicle history information by item, and calculates a status value for each item by matching them with each standard item code of the Automobile Performance Inspection Checklist notified by the Ministry of Land, Infrastructure and Transport, which corresponds to the exterior condition of the vehicle body, traces of accidents, and the condition of major structural parts.

[0112] A control unit according to one embodiment of the present invention corresponds vehicle status information generated during the vehicle shooting and information collection step with vehicle history information received through external linkage, item by item, based on the same vehicle. Here, vehicle status information refers to a set of information representing the physical state of a vehicle, such as the exterior condition of the vehicle body, traces of suspected accidents, traces of sheet metal work or painting, and the condition of the undercarriage, derived from the analysis results of captured images and input from the inspector; and vehicle history information refers to a set of information based on past history, such as registration information, accident history, and inspection history.

[0113] The control unit matches each item of the vehicle status information with the standard item codes specified in the Vehicle Performance Inspection Checklist notified by the Ministry of Land, Infrastructure and Transport. The Vehicle Performance Inspection Checklist defines the vehicle body exterior, major structural parts, and accident determination factors as coded standard items, and each item code is assigned to enable a one-to-one identification between the inspection target part and the inspection criteria. For example, the control unit matches the image analysis results corresponding to the vehicle body exterior condition with exterior item codes defined in the checklist, such as exterior panels, doors, fenders, and bumpers; matches information corresponding to accident traces or sheet metal repair marks with accident determination item codes; and matches the undercarriage condition with item codes corresponding to major structural parts, such as the frame, crossmember, and suspension joints.

[0114] Such matching can be performed through a pre-established item correspondence table. The item correspondence table defines the detailed item name of the vehicle status information, the standard item code corresponding to that item on the checklist, and the inspection criteria required by that code. The control unit sequentially queries each item of the vehicle status information, selects a standard item code that has the same or closest meaning, and applies a status value calculation procedure corresponding to that code.

[0115] The status value for each item is calculated by considering both vehicle condition information and vehicle history information. For example, in the case of the exterior item, the ratio of the damaged area detected in the captured image, the estimated damage depth, and the frequency of detection at the same location during repeated shooting can be compared with reference data to convert it into a stepwise status value such as normal, minor, or severe. For the accident trace item, the system can be configured to correlate the presence of accident history and the time of the accident included in the vehicle history information with the image-based identification results; this ensures that a higher-level status value is calculated if a past accident history exists and an exterior abnormality is detected in the same area. For the condition of major structural parts, the degree of deformation, asymmetry, and the difference from the reference shape in the captured image of the undercarriage are quantified for each item and compared with the standard range of the inspection checklist to convert them into a status value.

[0116] These criteria for calculating status values ​​can be established by comparing photographic data obtained through repeated inspections of actual vehicles with the results of manual performance checklists. For the same vehicle, the consistency of each item is verified by correlating the video-based analysis results with the entries on the official performance checklist; based on data organized through iterative verification to determine which status stage specific damage types correspond to, judgment criteria for each item are finalized. During this process, reference values ​​are adjusted by examining the distribution of results across multiple vehicles to ensure that specific items are not judged as excessively high or low.

[0117] Consequently, the control unit corresponds vehicle status information and vehicle history information by item, and by matching the exterior body condition, accident traces, and the condition of major structural parts with the standard item codes of the vehicle performance inspection checklist notified by the Ministry of Land, Infrastructure and Transport, it can calculate item-specific status values ​​that conform to the checklist format for each standard item, and this procedure can be sufficiently implemented by a skilled technician utilizing the structure of the standard checklist and actual vehicle inspection data.

[0118] A control unit according to one embodiment of the present invention identifies the item as a discrepancy item and classifies it as a peculiarity when the calculated status value differs from the information in the public history, automatically enters the classification result into an item of the performance inspection sheet, automatically generates an automobile performance inspection sheet by combining the real name and qualification information of the inspection subject and a shooting result with digital watermarking applied, and stores the generated performance inspection sheet in the storage unit.

[0119] A control unit according to an embodiment of the present invention compares whether there is a match with public history information based on the status value calculated for each item in the preceding step. Here, public history information refers to historical information managed by an authoritative external system, such as vehicle registration information received based on license plates, accident history, and inspection history, and the status value for each item is a value representing the current vehicle status calculated in correspondence with each standard item code of the vehicle performance inspection checklist. The control unit uses as a criterion for judgment whether the status value corresponding to the same standard item code and the content of the public history information show mutual contradiction or difference.

[0120] The above judgment is performed on an item-by-item basis. It determines whether there is a discrepancy between corresponding pieces of information, such as when traces of an accident are detected in a recording-based status value for an item confirmed to have no accident history in the public history, or when the current status value is calculated as normal for an item that has a past accident history. Since this comparison is performed based on discrete status classification values—such as the presence or absence of an accident, repair status, or structural damage—rather than a simple string comparison, a person of ordinary technical expertise can sufficiently implement it by referring to the field structure of the public history data and the item definitions of the performance checklist.

[0121] The control unit identifies items found to be inconsistent based on the comparison results as inconsistent items and classifies them as specific items. Items classified as specific items are automatically entered into the corresponding item area within the vehicle performance inspection checklist. The entered content may include the item code where the inconsistency occurred, the currently calculated status value, a summary of information from public history, and the result of the inconsistency determination. Through this, a third party viewing the inspection results can clearly identify which item and on what grounds the specific item occurred.

[0122] Subsequently, the control unit proceeds to the automatic generation stage of the performance checklist. The control unit combines the real name information of the inspector, qualification information verifying their eligibility to perform the inspection, and the shooting results to which digital watermarking generated during the shooting process has been applied, with the performance checklist data. The inspector's real name and qualification information are obtained through pre-registered user information or authentication procedures, enabling the clear identification of the party responsible for the inspection results. Since the shooting results to which digital watermarking has the shooting time, location, and vehicle license plate recognition results embedded within the image data itself, the authenticity of the shooting results can be verified retrospectively even after being combined with the performance checklist.

[0123] In this manner, data combining item-specific status values, classification results of specific details, inspection entity information, and imaging results is automatically arranged to fit the format structure of the vehicle performance inspection checklist, and each item is automatically entered into its corresponding standard item code position. During this process, the entire performance inspection checklist is generated without the need for separate manual input, and the generated checklist is stored in the storage unit as an electronic document. Along with the original performance inspection checklist data, the storage unit can also store the creation time, inspection entity identification information, and identifiers of referenced public history information, allowing the entire inspection process to be traced during subsequent viewing or verification.

[0124] This series of processing steps can be configured to accumulate and verify instances of discrepancies between status values ​​calculated by repeatedly applying actual vehicle inspection cases and public history information, and item-specific judgment criteria and automatic entry rules are established based on the results of comparing multiple actual inspection data with the official performance inspection checklist results. Therefore, since the above configuration is derived through a verification process based on actual data rather than arbitrary judgment, it can be fully implemented by a skilled technician who understands the vehicle performance inspection procedure and the public history linkage structure.

[0125] A control unit according to one embodiment of the present invention controls the generation of a vehicle performance information screen for public disclosure, which includes the results for each item of the performance inspection checklist, the real name information of the inspection subject, and the results of photographic fact verification, when a general user enters a vehicle number to search, while de-identifying and outputting the vehicle owner's personal information.

[0126] A control unit according to an embodiment of the present invention, when a general user requests a search by entering a vehicle number through a user terminal, uses the vehicle number as a search key value to retrieve data associated with an already generated and stored vehicle performance inspection sheet. Here, a general user refers to a user who has been granted the right to view public information regardless of whether they are the owner of the vehicle, and the vehicle number input can be performed through various input means, such as keyboard input, touch input, or character recognition using a camera. The control unit verifies whether the format of the entered vehicle number corresponds to a normal license plate format based on criteria such as character length and character combination rules, and then verifies whether performance inspection result data corresponding to the number exists.

