System for Automatically Inputting Document Classification and Text Extraction Results Based on Mobile Capture into a Hotel PMS Using RPA

Abstract

The present invention relates to a system for automatically inputting results of mobile-based document automatic classification and text extraction into a hotel PMS system in conjunction with RPA. The system comprises: a document receiving unit that receives image data containing a document from an external terminal and collects and records shooting time information (EXIF metadata) included in the file structure of the image data together with the image data; a recognition extraction unit that identifies the type of document based on the image data received from the document receiving unit and structures and extracts guest identity information from a text area within the image data by applying a recognition logic corresponding to the identified document type; and a verification unit that provides the guest identity information extracted by the recognition extraction unit to a user and determines whether to input the identity information into the hotel guest management system based on user confirmation input, wherein the verification unit controls the input of identity information by referring to the shooting time information (EXIF metadata) received from the document receiving unit so that the identity information can be input only when the image data was captured within a preset valid time range. The present invention includes an input processing unit that performs information input by transmitting the identity information through an application programming interface (API) of a hotel management system (PMS) or by automatically entering the identity information into an input field of the hotel management system, only when the input execution is determined by the verification unit. According to the present invention, by analyzing and processing objective image data such as passports, identification cards, and reservation confirmations using AI OCR (Optical Character Recognition) technology and RPA (Robotic Process Automation) technology, a customized automation solution for the hotel check-in process can be provided. This reduces unnecessary manual input processes while increasing the data accuracy required by the hotel, thereby significantly reducing guest waiting times and improving the work efficiency of hotel staff.

Description

Technology Field

[0001] The present invention relates to a system for automatically inputting results of automatic document classification and text extraction based on mobile camera capture into a hotel PMS system in conjunction with RPA. More specifically, it relates to a system for receiving document images, such as passports, identification cards, or reservation confirmations, captured through a mobile terminal device, automatically classifying the type of document and extracting identity information, and then automatically inputting the extracted information into a hotel PMS program after user verification. Background Technology

[0002] Accommodation facilities such as hotels, lodging establishments, and resorts must perform the essential task of verifying guest identity information through various documents, including passports, ID cards, and reservation confirmations, and inputting this data into a Property Management System (PMS) for check-in and accommodation management. This identity information entry is a necessary procedure in the operation of accommodation facilities, serving functions such as room allocation, bill settlement, access control, and compliance with legal reporting obligations.

[0003] Traditionally, it has been common practice for hotel front desk staff to personally collect passports or identification cards from guests, visually verify the information such as the guest's name, nationality, passport number, and date of birth, and manually input it into the PMS. However, with the recent proliferation of mobile devices, methods in which guests photograph documents using their own terminals or submit them as electronic documents are becoming increasingly widespread. Consequently, there is a growing demand for technologies that process information based on document images.

[0004] However, conventional guest information input and management methods face limitations in operational efficiency when there is a large volume of guests, as the information is not automatically entered into the hotel management system even when text is extracted from document images, and front desk staff often have to verify and manually input it. Furthermore, the lack of organic linkage between the user verification process for document recognition results and the automatic input process increases the possibility of input errors. In environments where a large number of documents must be processed simultaneously, such as during group visits or peak seasons with high reservation volumes, this can lead to issues such as increased workload on the front desk and longer customer waiting times.

[0005] Accordingly, there is a growing need for technology that can efficiently and reliably automate guest information entry tasks by automatically classifying document types based on images captured or submitted via mobile devices, extracting identity information, and automatically inputting it into the hotel management system after user verification. Prior art literature

[0006] Republic of Korea Registered Patent No. 10-1809362 The problem to be solved

[0007] The objective of the present invention is to solve the aforementioned conventional problems by providing an automatic input system for a hotel guest management program that can efficiently process multiple hotel guest documents by automatically identifying the type of document from document images captured via a mobile terminal or provided by a guest, extracting identity information, and then automatically inputting the extracted information into a hotel management system after user verification. means of solving the problem

[0009] The above objective is achieved by an automatic input system for a hotel guest management program, comprising: a document receiving unit that receives image data containing a document from an external terminal and collects and records shooting time information (EXIF metadata) included in the file structure of the image data together with the image data, according to the present invention; a recognition extraction unit that identifies the type of document based on the image data received from the document receiving unit and structures and extracts guest identity information from a text area within the image data by applying recognition logic corresponding to the identified document type; a verification unit that provides the guest identity information extracted by the recognition extraction unit to a user and determines whether to input the identity information into a hotel guest management system based on user confirmation input, wherein the verification unit controls the input of identity information only when the image data is captured within a preset valid time range by referring to the shooting time information (EXIF metadata) received from the document receiving unit; and an input processing unit that performs information input by transmitting the identity information through an application programming interface (API) of a hotel management system (PMS) or by automatically entering the identity information into an input field of the hotel management system, only when the input is determined to be performed by the verification unit.

