CORPORATE ASSET VERIFICATION SYSTEM

Abstract

This invention relates to a system (1) that enables the verification of asset images submitted by users in the assignment and reconciliation processes of corporate assets, comparison with past images, detection of image manipulations and comparison of asset information with corporate records.

Description

1 TARIFF CORPORATE ASSET VERIFICATION SYSTEM Technical Area This invention is in the fields of enterprise asset management and artificial intelligence, and deals with embezzlement and Verification of visual evidence presented in reconciliation processes, past comparison with images, detection of image manipulations, and with a system that enables the comparison of asset information with corporate records It is related. 10 Previous Technique In existing corporate asset management and liability reconciliation systems, Verification of assets held by users is mostly done by user 15 the statement made by and the system to support this statement These systems are based on uploaded photos or images. These systems use uploaded visuals. accepting the evidence as a mere file entry and the file format or standard Provided basic controls such as file size are met, the asset can be physically present in the employee's possession. It is assumed that it exists as such. However, high-budget or critical 20 In periodic audits of corporate assets, the accuracy of visual evidence and an autonomous control mechanism that examines its authenticity at the pixel level It is not available. Current applications digitally record the uploaded photo. whether it has been manipulated, whether it has been accessed by another user in the past. It cannot distinguish whether an uploaded photo is a copy or not. Therefore, 25 Before the image uploaded by the user is accepted, the following whether the image has been uploaded before, a digital image on the image whether changes have been made or whether the entity in the image is actually relevant an integrated check is performed to determine whether an asset is assigned to the user. A new system is needed. 30 2 Patent number US7421128B2, which is included in the prior art. the document determines the similarity of digital images and the images Techniques for comparison are described, despite various modifications. However, these techniques are used in corporate asset reconciliation processes, Images submitted by different users are being reused for the purpose of abuse. 5 detection of usage, examination of image manipulation and existence used to verify identity through corporate records It is not mentioned. Patent document number US10803542B2 describes image recognition of physical objects. A system for identifying these techniques using various methods is described. However, in this solution, the asset image shows past users. Comparison of uploads in terms of repetition and image manipulation by examining it from this perspective and evaluating corporate asset records together. There is no integrated structure for this purpose. 15 Brief Description of the Invention The purpose of this invention is to improve the accounting and reconciliation processes of corporate assets. the accuracy and reliability of visual evidence provided by the user 20 increasing, and allowing the same image to be reused by different users, a system that detects digital manipulations performed on images to present. Another purpose of the invention is to obtain 25 asset images from the user-uploaded image. The visual signature obtained is compared with the visual signatures of images previously uploaded to the system. comparison and determination of the level of similarity between the images to provide. 3 Detailed Description of the Invention "Corporate Entity Verification" was carried out to achieve the purpose of this invention. The "System" is shown in the attached figure; Figure 1. Schematic view of a system that is the subject of the invention. The parts shown in the figure are individually numbered, and these numbers... The corresponding answers are given below. 1. System 2. Electronic device 3. Database 4. Server Asset 15 uploaded to the system by the user during corporate asset reconciliation verification of the images, comparison with past images, to detect and alterations and manipulations made to the image. This allows the uploaded asset image to be verified by comparing it with the records. The subject of the invention is the system (1); - 20 for the user to obtain images and videos of the asset. entering the necessary permissions, the results of the analyses performed at least one that provides an interface to enable its presentation to the user electronic device (2), - user, asset, liability, historical video recordings, video signatures, Analysis results, asset characteristics, verification results, and abuse 25 to ensure the storage of information regarding their evaluations at least one database is structured (3) and, - processing images received via electronic device (2), images Creating visual signatures, comparing them with past images, images Detecting manipulations, obtaining asset identification information from the image 30 4 to record the information obtained in the database (3) compare the information and reach an agreement based on the results obtained. at least one server configured to evaluate the operation (4) It includes. The system in question consists of (1) electronic devices (2), smartphones, tablet computers or a device in the form of a portable computer. The preferred designation of the invention. In its application, the Internet is used as a data bus with electronic devices (2) and server (4). It is configured to exchange data using electronic means. device (2), asset information, assignment information and reconciliation by the user 10 entering information regarding the transaction, requesting asset information from the user. obtaining the necessary permissions for taking the images and the asset information to provide an interface that allows the user to input the image. is configured. Electronic device (2), performed by server (4) image