[0127] If search is possible, the control unit generates a vehicle performance information screen for public disclosure based on the item-specific results listed in the performance inspection checklist. Item-specific results refer to the results calculated in correspondence with each standard item code in the performance inspection checklist, such as the condition of the vehicle body exterior, the presence of accident traces, and the condition of major structural parts; each item is converted into a status value, such as normal, abnormal, or specific, or an equivalent expression, and displayed on the screen. In this process, the control unit may generate the screen by configuring the item names and status results with visual icons, color coding, or simple explanatory text, rather than exposing the original performance inspection checklist as is, so that general users can understand them.

[0128] In addition, the control unit includes the real-name information of the inspection entity. Here, the real-name information refers to the name of the mechanic or performance inspector who performed the relevant performance inspection, and this information is displayed on the screen as the minimum public information required to ensure the reliability of the inspection results. However, additional personal information such as the inspection entity's contact details or address is not included; the system is configured to output only information at the level of the real name.

[0129] In addition, the control unit includes the results of fact-checking based on shooting on the screen for public disclosure. The results of fact-checking based on shooting refer to information proving that the relevant inspection item was actually captured and verified, based on images or videos taken during the performance inspection process to which digital watermarking has been applied. The control unit may not disclose the entire original shooting file as is, but may include only the information within the scope necessary for fact-checking on the screen for each item, such as a representative image, shooting time, and summary information on the shooting location. In this case, since the shooting results to which digital watermarking has been applied combine the shooting time, location, and vehicle number with the image data itself, it can be subsequently confirmed that the shooting results displayed on the public screen are data generated during the actual inspection.

[0130] Meanwhile, the control unit controls the output of vehicle owner personal information by de-identifying it when generating a vehicle performance information screen for public disclosure. Vehicle owner personal information refers to information that can identify an individual, such as the vehicle owner's name, resident registration number, address, and contact information, and such information is not included in the public screen. De-identification processing may be performed using at least one of the following methods: simple deletion, partial character masking, or excluding the information itself from the search target. For example, if a field related to the vehicle owner exists in the performance inspection sheet data, the control unit prevents general users from identifying the owner by excluding the field during the search phase or processing it as blank during the screen output phase.

[0131] The vehicle performance information screen generated in this manner for public disclosure is configured to allow verification of key performance inspection results using only a single piece of information—namely, the vehicle license plate number input—while simultaneously being designed to protect personal information. This screen configuration method was derived by repeatedly comparing and verifying the differences between actual performance inspection data and the public screen; the display range of results by item and the criteria for anonymization processing can be established based on numerous inspection cases. Therefore, the above configuration is not an arbitrary setting but can be implemented through a method verified via actual data utilization and disclosure scenarios, and is concretized to a level sufficiently feasible for a person of ordinary technical expertise to implement the vehicle performance information disclosure system.

[0132] A control unit according to one embodiment of the present invention sequentially records shooting status information, including the time of shooting, shooting location coordinates, shooting angle, shooting target area, and the status of maintaining biometric authentication of the subject of shooting, throughout the entire shooting process in which the security processing image file is generated, at each shooting point.

[0133] A control unit according to one embodiment of the present invention is configured to sequentially record shooting status information at each shooting point from the time shooting begins to the time shooting ends, throughout the entire shooting process in which a security processing image file is generated. Here, the entire shooting process includes not only single image shooting but also continuous shooting, video shooting, and cases where shooting is paused and then resumed, and each shooting point can be classified by shutter input, frame capture, or video frame.

[0134] The control unit records the shooting time at each point in time. The shooting time is generated based on the clock of the user terminal or the internal system of the AI ​​security camera, and if connected to a network, it may use a time synchronized from a standard time server. The shooting time can consist of time information in units of year, month, day, hour, minute, and second, and in the case of continuous shooting or video recording, it can be configured to record the shooting time for each frame or at regular frame intervals. This method is intended to clearly distinguish the shooting order according to the flow of time within the same shooting session.

[0135] In addition, the control unit records the shooting location coordinates at each shooting point. The shooting location coordinates are based on GPS coordinates received via a location measurement means included in the user terminal or AI security camera, and may consist of pairs of latitude and longitude values. To account for cases where the GPS signal is temporarily weak, the most recently received coordinates may be used as a correction value, or the average of multiple coordinates may be calculated and recorded. This method of recording location coordinates enables subsequent verification of whether the footage of the same vehicle was actually taken at a specific location.

[0136] The control unit includes the shooting angle in the shooting status information. The shooting angle can be calculated using values ​​received from the camera's tilt sensor, gyroscope, or accelerometer, and can be expressed as an up-and-down angle relative to the horizontal plane, a left-and-right rotation angle, or a combination thereof. For example, if the camera is tilted at a certain angle toward the front of the vehicle, the corresponding tilt value is recorded as the shooting angle. This shooting angle information can be used to determine whether distortion occurs due to differences in shooting direction or viewpoint, even when photographing the same area.

[0137] In addition, the control unit records the target area as shooting status information. The target area refers to the part of the vehicle that the camera is pointing toward at the time of shooting, such as the vehicle exterior, specific panels, or undercarriage, and can be automatically identified based on predefined shooting procedure data or shooting stage information. For example, if the current stage in the shooting procedure is defined as 'left exterior of the vehicle,' the information for that stage is recorded as the target area. Alternatively, information regarding the area selected by the user on the shooting screen may also be recorded as the target area.

[0138] Furthermore, the control unit includes the biometric authentication maintenance status of the subject being filmed in the filming status information. The biometric authentication maintenance status refers to information indicating whether the subject being inspected maintains the authentication status as the same person even after the initial authentication during the filming process, and signifies the result of verifying the validity of the authentication at each filming point. For example, if the authentication is maintained by checking the facial recognition or fingerprint authentication status at regular intervals, it is recorded as valid; if the authentication status is deactivated or a discrepancy occurs, that status is recorded at the corresponding filming point. Through this, it can be confirmed solely from the filming records that no third party intervened during filming.

[0139] As such, shooting status information, including the time of capture, shooting location coordinates, shooting angle, target area, and biometric authentication maintenance status, is accumulated and recorded chronologically according to the sequence in which shooting proceeds. Each piece of shooting status information may be stored along with an index number, frame number, or shooting session identifier so that it can be linked to the captured image or video. This recording method has been verified through iterative testing based on actual shooting data and is designed to reflect all possible changes in time, location, and angle, as well as fluctuations in authentication status, that may occur during the shooting process.

[0140] Therefore, the sequential recording structure of the above-mentioned shooting status information corresponds to a specific implementation method that enables the reliability of the shooting result to be verified retrospectively, and is configured to be clear and feasible enough for a person skilled in the art to fully implement using a user terminal, a location information receiving means, a camera sensor, and an authentication means.

[0141] A control unit according to one embodiment of the present invention constructs a shooting progress sequence by combining a change in time interval between shooting points, a cumulative value of the distance traveled between shooting position coordinates, a change in the shooting angle, and a switching order of the shooting target area, based on the shooting state information.

[0142] A control unit according to one embodiment of the present invention constructs a shooting progress sequence based on shooting status information recorded sequentially. Here, shooting status information refers to a set of information including the shooting time, shooting location coordinates, shooting angle, and shooting target area recorded at each shooting point, and the shooting progress sequence refers to a data structure that expresses how such information has changed and accumulated in chronological order as a single continuous flow.

[0143] The control unit first calculates the trend of change in the time interval between shooting times. To this end, each shooting time included in the shooting status information is sorted in chronological order, and then the time difference between adjacent shooting times is calculated. For example, if the nth shooting time is t_n and the previous shooting time is t_(n-1), the time interval Δt_n between the two shooting times can be calculated as Δt_n = t_n - t_(n-1). By arranging these time interval values ​​according to the shooting order, a trend of change in the time interval is constructed, indicating whether shooting was performed continuously at a constant period or whether there were excessive gaps or abrupt changes in shooting frequency in specific sections.