[0010] In addition, the document receiving unit may be configured to allow a guest to directly transmit an image of the document captured by their own terminal to the server through an input page corresponding to a QR code provided to the guest terminal.

[0011] In addition, the document receiving unit may be configured to process image data containing multiple documents received simultaneously from multiple external terminals sequentially or in parallel.

[0012] In addition, the input processing unit may be configured to mask or delete some of the sensitive information among the identity information according to the input policy or settings of the hotel management system.

[0013] Additionally, the recognition extraction unit may be configured to perform one or more image quality correction processes, such as tilt correction, distortion correction, resolution correction, or noise removal, on a document image, identify the type of document, and if the reliability of the extraction result of the identity information is less than a preset threshold value, perform re-correction processing on the document image. Effects of the invention

[0014] According to the present invention, by analyzing and processing objective image data such as passports, identification cards, and reservation confirmations using AI OCR technology and RPA technology, an automated system optimized for the hotel check-in process can be provided. In particular, by reducing unnecessary manual input processes for individuals while accurately collecting guest personal information data required by the hotel, guest waiting times and hotel staff work efficiency can be improved.

[0015] In addition, by using a volatile memory processing method that ensures no guest information remains in relay servers and databases, and by comparing input completion signals in real time to exclude image or text data that could cause the leakage of guests' personal information from terminals and servers in advance, it is possible to increase service satisfaction for guests providing sensitive personal information and ensure security safety.

[0016] In addition, regarding document images uploaded to the server via a guest's terminal, by configuring the system to allow photo processing and input only for images taken within a certain period of time, for example, within 24 hours, based on the ID card shooting time information, it has the effect of preventing the reuse of identity information and ID card theft that may occur in conventional systems, and ensuring the reliability of the ID card.

[0017] In addition, by optimizing the data input process by considering various variables such as the hotel PMS environment or visits by group tourists, the present invention can be implemented even in outdated system environments where expensive API integration is difficult, thereby maintaining fast processing speeds and providing comfortable and prompt check-in services to group guests.

[0018] Furthermore, by automatically identifying passport layouts and notation formats that differ by country through an AI-based classification model and extracting corresponding identity information, the present invention enables stable processing of identity information without separate additional settings, even in hotel environments visited by foreigners of various nationalities.

[0019] In addition, the present invention stores and utilizes shooting time information, hotel check-in time information, room occupancy status, etc., as hotel operation data to store and analyze hotel guest accommodation data, thereby providing information that can be used for the efficient operation of the hotel, such as predicting check-in requests concentrated at specific times, room allocation, and front desk staff deployment.

[0020] Meanwhile, the effects of the present invention are not limited to those mentioned above, and various effects may be included within the scope obvious to a person skilled in the art from the contents described below. Brief explanation of the drawing

[0021] FIG. 1 illustrates, in its entirety, an automatic input system for a hotel guest management program according to one embodiment of the present invention, and FIG. 2 is a flowchart showing the overall flow of a distributed scan processing method using an automatic input system for a hotel guest management program according to an embodiment of the present invention, and FIG. 3 illustrates a screen for selecting a photograph of a location captured on a hotel guest terminal according to an embodiment of the present invention, and FIG. 4 illustrates an example of an image list verification and data extraction request screen according to an embodiment of the present invention, and FIG. 5 illustrates a user authentication and service access screen for a hotel guest management program according to one embodiment of the present invention. Specific details for implementing the invention

[0022] Hereinafter, some embodiments of the present invention will be described in detail with reference to exemplary drawings. It should be noted that in assigning reference numerals to the components of each drawing, the same components are given the same reference numeral whenever possible, even if they are shown in different drawings.

[0023] In addition, when describing embodiments of the present invention, if it is determined that a detailed description of related known configurations or functions would hinder understanding of the embodiments of the present invention, such detailed description is omitted.

[0024] In addition, terms such as first, second, A, B, (a), (b), etc., may be used when describing the components of the embodiments of the present invention. These terms are used merely to distinguish the components from other components, and the essence, order, or sequence of the components is not limited by such terms.

[0025] Additionally, in describing embodiments of the present invention, each "part," "module," or "step" may be implemented through a processor and memory. The term "processor" should be broadly interpreted to include general-purpose processors, central processing units (CPUs), microprocessors, digital signal processors (DSPs), controllers, microcontrollers, and state machines, etc. In some environments, the term "processor" may refer to an application-specific integrated circuit (ASIC), a programmable logic device (PLD), a field programmable gate array (FPGA), etc. The term "processor" may also refer to a combination of processing devices, such as, for example, a combination of a DSP and a microprocessor, a combination of multiple microprocessors, a combination of one or more microprocessors combined with a DSP core, or any other combination of such configurations.

[0026] Additionally, memory should be broadly interpreted to include any electronic component capable of storing electronic information. Memory may also refer to various types of processor-readable media, such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), eraseable-programmable read-only memory (EPROM), electrically eraseable PROM (EEPROM), flash memory, magnetic or optical data storage devices, registers, etc. If a processor can read information from memory or write information to memory, memory is said to be in an electronic communication state with the processor, and each "part," "module," or "stage" may be implemented through a program or application based on the processor and memory in an electronic communication state.

[0027] In addition, the communication network may be a wired or wireless communication network and may utilize at least one of LAN (local area network), WLAN (wireless LAN), Wibro (wireless broadband), WiMAX (world interoperability for microwave access), HSDPA (high speed downlink packet access), and various other communication standards. Furthermore, the communication network may be a mobile communication network, and it may be a synchronous mobile communication network or an asynchronous mobile communication network. As an example of the asynchronous mobile communication network, a WCDMA (Wideband Code Division Multiple Access) network may be cited. In this case, although not shown in the drawings, the mobile communication network may include, for example, an RNC (Radio Network Controller). Meanwhile, although the WCDMA network was cited as an example, it may be a next-generation communication network such as a 3GLTE network, 4G network, or 5G network, as well as other low-power communication networks such as Wi-Fi, Bluetooth, and Zigbee, and wired / wireless IP networks.

[0028] Furthermore, in the present invention, Artificial Intelligence (AI) refers to a technology that imitates human learning ability, reasoning ability, and perceptual ability, and implements them on a computer, and may include concepts such as machine learning and symbolic logic. Machine Learning (ML) is an algorithmic technology that classifies or learns the characteristics of input information on its own. AI technology can analyze input information as a machine learning algorithm, learn from the results of that analysis, and make judgments or predictions based on the results of that learning. Additionally, technologies that mimic the functions of the human brain, such as cognition and judgment, by utilizing machine learning algorithms can also be understood as falling within the category of AI. For example, technological fields such as linguistic understanding, visual understanding, reasoning / prediction, knowledge representation, and motion control may be included.

[0029] Artificial intelligence learning models or neural network models can be designed to implement the structure of the human brain on a computer and may include multiple network nodes that simulate neurons of a human neural network and have weights. These multiple network nodes can have interconnected relationships by simulating the synaptic activity of neurons, where they exchange signals through synapses. In an artificial intelligence learning model, multiple network nodes can be located in layers of different depths and exchange information according to convolutional connections. Artificial intelligence learning models can be machine learned using methods such as supervised learning, unsupervised learning, and reinforcement learning. Machine learning algorithms used to perform machine learning may include decision trees, Bayesian networks, support vector machines, perceptrons, and clustering.

[0031] From now on, with reference to the attached drawings, a system for automatically inputting mobile-capture-based document automatic classification and text extraction results into a hotel PMS system in conjunction with RPA according to one embodiment of the present invention will be described in detail.

[0032] FIG. 1 illustrates an automatic input system (100) for a hotel guest management program according to an embodiment of the present invention, FIG. 2 illustrates a process of inputting guest information through a QR code provided to a guest, FIG. 3 illustrates a screen for selecting a photograph of a location captured on a hotel guest terminal according to an embodiment of the present invention, FIG. 4 illustrates an example of a screen for checking an image list and requesting data extraction according to an embodiment of the present invention, FIG. 5 illustrates a screen for user authentication and service access for a hotel guest management program according to an embodiment of the present invention.

[0034] As illustrated in FIG. 1, a hotel guest management program automatic input system (100) according to one embodiment of the present invention may include a document receiving unit (110), a recognition extraction unit (120), a verification unit (130), and an input processing unit (140).

[0035] First, the document receiving unit (110) is configured to receive image data containing documents from a guest terminal or a hotel front terminal and is electrically connected to the recognition extraction unit (120). The document receiving unit (110) receives document images, such as passports and identification cards, taken by guests via a network and may be configured to record reception time information for the received document images. The reception time information for the document images collected during this process can be used to control the processing order of the document images in a subsequent processing stage, or for hotel guest management and hotel operation.

[0036] For example, the document receiving unit (110) may be configured to manage each document image by assigning individual identification information even when document images are received simultaneously from multiple external terminals, and accordingly, even when multiple identification document images of group guests are input simultaneously, the document images can be processed and managed without confusion between the document images.