[0144] Next, the control unit calculates the cumulative value of the distance traveled between shooting location coordinates. Each shooting location coordinate included in the shooting status information may consist of latitude and longitude, and the distance traveled is calculated using the difference in location coordinates between adjacent shooting points. For example, if the n-th shooting location coordinate is (lat_n, lon_n) and the previous shooting location coordinate is (lat_(n-1), lon_(n-1)), the distance traveled between the two coordinates can be calculated using a distance calculation formula based on the difference in coordinates. By accumulating and summing each calculated distance according to the shooting order, the cumulative value of the total distance traveled by the subject of the shot from the start of shooting until the current point in time is derived. This cumulative value can be used as an indicator to show whether shooting was performed repeatedly at the same location, or whether shooting proceeded while actually moving along the outer perimeter or surroundings of the vehicle.

[0145] In addition, the control unit calculates the direction of change of the shooting angle. The shooting angle refers to the tilt value or rotation angle recorded at each shooting point, and the control unit derives the amount of angle change by calculating the difference in shooting angles between adjacent shooting points. For example, if the n-th shooting angle is θ_n and the previous shooting angle is θ_(n-1), the amount of angle change Δθ_n can be calculated as Δθ_n = θ_n - θ_(n-1). At this time, by considering both the sign and magnitude of the amount of angle change, a direction of change is constructed that allows determining whether the shooting angle changes continuously in a specific direction or switches irregularly. This direction of angle change is utilized to determine whether the shooting was repeated in the same direction or whether the shooting was performed by naturally moving the viewpoint relative to the vehicle.

[0146] In addition, the control unit configures the transition sequence of the target areas. The target area refers to the parts of the vehicle recorded at each shooting point, and the control unit generates a transition flow of the target areas by listing this information according to the shooting order. For example, if the target areas transition sequentially as in front → left → rear → right, the transition sequence itself is recorded as a single sequence. Even if the same area appears repeatedly, the number of repetitions and the repetition intervals are reflected exactly according to the shooting order.

[0147] The control unit constructs a single shooting sequence by combining the calculated trends in time interval changes, accumulated values ​​of movement distances, directional changes in shooting angles, and the switching order of the target area. This shooting sequence is not merely a list of individual numerical values ​​or items, but rather corresponds to a data structure that expresses the entire shooting flow from the start to the end of shooting as a continuous change in time, space, direction, and target area. Since each component is calculated, accumulated, and aligned based on values ​​directly derived from actual shooting status information, no arbitrary criteria or abstract judgments are involved.

[0148] This method of configuring the shooting sequence is a structure confirmed to clearly distinguish between cases where shooting is performed continuously and cases where it is not when tested based on repetitive actual shooting data; it is configured to be specific and feasible enough for an ordinary technician to fully implement using information on the shooting time, position coordinates, angle, and shooting area.

[0149] A control unit according to an embodiment of the present invention verifies step by step based on whether the shooting sequence involves a shooting sequence involving continuous movement based on the outer perimeter of the vehicle, whether excessive repeated shooting occurs at the same shooting location, and whether the switching of the shooting target area does not conflict with a predefined inspection procedure sequence; and if any of the step-by-step verification processes of the shooting sequence do not satisfy the criteria, the entire shooting result is classified as not being data generated by normal performance inspection activities and controlled to be excluded from the vehicle status information generation step.

[0150] A control unit according to one embodiment of the present invention takes a previously configured shooting sequence as an input value and verifies in stages whether the corresponding shooting was performed by normal performance inspection activities at an actual vehicle site. Here, the shooting sequence refers to a data set in which changes in time intervals between shooting points, movement paths of shooting position coordinates, directionality of changes in shooting angles, and the switching order of shooting target parts are combined in chronological order, and the control unit sequentially applies a plurality of verification conditions based on this sequence.

[0151] First, the control unit verifies whether the shooting sequence involves continuous movement relative to the vehicle's outer perimeter. To this end, it considers both the cumulative distance traveled between shooting location coordinates and the direction of movement between adjacent shooting points. For example, if the distance traveled between adjacent shooting points repeatedly remains close to zero, or if a pattern of abrupt reversal in the direction of movement within a certain interval is repeated, it can be determined that it is difficult to view the subject of the shoot as having moved along the vehicle's outer perimeter while shooting. Conversely, if the shooting location coordinates show a pattern of gradual movement along the vehicle's length or width direction, and the cumulative distance traveled continuously increases according to the shooting sequence, it is determined that the shooting sequence involves continuous movement relative to the vehicle's outer perimeter. This determination can be established based on movement patterns confirmed in multiple actual measurement data obtained by moving around the vehicle at least once.

[0152] Next, the control unit verifies whether excessive repetitive shooting has occurred at the same shooting location. To this end, it calculates the number of times identical or error-bound shooting location coordinates appear consecutively or cumulatively within the shooting sequence. For example, if images captured at the same coordinates or within a set allowable radius account for more than a certain proportion of the total number of shots, it can be determined that the shooting was performed repeatedly at a specific location rather than moving shots in accordance with the actual inspection procedure. In this case, the allowable radius and the number of allowed repetitions may be values ​​derived from multiple experimental shooting data to distinguish between cases where the photographer captures shots from multiple angles at a single location at the actual inspection site and cases where arbitrary shooting is repeated without movement.

[0153] In addition, the control unit verifies whether the switching of the target areas for imaging does not conflict with the predefined inspection procedure sequence. The inspection procedure sequence refers to the order of target areas required during a general vehicle performance inspection process, such as the front, sides, rear, and undercarriage. The control unit compares the sequence of switching target areas included in the imaging sequence with this inspection procedure sequence. For example, if the predefined procedure is the order of front → left side → rear → right side → undercarriage, it is verified whether the imaging sequence does not significantly deviate from this order and demonstrates reasonable transitions between adjacent steps. If there is a frequent occurrence of reverse movement to the same area or switching of areas lacking procedural continuity, it can be determined that the imaging is inconsistent with the inspection procedure.

[0154] The control unit performs verification of continuous movement, verification of repeated shooting, and verification of switching of the target area in stages as described above. If any of these verifications fails to meet the established criteria, the entire shooting sequence is classified as data not generated by normal performance inspection activities. In this case, the control unit controls the exclusion of the corresponding shooting result during the subsequent stage of generating vehicle status information, thereby ensuring that data that does not satisfy the reliability of the inspection procedure, even if the shooting itself exists, is not used for determining the vehicle status.

[0155] This step-by-step verification method applies criteria established based on the results of comparing and analyzing actual field inspection shooting data with abnormal shooting data, and judgments are made based solely on directly observable data such as shooting location, movement path, and switching of the target area. Therefore, a person of ordinary skill in the art can implement the verification logic using shooting time, location coordinates, shooting angle, and shooting area information, and can practically determine the suitability of the shooting sequence without arbitrary estimation or unclear judgment factors.

[0156] A control unit according to one embodiment of the present invention is configured such that, only when the shooting sequence satisfies all of the criteria, the shooting result is confirmed as the result of an inspection action performed by an inspector moving around the vehicle at an actual site, and the confirmed result is used to automatically generate a vehicle performance inspection report, thereby determining whether the performance inspection report can be created based on whether the method of proceeding with the shooting action itself faithfully followed a predefined inspection procedure, regardless of whether the shooting images were simply collected.

[0157] A control unit according to an embodiment of the present invention confirms the shooting result as the result of an inspection action performed by an inspector moving relative to the vehicle at an actual site only when the previously generated shooting sequence satisfies all predefined criteria. Here, satisfying all criteria means that the change in time intervals between shooting points is not abnormally interrupted during the inspection process, the shooting location coordinates are confirmed to have moved continuously along the outer edge of the vehicle, and the direction of change in the shooting angle and the sequence of transitions of the shooting target parts substantially match the flow defined in the inspection procedure data. The control unit synthesizes the judgment results for each of these criteria to finally determine whether the entire shooting sequence was generated by a single continuous on-site inspection action.

[0158] Specifically, the control unit can be configured to record the pass status of each shooting step constituting the shooting sequence as a binary value, and to confirm the corresponding shooting result as a valid inspection result only when all steps have been passed. For example, the state transitions to a confirmed state only when verification of shooting position movement, verification of target area switching, and verification of repeated shooting are all passed; if any of these are judged to be unsatisfactory, the state does not transition to a confirmed state. This method reflects the results of analyzing only the cases where each verification item is simultaneously satisfied, based on multiple inspection shooting data normally performed in the actual field, and is configured to evaluate the overall process of the shooting activity itself rather than relying on a single criterion.

[0159] The control unit ensures that only confirmed shooting results are used for the automatic generation of the vehicle performance inspection report. In other words, regardless of whether captured images or videos simply exist or if a certain number have been collected, the criterion determining whether the performance inspection report can be generated is whether the shooting was conducted in faithful adherence to the actual inspection procedures. Consequently, even if the same number of captured images exist, if the shooting sequence does not meet the criteria, the images are not passed to the performance inspection report generation stage; conversely, they are utilized for generating the report only if all criteria are satisfied.

[0160] Since this configuration includes as judgment factors whether the temporal and spatial progression of the shooting activity, as well as the content of the shooting result itself, conforms to the inspection procedure, a person skilled in the art can easily implement this embodiment by configuring a shooting sequence using information on the time of shooting, location coordinates, shooting angle, and the shooting target area, and determining step-by-step whether each criterion is satisfied. In addition, by applying criteria derived from comparing actual field inspection data with abnormal shooting data, it is possible to prevent the shooting result from being arbitrarily selected or determined, and to ensure that the feasibility of creating a performance inspection checklist is determined based on the progression of the shooting activity itself.

[0161] A control unit according to one embodiment of the present invention generates a plurality of independent record paths based on the time of generation, generation order, and reference relationship of each information regarding vehicle status information, vehicle history information, and shooting sequence used in the automatic generation process of the vehicle performance inspection checklist, and performs internal consistency verification for each of the record paths based on whether the combination order of the vehicle number, shooting time, inspection subject identification information, and performance inspection checklist item code is the same.

[0162] A control unit according to an embodiment of the present invention generates multiple independent record paths based on when the vehicle status information, vehicle history information, and shooting sequence used in the automatic generation process of a vehicle performance inspection report were generated, in what order they were processed, and what reference relationships they have with one another. Here, a record path refers to a logical flow in which the generation times of each piece of information—such as the time when vehicle status information is generated, the time when vehicle history information is received, and the time when the shooting sequence is finalized—are arranged in chronological order for the same inspection case, and the connection relationships express which information is subsequently combined with or referenced. The control unit is configured not to use only a single path for a single inspection case, but to generate two or more record paths in parallel by independently reconstructing the generation times and reference relationships based on the same source data.

[0163] In this case, each record path uses the same vehicle number as the reference identifier, but explicitly includes the order in which the shooting time, inspection entity identification information, and performance checklist item code are combined. For example, one record path may have an order in which the vehicle number is determined first, followed by the shooting time, then the inspection entity identification information, and finally the performance checklist item code is mapped; in other record paths, a separate processing flow is reconfigured to ensure that the same elements are combined in the same order. For each of these record paths, the control unit uses not only the types of elements used for combination but also whether the combination order itself is identical as a criterion for internal consistency verification.

[0164] During the internal consistency verification process, the control unit compares the combination order included in each record path to sequentially check whether the shooting time is combined after the vehicle number, whether the inspection entity identification information is linked after the shooting time, and whether the performance checklist item codes correspond in the same order thereafter. In this process, if the combination order differs in any single record path, if specific elements are omitted, or if reference relationships are formed differently, it can be determined that an internal inconsistency exists between the relevant record paths. This determination can be established based on the results of confirming through multiple actual processing logs that record paths generated normally during the actual inspection data processing process repeatedly exhibit the same combination order.

[0165] In this way, the control unit does not merely compare the contents of a single data set, but reconstructs the processing flow itself—in which data is generated and combined—into multiple independent paths and verifies the identity of those paths. Accordingly, a person skilled in the art can easily implement this embodiment by generating a processing log that sequentially links the vehicle number, shooting time, inspection entity identification information, and performance checklist item code, and by independently repeating the same processing logic to compare the combination order between the generated record paths. This allows for clear verification of whether the order of data generation, reference, and combination has been consistently maintained during the automatic generation of the vehicle performance checklist, and effectively prevents errors caused by arbitrary data combinations or changes in the processing order from being reflected in the inspection results.

[0166] A control unit according to one embodiment of the present invention controls the performance checklist to be excluded from public disclosure by classifying it as not being a properly confirmed check result if it is determined that any one of the plurality of recording paths has a different creation order, reference relationship, or combination result for the same item.

[0167] Here, a difference in generation order refers to cases where the temporal sequence in which vehicle status information, vehicle history information, or shooting sequences are combined for the same item varies across different recording paths, and a difference in reference relationship refers to cases where the preceding information referenced by a specific item is connected differently depending on the recording path. Furthermore, a difference in combination result includes cases where, even if they correspond to the same vehicle number and the same performance inspection checklist item code, the components or arrangement state of the finally combined value do not match across different recording paths.

[0168] To determine whether such discrepancies exist, the control unit sequentially compares the processing results generated for the same item in each recording path. For example, if in one recording path vehicle status information is generated first, followed by vehicle history information and referenced shooting sequence, whereas in another recording path the shooting sequence is combined first and then vehicle status information is linked, the control unit can identify this as a discrepancy in the generation order. Additionally, if the shooting time or inspection entity identification information referenced by the same item is linked differently depending on the recording path, it can be determined as a discrepancy in the reference relationship; furthermore, it can be confirmed through the analysis of multiple processing logs that such discrepancies are patterns not repeatedly observed during the actual normal inspection process.

[0169] If the aforementioned discrepancy exists in any of the multiple recording paths, the control unit classifies the entire performance checklist as not being a properly confirmed inspection result. This is because a breakdown in the consistency of the processing flow in even a few recording paths implies the possibility of arbitrary intervention or abnormal processing during the process of generating and combining inspection results. Therefore, in such cases, the control unit controls the performance checklist to be excluded from public disclosure, ensuring that it is not displayed on public screens or in response to external inquiry requests.

[0170] This configuration uses not only the content of the performance checklist itself but also the identity of the internal processing path through which the content is generated as a verification criterion. A person skilled in the art can easily implement this embodiment by storing the generation order and reference relationships for each recording path in a log format and comparing them. By ensuring that the performance checklist is disclosed only when a normal processing flow is confirmed, it is possible to effectively prevent inconsistent data that could undermine the reliability of the inspection results from being provided externally.

[0171] A control unit according to one embodiment of the present invention controls the vehicle performance inspection table to be identified as a finalizable result and recorded in a storage unit in a finalized state only when it is determined that all of the plurality of recording paths have the same generation order and reference relationship, thereby configuring the performance inspection table to be disclosed only when the consistency of the internal processing paths in which the inspection results are generated, combined, and finalized is confirmed.

[0172] A control unit according to an embodiment of the present invention identifies the vehicle performance inspection sheet as a finalizable result only when, as a result of comparing a plurality of previously generated recording paths, it is determined that all recording paths have the same generation order and the same reference relationship. Here, the same generation order means that the temporal flow in which vehicle status information, vehicle history information, and the shooting progress sequence are generated and combined matches across all recording paths, and the same reference relationship means that the preceding information referenced by each information and the target of combination remain the same for each recording path. The control unit does not merely verify whether the final result value is the same, but uses as a judgment criterion whether the internal processing flow itself in which each information is generated, combined, and finalized is completely identical.