[0037] Next, the recognition extraction unit (120) is configured to receive document image data transmitted from the document receiving unit (110), perform preprocessing and quality correction processing on the document image, identify the type of document, and extract guest identity information included in the document, and is electrically connected to the document receiving unit (110) and the verification unit (130). The recognition extraction unit (120) can perform at least one of tilt correction, distortion correction, and noise removal by considering the shooting angle of the document image, whether there is image distortion, resolution, and degree of noise inclusion.

[0038] Additionally, the recognition extraction unit (120) may be configured to identify the type of scanned document using an AI-based classification model generated based on the results of learning multiple document images, and the method of extracting identity information or the extraction items may be different depending on the document type. The AI-based classification model can determine whether the input document corresponds to a passport, driver's license, national ID card, or other identification card by comprehensively analyzing the overall layout of the document image, the text arrangement form, the presence or absence of specific areas, document color patterns, or location information of security elements.

[0039] For example, if the input document image is a passport image, the recognition extraction unit (120) can identify the document type as a passport based on the passport's unique cover design, the location and shape of the machine-readable area (MRZ), the country code area, etc. When the document type is identified as a passport, the recognition extraction unit (120) can be configured to extract identity information by setting items such as name, passport number, nationality, date of birth, and issuing country placed on the internal pages of the passport as targets for identity information extraction.

[0040] Meanwhile, if the input document image is a driver's license image, the recognition extraction unit (120) can identify the document type as a driver's license based on the unique card shape of the driver's license, the arrangement of the photo area and text area, the license number display area, etc. When the document type is identified as a driver's license, the recognition extraction unit (120) can be configured to extract identity information by setting items such as the name, license number, issuing agency, and validity period listed on the driver's license as targets for identity information extraction.

[0041] In this way, the recognition extraction unit (120) applies different target items for identity information extraction, extraction locations, or extraction methods depending on the document type, thereby enabling accurate extraction of identity information even when document images with different formats and layouts are input. Accordingly, even when passports with different formats by country or driver's licenses with different designs are input, extraction processing optimized for each document type can be performed without relying on a single recognition logic.

[0042] Next, the verification unit (130) provides hotel guest identity information extracted by the recognition extraction unit (120) to the user, receives confirmed or modified identity information from the user, and determines whether to input the identity information into the hotel management system. The verification unit (130) is electrically connected to the recognition extraction unit (120) and the input processing unit (140).

[0043] Specifically, the verification unit (130) receives guest identity information automatically extracted based on a document image provided by the recognition extraction unit (120), transmits it to a user or system administrator, and determines whether to transmit the identity information to the input processing unit (140) based on the confirmation input of the user or system administrator.

[0044] The verification unit (130) can be implemented through a display screen of a guest terminal or a hotel front terminal and can be configured to visually provide the user with key items necessary for guest management, such as name, date of birth, and document number among the extracted identity information. At this time, the verification unit (130) can induce the user to verify the accuracy of the extracted identity information or to modify some identity information items. The verification unit (130) can be configured to receive at least one of a confirmation input, a modification input, or an input cancellation input from the user, and selects a subsequent information processing flow according to the type of user input received. For example, if the user performs a confirmation input regarding the extracted identity information, the verification unit (130) can control the extraction identity information to be transmitted to the input processing unit (140) so that automatic input to the hotel management system is performed. At this time, if the user performs a modification input regarding some identity information, it can be configured to transmit the modified identity information to the input processing unit (140).

[0045] Additionally, if the user selects to cancel the input, the verification unit (130) may be configured to stop the input processing of the corresponding identity information and transmit a request for reprocessing or re-photography to the recognition extraction unit (120) or the document receiving unit (110). In this case, the verification unit (130) may act as an intermediate control node that controls whether to transmit guest identity information between the recognition extraction unit (120) and the input processing unit (140).

[0046] Meanwhile, the verification unit (130) may be configured such that, depending on the implementation method, the entity verifying the identity information is either the guest themselves or at least one of the hotel manager. For example, if a guest takes a picture of a document using their mobile terminal and uploads it to a relay server, the verification unit (130) may be configured so that the guest themselves directly verifies the extracted identity information and approves the server upload. On the other hand, if a document image is input through a hotel front terminal, the verification unit (130) may be configured so that the hotel manager verifies and approves the identity information extracted through the front terminal. In this way, the verification unit (130) verifies the result of automatic extraction of identity information through a user verification procedure, thereby preventing errors that may occur during the automatic recognition process by the machine system in advance and allowing the entity approving the input of identity information to be flexibly set according to the situation.