[0173] Specifically, the control unit sets the vehicle number as a reference identifier for each recording path, and then aligns the time at which vehicle status information is generated, vehicle history information is combined, and the shooting sequence is referenced in the order corresponding to the vehicle number on the time axis. Subsequently, it compares these alignment results between each recording path to verify whether the same information is combined in the same order. In addition, by comparing the reference relationships regarding which information the performance checklist item code is combined with at which stage, it is configured to identify even minute differences in the processing flow. This comparison process can be established based on creation and reference patterns that are repeatedly confirmed in the actual normal inspection processing logs.

[0174] Based on the comparison results described above, the control unit identifies the vehicle performance inspection sheet as a finalizable result only when it is determined that multiple recording paths all have the same generation order and reference relationship, and records it in the storage unit as a finalized state. Conversely, if even a part of the recording paths differs in generation order or reference relationship, it is not converted to a finalized state and may remain in the storage unit as an unfinalized state. Thus, the performance inspection sheet is managed as a finalized result only when the consistency of the internal processing paths through which the inspection result is generated, combined, and finalized is confirmed.

[0175] With this configuration, the disclosure of the performance checklist is not determined by the mere existence of data or the similarity of result values, but rather by whether the flow of the entire process in which the inspection results are internally processed has been consistently maintained. Therefore, a person skilled in the art can easily implement this embodiment by recording the creation time and reference relationships of each processing step in a log format, reconstructing them into multiple paths, and comparing them, and can control the disclosure of the performance checklist only when a normal processing flow is confirmed without arbitrary data combinations or changes in the processing order.

[0176] A control unit according to one embodiment of the present invention defines the pattern of how an inspection activity is repeated and changed over time as an inspection activity pattern, based on the shooting progress sequence, vehicle status information, and the frequency of shooting intervention by the inspection subject collected throughout the process of generating the vehicle performance inspection checklist.

[0177] A control unit according to one embodiment of the present invention manages the shooting progress sequence, vehicle status information, and the frequency of shooting intervention by the inspection subject collected throughout the process of generating a vehicle performance inspection checklist by arranging them in chronological order. Here, the shooting progress sequence refers to a set of time information in which the time when the inspection subject starts shooting while moving relative to the vehicle, the duration of shooting at each shooting location, and the time when shooting ends are recorded continuously. The vehicle status information refers to a set of status information directly corresponding to inspection items, such as the vehicle exterior, instrument status, and whether warning lights are illuminated, which are generated in conjunction with each shooting time. In addition, the frequency of shooting intervention by the inspection subject is defined as a value counted on the time axis of the number of times the inspection subject manually starts shooting or performs reshooting, the number of times the shooting angle is changed, and the number of times the shooting target is re-designated during the automatic shooting flow.

[0178] To map each of the above information to the same time standard, the control unit sets the shooting time as the reference time and sequentially corresponds changes in the shooting sequence, updates to vehicle status information, and the occurrence of shooting intervention events centered on that time. For example, if shooting images are generated at regular intervals after the start of shooting and manual intervention by the inspection entity occurs at a specific point in between, the shooting intervals before and after that point, the shooting target, and changes in vehicle status information are recorded together. This time-aligned data is used as foundational data to indicate that the inspection activity is not a set of single operations, but a continuous activity that is repeated and changes over time.

[0179] The control unit analyzes changes in the shooting progression sequence, vehicle status information, and shooting intervention frequency aligned on the time axis to identify shooting intervals that appear repeatedly within a certain range, updates of the same type of vehicle status information, and similar shooting intervention patterns as a single group. In this case, repetition can be determined based on whether identical or similar shooting intervals appear consecutively or whether vehicle status information corresponding to the same inspection item is repeatedly generated in a specific order, while patterns of change can be determined based on increases or decreases in shooting intervals, abrupt changes in the frequency of shooting intervention, or shifts in the update timing of specific vehicle status information. Through this combination of repetition and change, the temporal flow of the entire inspection activity is structured into a single continuous pattern.

[0180] Specifically, the control unit calculates the time difference between adjacent shooting points in the shooting sequence and extracts sections where that time difference repeats within a certain range. It also checks whether the generation order and item composition of vehicle status information are repeated within the same section, and simultaneously comprehensively considers whether the frequency of the inspection entity's shooting intervention is maintained within a specific range or changes in stages. The arrangement of repeating and changing sections derived in this manner is defined as a single inspection behavior pattern that indicates how the inspection activity was actually performed.

[0181] These inspection behavior patterns can be established based on the temporal and behavioral flows commonly observed in numerous inspection cases repeatedly performed in an actual inspection environment, and can be generated by directly utilizing shooting logs and vehicle status information logs. Therefore, a typical technician can implement inspection behavior patterns using only the shooting time, number of shots, the point in time when vehicle status information is generated, and records of shooting intervention events, without the need for separate abstract criteria or arbitrary judgments. In this way, the repetitive and changing aspects of inspection behavior over time are defined as a concrete data flow, and this definition is configured to be reproducible based on data collected during the actual inspection process.

[0182] A control unit according to one embodiment of the present invention calculates a pattern stability index for the inspection behavior pattern based on the change in the distribution of intervals between shooting times, the temporal variation in the frequency of switching of the shooting target area, and the degree to which the timing of intervention by the inspection subject is concentrated in a specific section.

[0183] A control unit according to one embodiment of the present invention uses time flow data constituting a previously defined inspection behavior pattern as an input value to calculate a pattern stability index indicating how uniformly the pattern is maintained over time. Here, the inspection behavior pattern consists of a data set in which the sequence of shooting times, the sequence of switching of the parts subjected to shooting, and the time of the inspection subject's intervention in shooting are aligned on a single time axis.

[0184] The control unit first calculates the distribution change of intervals between shooting points. To this end, it generates an array of shooting intervals by sequentially calculating the time difference between two adjacent shooting points, and calculates a variance value based on the average interval of the array and how much each shooting interval deviates from the average. If the variance of the shooting intervals is small, it is determined that shooting was performed in a constant rhythm; if the variance is large, it is determined that shooting was performed irregularly. Since this calculation can be implemented using only the timestamps recorded in the shooting log, a person of ordinary skill can reproduce it as is without any separate abstract criteria.

[0185] Next, the control unit calculates the temporal variation in the frequency of switching between target areas. To this end, it identifies the target area for each consecutive image and records the point in time when the image switches from the same area to another. Subsequently, it aggregates the number of switching intervals and compares the differences in switching frequency across each interval. If switching is excessively concentrated in a specific interval, or conversely, if no switching occurs for an extended period, the shooting flow is determined to be uneven. Since this switching frequency data can be calculated solely by combining target area identification information and shooting time information, the implementation process is clear.

[0186] In addition, the control unit calculates the degree to which the timing of the inspection entity's intervention is concentrated in specific segments. The timing of the inspection entity's intervention is defined as the time when events such as manual recording initiation, re-recording, or adjustment of the recording angle occur. The entire inspection time segment is divided into intervals of a fixed length, and the number of intervention events for each segment is calculated. Subsequently, by calculating the ratio of the number of interventions concentrated in specific segments relative to the total number of interventions, it is determined whether the interventions are evenly distributed over the entire time or clustered at specific points in time. This can also be derived solely from intervention event logs and time information.

[0187] The control unit quantifies the changes in the distribution of intervals between shooting times, the temporal fluctuations in the frequency of switching target areas, and the degree of bias in the timing of inspection agent intervention, all calculated as described above, and then combines these values ​​to generate a stability index for the inspection activity pattern. For example, if the values ​​calculated for each item fall within a certain range, the pattern is judged to be temporally stable; conversely, if abrupt fluctuations are observed in one or more items, the stability of the pattern may be judged to be low. These criteria can be applied directly using verified range values ​​based on shooting logs and intervention logs repeatedly collected during the actual inspection process.

[0188] Since this process for calculating stability indicators is constructed by directly utilizing the statistical distributions of time and event data generated in actual inspection environments, ordinary technicians can implement it in the same manner without separate training data or abstract models. Consequently, the stability of inspection activity patterns is defined not by arbitrarily set standards, but by a value derived by directly reflecting the temporal characteristics of actual footage and inspection data.

[0189] A control unit according to one embodiment of the present invention controls the vehicle performance inspection checklist to be maintained in an unconfirmable state by classifying the inspection behavior pattern as different from the stable behavior pattern observed in normal field inspection activities when the stability index changes rapidly over a short period or exhibits a pattern of discontinuous distortion in a specific section.

[0190] A control unit according to one embodiment of the present invention arranges the previously calculated stability index in chronological order and continuously tracks how the stability index changes during the inspection process. Here, since the stability index is a value already quantified by reflecting the distribution of intervals between shooting times, the frequency of switching of the target area being photographed, and the degree of bias in the timing of the inspection agent's intervention, the control unit only needs to additionally analyze the temporal change of the index.

[0191] The control unit sequentially calculates the difference in stability indicators between two consecutive points in time to determine whether the stability indicator changes rapidly over a short period. For example, if the difference between stability indicator values ​​calculated at regular time intervals occurs repeatedly and significantly, it is determined that the method of performing the inspection activity has changed rapidly within a short period. Since this determination of rapid change can be implemented simply by comparing the differences between continuous values, a person of ordinary skill can reproduce it without complex models or learning processes.

[0192] In addition, the control unit determines whether the stability index exhibits a pattern of discontinuous distortion in a specific section. To this end, the entire inspection time interval is divided into multiple continuous time intervals, and the flow of stability index values ​​calculated in each interval is compared. If the stability index values ​​between adjacent intervals do not continue continuously but rise or fall abruptly, it is determined that the inspection activity in that interval was performed differently from the general field inspection flow. This determination of discontinuity can also be implemented solely by comparing the index for each time interval.

[0193] If such abrupt changes or discontinuous distortions are identified, the control unit classifies the corresponding inspection behavior pattern as different from the stable behavior patterns observed in normal field inspection activities. In this context, the stable behavior patterns observed in normal field inspection activities can be based on patterns identified as cases where stability indicators change relatively gradually, derived from recording logs and intervention logs collected during multiple actual inspection processes. In other words, the judgment is made by applying the range of changes repeatedly observed in actual inspection data directly, rather than relying on separate assumptions or abstract criteria.

[0194] If the control unit classifies the inspection activity pattern as unstable as described above, it controls the vehicle performance inspection checklist to remain in an unconfirmed state rather than confirming the corresponding inspection result as a reliable field inspection result. Accordingly, regardless of whether sufficient images have been collected, if the temporal flow and stability of the inspection activity do not correspond to normal field inspection activities, the performance inspection checklist is processed so that it is not confirmed.

[0195] Since this control method makes judgments based solely on the temporal changes of stability indicators, it is clear that the stability indicators accurately reflect the flow of data generated during the actual inspection process. Therefore, a skilled technician can implement the same judgment process using only the shooting time, information on the captured part, and logs of the inspector's intervention; furthermore, it is evident that the classification of inspection behavior patterns and the control of whether the performance checklist is finalized can be realized based on actual data.

[0196] A control unit according to one embodiment of the present invention controls the transition to a performance checklist confirmation stage by identifying the inspection activity pattern as a result of inspection activities performed at the actual site only when the stability index shows a gradual change or a repetitive convergence pattern throughout the entire inspection process, thereby configuring the system to determine whether the performance checklist can be confirmed based on the stability of the temporal and behavioral patterns appearing throughout the inspection activity.

[0197] A control unit according to one embodiment of the present invention continuously aligns the previously calculated stability index along the entire time axis from the start of the inspection to the end of the inspection, and determines whether the change pattern of the stability index is maintained gradually or converges repeatedly. Here, a gradual change refers to a state in which the stability index gradually increases or decreases over time, but the difference between adjacent points in time does not rapidly expand, and a repetitive convergence pattern refers to a state in which, even if the stability index oscillates within a certain range, it repeatedly exhibits a tendency to gradually return to the same value or a value close to it.

[0198] To this end, the control unit calculates the differences between consecutive stability index values ​​in chronological order and verifies whether these difference values ​​do not increase intensively in specific sections of the entire inspection period. For example, if the difference between consecutive stability index values ​​is repeatedly maintained within a certain range or gradually decreases, it can be determined that the entire inspection process was performed according to the same operational rhythm and procedure. Since this determination is possible through simple value comparison and cumulative observation alone, it can be implemented without complex prediction models or learning processes.

[0199] In addition, the control unit verifies whether the stability index exhibits a pattern of repeated convergence over time. To this end, the entire inspection process is divided into time intervals of a fixed length, and the average or representative value of the stability index calculated in each interval is compared. If a pattern repeats where the representative values ​​between adjacent time intervals do not deviate significantly and return to a similar range, it is determined that the inspection activity was maintained in a consistent manner without being disrupted by accidental actions at a specific point in time.

[0200] The control unit identifies the inspection behavior pattern as the result of inspection activities performed by an inspector moving around the vehicle in the actual field, only when the stability indicator exhibits a gradual change or a pattern of repetitive convergence throughout the entire inspection process as described above. This judgment is based not merely on the existence of captured data, but on whether the timing of the capture, the switching of the target area, and the frequency of the inspector's intervention flowed naturally over time.

[0201] In this way, the control unit controls the transition to the performance checklist confirmation stage only when identified as a result of actual field inspection activities. Therefore, the performance checklist moves to a state where it can be confirmed only when the stability of the temporal and behavioral patterns appearing throughout the inspection activities is secured, and it does not transition to the confirmation stage if the pattern of change in the stability index is irregular or skewed to a specific section.

[0202] Since this control method is based on the results of a chronological analysis of the actual shooting sequences and intervention logs collected during the inspection process, a standard technician can implement the same judgment process using only the same log data. In other words, the assessment of the stability of the inspection activity is derived step-by-step based on the actual data flow, and the system is specifically configured to enable the determination of whether to finalize the performance checklist based on the stability of the inspection activity pattern.

[0203] A control unit according to one embodiment of the present invention identifies as a disturbance event a case in which a change in the external environment or an interruption of the inspection flow occurs at the time of intervention by the inspection subject, the shooting progress sequence, vehicle status information, and the inspection subject during the process of forming the inspection behavior pattern.

[0204] A control unit according to one embodiment of the present invention observes the shooting progress sequence, vehicle status information, and the point of intervention of the inspection subject in chronological order, which are generated during the entire process in which an inspection activity pattern is formed, and detects sections where the flow of these information is not maintained continuously. Here, the shooting progress sequence refers to a series of records in which the time interval between shooting points, the relationship of movement of shooting positions, and the sequence of transitions of the shooting target parts are connected on a time axis, and the vehicle status information includes the continuity of status values ​​generated at the same point in time or adjacent points in time.

[0205] The control unit detects cases where the interval between shooting points in the shooting sequence increases rapidly compared to usual, or where the shooting position coordinates remain unchanged for a certain period of time. This can be interpreted as reflecting a situation where the shooting activity has been interrupted due to the influence of the external environment or where the inspection flow has been temporarily broken. For example, if a section occurs where the continuous shooting interval expands to more than twice the average interval, or if the same shooting position is repeatedly maintained without transitioning to another target area, it can be determined that the natural progression of the inspection flow has been hindered.

[0206] In addition, the control unit checks whether the shooting sequence and vehicle status information change simultaneously based on the point of intervention by the inspection entity. If no significant change occurs in the shooting target area, shooting angle, or vehicle status information despite the intervention by the inspection entity, the intervention may be determined to be a interruption or disturbance caused by external factors, rather than a normal part of the inspection activity. Conversely, if the shooting flow resumes or switches to a different pattern after the point of intervention, it may also be interpreted as the inspection flow not being consistently maintained.