[0047] In an automatic input system (100) for a hotel guest management program according to an embodiment of the present invention, the verification unit (130) may refer to the shooting time information of a document image corresponding to the guest identity information extracted by the recognition extraction unit (120) together, and may be configured to allow input to the hotel guest management program only when the document image is taken within a preset valid time range. Specifically, when the document receiving unit (110) receives document image data from an external terminal, it may be configured to acquire and record shooting time information based on the shooting information included in the document image. The recorded shooting time information may be transmitted to the verification unit (130) via the recognition extraction unit (120). The verification unit (130) may compare the transmitted shooting time information with a reference time or check-in reference time preset by the hotel and may be configured to allow input of identity information only when the document image was taken within 24 hours. For example, based on the time when a guest performs check-in, the verification unit (130) may be configured to allow user verification input and activate the input processing unit (140) only if the document image was taken within 24 hours prior to that time. According to this embodiment, the reuse of document images taken in the past or unauthorized use of document images by a third party can be prevented, and the input of guest identity information can be restricted to using only the latest document image taken directly by the actual guest at the time of check-in, thereby improving the reliability of guest identity information and the security of hotel guest personal information management.

[0048] Additionally, the verification unit (130) may be configured to perform valid shooting time verification to determine whether the time of shooting of the document image corresponds to a time period adjacent to the time of hotel check-in.

[0049] Specifically, the verification unit (130) can determine whether the document image was taken within a pre-set valid shooting time range by comparing the shooting time information (EXIF metadata) included in the image file containing the document with the time the document image was received on the server.

[0050] For example, the verification unit (130) may be configured to determine whether the time of capture of the document image is within 24 hours based on the current server time. In this case, in addition to 24 hours, it may be variably changed to 6 hours or 12 hours, etc., depending on the hotel operation policy or check-in policy.

[0051] And, if it is determined that the captured document image is outside the valid shooting time range, the verification unit (130) may be configured to perform at least one of the following.

[0052] 1. A step of providing a user interface that induces reshooting.

[0053] 2. Step to block the upload of the relevant document image

[0054] 3. A step of temporarily disabling the automatic input function of the input processing unit (140) for the corresponding check-in session.

[0055] Accordingly, it is possible to prevent the reuse or unauthorized submission of document images taken in the past regardless of the time of hotel check-in, and to have the effect of reflecting only documents linked to the actual guest's on-site verification procedure in the system (100) according to one embodiment of the present invention.

[0056] Additionally, the verification unit (130) can be configured to perform a double verification structure combined with server-based session information, without relying solely on EXIF ​​metadata, in consideration of the possibility of forgery and alteration of the shooting time information.

[0057] Specifically, at the step where the hotel generates the QR code, the server can issue a session token corresponding to the check-in session, and the session token can be configured to include issuance time and expiration time information.

[0058] The verification unit (130) can determine whether the time of capture of the uploaded document image is within the session validity period range by comparing the time of capture of the uploaded document image with the time of issuance and expiration of the session token. For example, if the time of capture of the document image is before the issuance of the session token, or if the time of capture of the document image is after the expiration of the session token, the uploaded document image may be determined to be invalid.

[0059] Accordingly, even if the guest's terminal arbitrarily modifies the EXIF ​​capture time of a document image, it is possible to determine whether the capture time has been manipulated through cross-verification with the session issuance time information held by the hotel server. Through this dual verification structure, it can function as a security verification mechanism to determine whether the document image capture time information has been forged.

[0060] Next, the input processing unit (140) is activated when the verification unit (130) determines that the input of identity information is to be performed, and can perform the function of automatically inputting the extracted and verified guest identity information into the hotel guest management program. The input processing unit (140) is electrically or telecommunicationally connected to the verification unit (130) and the hotel guest management program.

[0061] The input processing unit (140) may be configured to selectively perform at least one of several automatic input methods depending on the structure or integration environment of the hotel guest management program. Specifically, the input processing unit (140) may be configured to transmit guest identity information to the hotel guest management program and input identity information using an application programming interface (API) provided by the hotel guest management program. If the hotel guest management program is a model that does not support external API integration, the input processing unit (140) may be configured to perform information input by recognizing an input field displayed on the input screen of the hotel guest management program and automatically entering identity information into the corresponding input field. In this case, the input processing unit (140) may perform automatic input processing based on the location of the input field on the input screen, the order of items, or input rules. Additionally, the input processing unit (140) may be configured to transmit the result of the identity information input to the verification unit (130) or a separate record management module. In this case, the input processing unit (140) may record information such as whether the input was successful, the time at which the input failure occurred, an error code or error type corresponding to the cause of the failure, and the number of retries. Additionally, the input processing unit (140) may be configured to perform an input processing retry process based on the recorded cause of failure and a pre-set retry condition when input failures occur repeatedly, and may stop input processing and provide a notification to the user or administrator if the number of retries exceeds a threshold value. According to this embodiment, system overload caused by repeated errors in hotel guest identity information input can be prevented.

[0063] Next, with reference to FIG. 2, a distributed scan processing method (S100) using a hotel guest management program automatic input system according to one embodiment of the present invention will be described in detail.

[0064] As illustrated in FIG. 2, a hotel guest management program automatic input system (100) according to one embodiment of the present invention can perform a distributed scan processing method (S100) using the hotel guest management program automatic input system from the document image collection step to the guest identity information automatic input step.

[0065] A distributed scanning method (S100) using an automatic input system for a hotel guest management program according to an embodiment of the present invention may be performed by the following steps: presenting a QR code for a group from the hotel (S110); a guest terminal recognizing the QR code and connecting to a check-in session (S120); when a guest photographs and uploads an identity verification document, an AI-based model automatically classifies the document type (S130); extracting OCR from the document image and self-detecting data (S140); transmitting encrypted text data to a server (S150); aggregating the data transmitted to the server by group session (S160); and automatically inputting hotel guest information into the PMS through front approval (S170).

[0066] First, the step (S110) of presenting a QR code for a group by the hotel can be performed at a hotel front desk terminal or a hotel guest management program. The QR code for a group can be generated to include reservation information for check-in, check-in session identification information, and input page access information, and can be provided as an identification means to allow multiple guests to access the same check-in session. A hotel employee can present the generated QR code to the guest in the form of a printout, a kiosk screen, or a mobile screen.

[0067] Next, the step (S120) of a guest recognizing a QR code with a mobile terminal can be performed by the guest using their own mobile terminal. The guest terminal can recognize the QR code using a camera or a QR scanner application and access the input page of the corresponding guest management program. At this time, multiple guest terminals can be connected to the same check-in session while being provided with independent input environments, and accordingly, a distributed scan input environment that allows for simultaneous input can be provided.

[0068] Next, the step (S130) in which an AI-based model automatically classifies the document type when a guest photographs and uploads an identity verification document can be performed through cooperative processing between the guest terminal and the server. The guest may photograph and upload an identity verification document, such as a passport, resident registration card, driver's license, or alien registration card, using the camera of a mobile terminal. The uploaded document image can be automatically analyzed by an AI-based classification model generated based on the results of learning multiple document images, thereby identifying the document type. Meanwhile, in the case of a passport, country information can also be identified by taking into account differences in passport formats by country.

[0069] Next, the step (S140) of OCR extraction and self-detection of data from a document image can be performed by the recognition extraction unit. At least one quality correction process among tilt correction, distortion correction, resolution correction, or noise removal can be performed on the document image, and subsequently, identity information such as name, date of birth, and passport number can be extracted through character recognition (OCR). At this time, if the reliability of the identity information extraction result is below a threshold value, re-correction or re-extraction can be performed.

[0070] Next, the step of transmitting encrypted text data to the server (S150) can be performed by a secure transmission protocol. Identity information extracted from the guest terminal can be converted into an encrypted text data format and transmitted to the server, and in this process, the server transmission section can be encrypted to protect personal information.

[0071] Next, the step (S160) of aggregating data transmitted to the server by group session allows guest information corresponding to QR codes to be grouped and managed based on check-in session identification information, terminal identification information, and reception time information included in each document data. Through this process, information on group guests can be managed collectively and utilized in conjunction with hotel operational information, such as room assignment.

[0072] Finally, a step (S170) in which hotel guest information is automatically entered into the PMS through front approval can be performed. The hotel manager or guest can verify or modify the extracted identity information, and only approved information can be automatically entered into the hotel guest management program. At this time, automatic entry can be performed by at least one of a transmission method via a program interface (API) or an automatic entry method for input fields.

[0073] According to the distributed scanning method (S100) using an automatic input system for a hotel guest management program according to an embodiment of the present invention as described above, it is possible to resolve the limitations of a centralized input structure unified by front desk staff that occurred during the conventional hotel check-in process. Conventionally, since hotel staff had to collect guests' passports or identification cards in bulk and then sequentially input information using a terminal, there was a problem where check-in waiting times became long and front desk operations became heavy when group guests arrived simultaneously. According to the distributed scanning method (S100) using an automatic input system for a hotel guest management program according to an embodiment of the present invention, a distributed input structure is formed in which multiple guests each independently photograph and upload documents while connected to the same check-in session using their own mobile terminals. Consequently, input operations can be performed in parallel, there is no manual input step at the hotel front desk, and the input time, which used to increase in proportion to the number of guests, can be significantly reduced. Furthermore, since the server can automatically group guest information based on check-in session information and reception time information included in each document, the hotel manager can process the verification of group guest information and room assignment tasks in bulk. In addition, through AI-based document classification and OCR extraction, input items are automatically configured according to document type and automatically entered into the PMS after a verification process, effectively preventing typos or omissions that occur during manual input.