[0207] In this way, the control unit comprehensively compares the temporal continuity of the shooting sequence, the flow of changes in vehicle status information, and pattern changes before and after the point of intervention by the inspection entity to identify sections where changes in the external environment or interruptions in the inspection flow occur as disturbance events. Since this identification is achieved by simply comparing the time when each piece of information was actually generated with its preceding and subsequent relationships, it can be implemented without a separate, complex reasoning process.

[0208] Sections identified as disturbance events are managed separately from the normal flow during the formation of inspection activity patterns, and can subsequently be utilized as reference information when evaluating the stability of the inspection activity patterns or determining whether to finalize the performance checklist. Due to this configuration, changes in the external environment or discontinuous flows occurring during the inspection process are clearly distinguished based on actual data, allowing even a skilled technician to perform the disturbance event identification process exactly as is using the same recorded information.

[0209] A control unit according to one embodiment of the present invention compares inspection behavior patterns before and after the occurrence of the disturbance event and determines whether the pattern is restored based on whether the shooting interval, the sequence of switching the shooting target area, and the frequency of intervention by the inspection subject converge back to the pattern range prior to the disturbance.

[0210] A control unit according to one embodiment of the present invention separates and compares, in chronological order, an inspection behavior pattern formed before a disturbance event occurs and an inspection behavior pattern formed again after the disturbance event ends. Here, the inspection behavior pattern prior to the disturbance event is a reference pattern calculated during a period in which a normal shooting sequence was maintained, and refers to a pattern formed by repeating the shooting interval, the order of switching the shooting target area, and the frequency of intervention by the inspection subject within a certain range.

[0211] The control unit first collects interval data between shooting points in the section prior to the disturbance, and calculates the average and variance values ​​of the intervals to set a reference shooting interval range. For example, if the continuous shooting intervals are defined as t₁, t₂, …, t, the average interval is calculated as (t₁+…+t) / n, and the variance can be calculated based on the sum of the squared differences between each interval and the average interval. Based on the average and variance calculated in this way, the range in which the shooting intervals prior to the disturbance were typically distributed is defined as the reference range.

[0212] Subsequently, the control unit sequentially checks whether the newly collected shooting interval data in the interval following the disturbance event re-enters the aforementioned reference range. If a certain number of shooting intervals or more after the disturbance fall within the reference range, it can be determined that the pattern in terms of shooting intervals is reconverging to the state prior to the disturbance. Since this process is performed simply by comparing each shooting interval value with the reference range, a person skilled in the art can implement it without a separate, complex model.

[0213] Regarding the sequence of target area transitions, the transition sequences observed in the pattern prior to disturbance are listed in chronological order, and a comparison is made based on whether the sequence appears repeatedly. For example, if a transition sequence such as outer → front → lateral → rear was repeatedly identified prior to disturbance, it is checked whether the target area transitions occur in the same or a similar sequence after disturbance. If the transition sequence after disturbance matches the reference sequence by a certain percentage or more, it can be determined that pattern recovery has occurred in terms of target area transitions as well.

[0214] Regarding the frequency of intervention by the inspection entity, a reference frequency is established by aggregating the number of interventions that occurred per unit of time in the period prior to the disturbance, and whether the number of interventions in the same unit of time after the disturbance returns to the reference frequency range is compared. For example, if an average of 1 to 2 interventions occurred per minute prior to the disturbance, the recovery of the pattern can be determined by checking whether the same level of intervention frequency is maintained after the disturbance.

[0215] The control unit compares the pattern range formed prior to the disturbance with the data after the disturbance for each item—such as the shooting interval, the sequence of switching target areas, and the frequency of intervention by the inspection agent—and determines that pattern recovery has occurred if at least a certain proportion of the multiple items converges back to the reference pattern range. Since this determination is performed using the time information and occurrence frequency data collected during the actual inspection process, it is possible to clearly verify, based on data, whether the inspection behavior pattern before and after the disturbance event has returned to the actual field inspection flow.

[0216] A control unit according to one embodiment of the present invention controls the vehicle performance inspection checklist to be maintained in an unconfirmable state by classifying the inspection activity as an abnormal inspection flow if the inspection activity pattern does not recover to the existing pattern range even after a disturbance event, or if discontinuous changes in the shooting order or continuous omission of shooting of specific parts occur during the recovery process.

[0217] A control unit according to one embodiment of the present invention continuously checks, in chronological order, whether the inspection behavior pattern after a disturbance event recovers to an existing pattern range formed prior to the disturbance. Here, the existing pattern range refers to a state in which the distribution of shooting intervals, the continuity of the sequence of switching of the target area being photographed, and the frequency of intervention by the inspection subject, which were repeatedly observed in the section prior to the disturbance, were stably maintained within a certain range. This is defined by aggregating the shooting time, identification information of the target area being photographed, and the time of intervention collected during the actual inspection process.

[0218] The control unit compares the inspection behavior patterns formed during a certain period following the termination of the disturbance event with the existing pattern range to determine whether the shooting interval re-enters the average value and allowable deviation range, whether the switching sequence of the target areas follows the order repeatedly verified prior to the disturbance, and whether the frequency of intervention by the inspection entity recovers to the level prior to the disturbance. In this case, the recovery of the shooting interval is determined by calculating the time difference between consecutive shooting points and comparing it simply with the reference range, while the switching sequence of the target areas is verified by sequentially comparing it with the reference sequence defined prior to the disturbance.

[0219] If the shooting interval does not return to the standard range even after a disturbance event, or if discontinuous changes unrelated to the existing sequence repeatedly occur during the process of switching target areas, the control unit determines that the inspection activity has failed to restore the normal inspection flow. Additionally, if a specific target area is not continuously shot more than a certain number of times in the section following the disturbance, despite being mandatory for inclusion in the standard inspection procedure, it is identified that a continuous omission of shooting that area has occurred.

[0220] The control unit classifies an inspection activity as an abnormal inspection flow that deviates from the normal field inspection flow if the inspection activity pattern fails to recover to the existing pattern range even after a disturbance event, or if discontinuous changes in the shooting sequence or continuous omissions in shooting specific parts are confirmed during the recovery process. Based on this classification result, the vehicle performance inspection checklist generated from the relevant inspection activity is controlled to remain in an unconfirmable state rather than being converted to a confirmable state; this configuration ensures that the performance inspection checklist is not confirmed when the continuity and completeness of the inspection procedure are not verified at the data level.

[0221] A control unit according to one embodiment of the present invention controls the transition to a performance checklist confirmation stage by identifying the inspection activity as an inspection activity maintained normally despite external influences only when it is determined that the inspection activity pattern recovers to the existing pattern range even after a disturbance event, thereby determining whether the performance checklist can be confirmed based on whether the inspection activity pattern is stably restored even after a disturbance.

[0222] A control unit according to one embodiment of the present invention sets an inspection behavior pattern formed in the section prior to the disturbance as a reference pattern in order to objectively determine whether the inspection behavior pattern after a disturbance event recovers to the existing pattern range. This reference pattern is defined by utilizing the shooting time information, records of switching of the shooting target area, and the point of intervention of the inspection subject, which were sequentially collected during the actual inspection process prior to the disturbance, and is composed of a range derived from the results of repeated observation of actual field inspection data, rather than a specific algorithm or abstract standard.

[0223] The control unit compares the inspection behavior pattern following the termination of the disturbance event with the reference pattern defined prior to the disturbance in chronological order. In this process, the intervals between shooting points are calculated by directly determining the difference between consecutive shooting times and organized into an array; it then verifies whether the values ​​in this array re-enter the shooting interval range observed in the reference pattern. For the switching of target areas, the sequence of area switching that appeared repeatedly prior to the disturbance is set as the reference sequence, and the system sequentially compares this to determine if the same sequence is continuously formed again in the post-disturbance period. Similarly, the timing of the inspection entity's intervention is determined using a simple aggregation method, based on the intervention frequency and timing distribution observed prior to the disturbance, to determine whether it recovers to the same level in the post-disturbance period.