[0074] Next, the recognition extraction unit (120) processes the document images received by the document receiving unit (110) at a central server and performs document type identification and guest identity information extraction for each document image through OCR. At this time, the recognition extraction unit (120) can automatically determine the document type using an AI-based classification model generated based on the results of learning multiple document images, and if the reliability of the extracted identity information is below a threshold value, the information OCR extraction process can be performed again.

[0075] Next, the verification unit (130) provides the guest identity information extracted by the recognition extraction unit (120) to the user and may request and receive confirmation or modification input regarding the identity information from the user. At this time, the verification unit (130) determines whether to perform the input of identity information based on the user verification result, and activates the input processing unit (140) only if the input performance is approved. On the other hand, if the input cancellation or document re-photography is requested, the input processing for the corresponding identity information may be stopped and a request for re-input of the document image may be made. At this time, the verification subject may be the guest themselves or the hotel manager, and the verification unit may be implemented through the guest terminal or the hotel front terminal.

[0076] Finally, the input processing unit (140) can automatically input guest identity information, for which input execution has been determined by the verification unit (130), into a hotel guest management program. At this time, an RPA bot can automatically perform input through the PMS system UI or API and verify the input results. With this configuration, the system of the present invention can efficiently perform distributed scanning using guest terminals and document recognition, verification, and input processing based on a central server even in situations where multiple guests check in simultaneously, and can provide the effect of reducing the burden of hotel front desk operations.

[0078] Next, according to one embodiment of the present invention, a guest can recognize a QR code and take a picture of a document using their mobile terminal, and the mobile terminal screen may include a user interface to guide the guest to accurately take a picture of an identity document.

[0079] Referring to FIG. 3, the mobile terminal screen may include a guidance message instructing the user to place documents to be scanned one by one within the screen for shooting. For example, a phrase such as "Illuminate and take a picture of each document so that they are within the screen one by one" may be displayed at the top of the screen to prevent multiple documents from being photographed simultaneously or documents from moving outside the screen area. Additionally, a guide frame indicating a document shooting area may be displayed in the central area of ​​the mobile terminal screen, and the user may place identification documents such as passports, driver's licenses, and alien registration cards within the guide frame and then take a picture. Accordingly, the position, size, and orientation of the documents are maintained within a certain range, thereby improving the document recognition accuracy by the recognition extraction unit (120) thereafter. A preview list of captured images may be displayed in the bottom area of ​​the mobile terminal screen, and the user may view or select the document images that have already been captured. Additionally, the user may delete specific images among the captured images or perform continuous shooting through additional shooting. Additionally, a shooting button, a save button, and a reshoot button may be provided on the mobile terminal screen, and the guest can perform a saving operation to transmit the document image to the server after checking the shooting result. When the image saving operation is performed, the captured document image is transmitted to the document receiving unit (110) and subsequently transmitted to the document type identification and identity information extraction stage by the recognition extraction unit (120). Through such a mobile terminal user interface, the guest can directly perform identity document shooting using their own terminal without visiting or waiting at the hotel front desk, and can minimize errors that may occur during the shooting process. By inducing the hotel to meet the identity document shooting conditions required from the document shooting stage in this way, the accuracy of the identity information processing process can also be improved.

[0081] Next, according to one embodiment of the present invention, an interface may be provided that allows a guest to check the quality of an image and select whether to request data extraction after taking a picture of an identity document using a mobile terminal. Referring to FIG. 4, the captured document image is displayed on the mobile terminal screen, and the user can visually inspect the image to check whether the entire area of ​​the document is included and whether the resolution and clarity capable of character recognition are maintained. At this time, the interface screen may be configured to highlight the currently selected image among the captured images so that the user can intuitively recognize the image to be reviewed. In addition, after taking the identity document, a guidance message may be displayed on the mobile terminal screen instructing the user to inspect the quality of the captured image and then select to take additional photos or save it. For example, the guest may be guided through a phrase such as, "Inspect the quality of the captured image and select to take additional photos or save it." Through the provided interface screen, the guest can select and delete some or all of the captured images, and if necessary, retake the image or perform continuous shooting, thereby removing blurry images, document images with some identity information missing, and duplicate images in advance. Additionally, when the user selects the save button, a confirmation popup or guidance screen requesting the extraction of guest identity information from the captured image may be provided. Through the guidance screen, the user may approve or cancel the data extraction request, and upon approval, the captured document image is transmitted to the server, and the guest identity information extraction process may be carried out by the recognition extraction unit (120).The step of providing an interface that allows checking the quality of such captured images and selecting whether to request data extraction has the effect of preventing unwanted information processing or automatic recognition of incorrect images through a verification process based on the user's intent before identity information is automatically extracted.