[0224] The control unit synthesizes these comparison results and identifies the inspection activity as having been maintained normally despite external influences only if the shooting interval, the sequence of switching between target areas, and the frequency of intervention by the inspection agent all recover to the reference pattern range even after the disturbance event. In this process, no separate estimations or assumptions are used, and the judgment is made based solely on the results of a direct correspondence comparison between the actual inspection data before and after the disturbance.

[0225] In this manner, the control unit controls the transition of the inspection activity to the performance checklist confirmation stage only when it is determined that the inspection activity pattern has been stably restored even after the disturbance. Conversely, if recovery to the reference pattern is not confirmed, the system is maintained so that it does not transition to the confirmation stage; thus, the system is configured to determine whether the performance checklist can be confirmed based on whether the inspection activity pattern has recovered the stable behavior observed in actual field inspections even after the external disturbance. Explanation of the symbols

[0226] 100: User terminal 110: User Input Section 120: Display section 200: Server 210: Communications Department 220: Storage section 230: Control unit

Claims

Claim 1 A platform system for generating vehicle performance checklists based on AI security cameras and disclosing information to the public comprises: a user terminal equipped with a display unit and a user input unit; and a server connected to the user terminal via a wired or wireless network, wherein the server comprises a communication unit and a storage unit. and a control unit; wherein the control unit, when a mechanic or performance inspector accesses through the user terminal, performs an authentication procedure using biometric information to identify the inspector, generates identity information including a license number, user identifier, access time, and access location information corresponding to the identified inspector, and stores it in a storage unit; during the process of photographing a vehicle through an AI security camera linked to the user terminal, generates shooting information by combining the vehicle license plate recognition result, GPS coordinates of the shooting location, the shooting time, and the identity information, and creates a security processing image file by inserting the shooting information into the pixel data area of ​​the captured image or video using a digital watermarking method; after the start of shooting, identifies the shooting sequence, shooting angle, shooting area, and shooting conditions step by step according to predefined inspection procedure data, judges the shooting quality by comparing the illumination, focus, shake, and field of view of the captured image at each step with reference data, outputs shooting guidance information to the user terminal to induce re-shooting if the judgment result regarding the shooting quality does not meet a pre-set standard, and includes the vehicle body exterior, suspected accident area, sheet metal or paint traces, and undercarriage condition from the captured security processing image file. The system is configured to identify vehicle status information using an AI image recognition method and simultaneously receive supplementary opinions through input from the inspection entity to generate inspection data merged with the vehicle status information, and the control unit receives vehicle history information including vehicle registration information, accident history, and inspection history by linking with at least one of the Ministry of Land, Infrastructure and Transport’s automobile information system, the Korea Insurance Development Institute’s accident history system, and the vehicle registration information system based on the vehicle license plate recognition result.The above vehicle condition information and the above vehicle history information are matched by item, and a status value for each item is calculated by matching them with each standard item code of the Automobile Performance Inspection Checklist notified by the Ministry of Land, Infrastructure and Transport, which corresponds to the condition of the vehicle body exterior, accident traces, and the condition of major structural parts; if the calculated status value differs from the information in the public history, the corresponding item is identified as a discrepancy item and classified as a peculiarity, and the classification result is automatically entered into the performance inspection checklist item; the Automobile Performance Inspection Checklist is automatically generated by combining the real name and qualification information of the inspection subject and the shooting result with digital watermarking applied; after the generated performance inspection checklist is stored in the storage unit, when a general user searches by entering a vehicle number, a vehicle performance information screen for public disclosure is generated that includes the item-by-item results of the performance inspection checklist, the real name information of the inspection subject, and the shooting-based fact verification result, while controlling the output so that the vehicle owner's personal information is anonymized; and the control unit sequentially generates shooting status information including the shooting time, shooting location coordinates, shooting angle, shooting target part, and the status of maintaining the biometric authentication of the shooting subject at each shooting point throughout the entire shooting process in which the security processing image file is generated. Recording, and based on the above shooting status information, constructing a shooting progression sequence by combining the trend of change in time intervals between shooting points, the cumulative value of the distance traveled between shooting location coordinates, the directionality of change in shooting angle, and the switching order of the shooting target area; verifying step by step based on whether the above shooting progression sequence is a shooting sequence involving continuous movement based on the vehicle outer perimeter, whether excessive repeated shooting has occurred at the same shooting location, and whether the switching of the shooting target area is inconsistent with the predefined inspection procedure sequence; and if any of the step-by-step verification processes of the above shooting progression sequence do not satisfy the criteria, classifying the entire shooting result as not being data generated by normal performance inspection activities, and controlling it to be excluded from the vehicle status information generation stage.An AI security camera-based vehicle performance inspection report generation and public information disclosure platform system characterized by being configured such that, only when the above-mentioned shooting sequence satisfies all of the above-mentioned criteria, the above-mentioned shooting result is confirmed as the result of an inspection action performed by an inspection entity moving around the vehicle at the actual site, and the confirmed result is controlled to be used for the automatic generation of a vehicle performance inspection report, thereby determining whether the performance inspection report can be created based on whether the method of proceeding with the shooting action itself faithfully followed a predefined inspection procedure, regardless of whether the shooting images were simply collected. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 In claim 1, the control unit generates a plurality of independent record paths based on the time of generation, generation order, and reference relationship of each information regarding vehicle status information, vehicle history information, and shooting sequence used in the automatic generation process of the vehicle performance inspection checklist, and for each of the record paths, performs internal consistency verification based on whether the combination order of the vehicle number, shooting time, inspection subject identification information, and performance inspection checklist item code is the same, and if any of the plurality of record paths is determined to have different generation orders, reference relationships, or combination results for the same item, controls the performance inspection checklist to be classified as not being a normally confirmed inspection result and excluded from public disclosure, and only when it is determined that all of the plurality of record paths have the same generation order and reference relationship controls the vehicle performance inspection checklist to be identified as a confirmable result and recorded in a storage unit in a final confirmed state, so that the performance inspection checklist is disclosed only when the consistency of the internal processing paths in which the inspection result is generated, combined, and confirmed is confirmed. Claim 6 In claim 5, the control unit defines the pattern of repeated and changing inspection activities over time as a single inspection activity pattern based on the shooting progress sequence, vehicle status information, and the frequency of the inspection subject's shooting intervention collected throughout the process of generating the vehicle performance inspection checklist; calculates a stability index of the pattern based on the change in the distribution of intervals between shooting points, the temporal variation in the frequency of switching of shooting target parts, and the degree to which the timing of the inspection subject's intervention is concentrated in a specific section for the inspection activity pattern; if the stability index shows a pattern of rapid change in a short period or discontinuous distortion in a specific section, it classifies the inspection activity pattern as different from the stable activity pattern observed in normal field inspection activities and controls the vehicle performance inspection checklist to remain in an unconfirmed state; and only when the stability index shows a pattern of gradual change or repetitive convergence throughout the entire inspection process, it controls the inspection activity pattern to be identified as the result of inspection activities performed in the actual field and transition to the performance inspection checklist confirmation stage, thereby configuring the system so that the possibility of confirming the performance inspection checklist is determined based on the stability of the temporal and behavioral patterns appearing throughout the inspection activity. AI security camera-based vehicle performance checklist generation and public information disclosure platform system featuring

Citation Information

Patent Citations

  • Intelligent performance inspection service system and method for used automobiles

    JP2023553497A

  • Car tire outfit auto test method using ai deep learning method

    KR1020220104857A

  • System and method for selling customized used car based on legal performance inspection

    KR102736210B1

  • Operating method of the diagnostic system for automobiles

    KR102866257B1