[0083] Next, according to one embodiment of the present invention, an automatic input result and a sign-up authentication guidance screen may be provided in a hotel guest management program. Referring to FIG. 5, the automatic input result and sign-up authentication guidance screen in the hotel guest management program may include an initial screen for logging into the hotel guest management program, a login screen for user account authentication, and a confirmation guidance screen for checking the automatic input result. For example, a guest user may be guided on using the hotel guest management program according to the present invention in a mobile terminal environment accessed by recognizing a QR code provided by the hotel, and may perform a user login or account authentication procedure using an email and password at a stage before or after the automatic input processing. In addition, as illustrated in FIG. 5, the login and authentication guidance screen may provide the guest user with guidance functions regarding the method of use of the hotel guest management program automatic input system according to one embodiment of the present invention, the automatic input procedure, and the method of processing personal information, and may include a sign-up guidance function for new users. By introducing guest user accounts and account authentication procedures into the automatic input system of a hotel guest management program according to one embodiment of the present invention, the possibility of unauthorized use of document images or identity theft by a third party can be reduced, and the effect of enhancing personal information protection and security can be provided.

[0085] Although all components constituting the embodiments of the present invention have been described above as being combined or operating in combination, the present invention is not necessarily limited to such embodiments. That is, within the scope of the purpose of the present invention, all components may be selectively combined in one or more ways to operate.

[0086] Furthermore, terms such as "include," "compose," or "have" described above, unless specifically stated otherwise, mean that the relevant component may be inherent; therefore, they should be interpreted as allowing for the inclusion of additional components rather than excluding them. All terms, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains, unless otherwise defined. Commonly used terms, such as those defined in advance, should be interpreted in accordance with their meaning in the context of the relevant technology and should not be interpreted in an ideal or overly formal sense unless explicitly defined in the present invention.

[0087] Furthermore, the above description is merely an illustrative explanation of the technical concept of the present invention, and those skilled in the art to which the present invention pertains will be able to make various modifications and variations within the scope of the essential characteristics of the present invention.

[0088] Accordingly, the embodiments disclosed in this invention are intended to illustrate, not limit, the technical concept of the invention, and the scope of the technical concept of the invention is not limited by these embodiments. The scope of protection of this invention shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of this invention. Explanation of the symbols

[0089] 100: Automatic input system for a hotel guest management program according to an embodiment of the present invention 110 : Document receiving unit 120 : Recognition extraction unit 130 : Verification section 140 : Input processing unit

Claims

Claim 1 A hotel guest management program automatic input system comprising: a document receiving unit that receives image data containing a document from an external terminal and collects and records shooting time information (EXIF metadata) included in the file structure of the image data together with the image data; a recognition extraction unit that identifies the type of document based on the image data received from the document receiving unit and structures and extracts guest identity information from a text area within the image data by applying recognition logic corresponding to the identified document type; a verification unit that provides the guest identity information extracted by the recognition extraction unit to a user and determines whether to input the identity information into a hotel guest management system based on user confirmation input, wherein the verification unit controls the input of identity information only when the image data is captured within a preset valid time range by referring to the shooting time information (EXIF metadata) received from the document receiving unit; and an input processing unit that performs information input by transmitting the identity information through an application program interface (API) of a hotel management system (PMS) or by automatically entering the identity information into an input field of the hotel management system, only when the input is determined to be performed by the verification unit. Claim 2 A hotel guest management program automatic input system according to claim 1, wherein the document receiving unit is configured to allow a guest to directly transmit an image of the document captured by the guest using their terminal to a server through an input page corresponding to a QR code provided to the guest terminal. Claim 3 The automatic input system for a hotel guest management program according to claim 1, wherein the document receiving unit is configured to process image data containing a plurality of documents received simultaneously from a plurality of external terminals in a sequential or parallel manner. Claim 4 A hotel guest management program automatic input system according to claim 1, wherein the input processing unit is configured to selectively input only some information among the identity information according to the input policy or settings of the hotel management system (PMS), or to mask or delete some sensitive information among the identity information before or after input. Claim 5 An automatic input system for a hotel guest management program according to claim 1, wherein the recognition extraction unit performs one or more quality correction processes among tilt correction, distortion correction, resolution correction, or noise removal on a document image, identifies the type of document using an AI-based classification model generated based on the results of learning a plurality of document images, and performs re-correction processing on the document image if the reliability of the extraction result of the identity information is less than a preset threshold value.

Images