analysis, reconciliation process status, analysis results, the 15 information regarding whether the application is under review or rejected It is configured to enable display. In the system that is the subject of the invention, the database (3) in (1) is in communication with the server (4) and is configured to be managed by the server (4). The invention is preferred In the implemented application, the database (3) stores user information, user information and 20 Associated asset and liability information, asset identification information, product type, Brand, model and color information, asset images, image uploads. their times and uploading user information, image signatures, history screenshots records, visual similarity values, similarity distances, images manipulation analysis results, manipulation risk values, 25 obtained from the image the received identity information, expected and detected entity characteristics, verification and non-compliance results, fraud risk scores, reconciliation transaction statuses, transaction times, analysis results, records of control checks performed, and It is structured to store the generated assessment results. The server (4) in the system (1) which is the subject of the invention, any remote communication Communicating with electronic devices (2) and databases (3) using the protocol and data through this communication established with electronic device (2) and database (3) It is configured to carry out the exchange. The server in question (4) is configured to process the data. Making new data entries into the database (3), 5 entries in the database (3) deletion of data or modification of registered data in the database (3), Updating and retrieving registered data in the database (3) It is structured to manage the database (3) with operations such as these. Server (4) receives user information from the user via electronic device (2), user asset information. 10. Ensure that the necessary sharing permissions are requested in order to obtain the relevant images. It is structured in such a way. The server (4) receives from the electronic device (2). to receive the asset reconciliation request and, within the scope of that request, to process the user's information. personal and / or financial information, embezzlement information, and information requested to be verified to retrieve the records relating to the asset from the database (3) is configured. Server (4), electronic device (2) 15 by the user to receive the uploaded raw asset photos from there It is configured. The server (4) displays the image received from the electronic device (2). It is configured to remove noise and resize. The server (4) uses perceptual summarization to create the visual signature of the image. (perceptual hashing) and / or convolutional neural network 20 It is structured to use the methods. The server (4) is another of the invention. deep embedding for creating a visual signature in the application embeddings) and / or convolutional neural network It is configured to use the models. The server (4) is created visual signature with past image records in database (3) 25 It is configured to compare the images. The server (4) is configured to compare the images. Hamming distance and / or Euclidean distance to determine similarity It is configured to use Euclidean distance methods. Server (4) provided a vector for the search operation in another application of the invention. search library (vector search library) and / or vector database (vector 30 The server (4) is configured to use the database. 6 Hamming distance with a predetermined threshold value It is configured to compare. The server (4) is the Hamming distance. If the subject is below the threshold value, the server (4) displays the image as it was previously displayed. To identify duplicate usage with an uploaded image and perform the reconciliation process. It is configured to stop. The server (4) can stop the situation again. 5 Creating an abuse alert by identifying it as anti-replay. It is structured accordingly. The server (4) shows the image more according to the similarity result. to determine whether it has been used before and whether it is a duplicate use is configured. The server (4) ensures that the image is not duplicated. After identification, digital manipulation was found on the image and 10 error level analysis to determine if it is not present It is configured to implement the method. The server (4), in the image Determining the manipulation risk value by analyzing regional compression differences. It is configured to create the server (4), label, barcode and 15 object detection method to determine the identity area It is configured to use. The server (4) uses serial from the specified area. optical character recognition to obtain the number and other identification information It is configured to use optical character recognition (OCR) technology. Server (4) retrieves the serial number obtained from the image into the database (3) It is configured to compare with asset records. Server (4), serial 20 The brand, model, color, and visual characteristics obtained from the image using the number are data. The server is configured to compare with the records in the database (3). (4), the comparison in question is semantic matching It is configured to be carried out by the method. Server (4), serial number If there is a discrepancy between the visual and physical characteristics, the physical label will be different. 25 spoofing detection related to the use of counterfeit materials on an asset It is configured to perform. The server (4) displays another image. an image taken from the display screen or a previously created image to identify instances of reuse (anti-counterfeiting protection) 30 to use spoofing and anti-replay detection methods is being configured. The server (4) states that the image has not been used before. 7 After it was determined whether digital manipulation was found on the image error level analysis to determine if it is not present It is structured to implement the method. The invention is a preferred method. In the application, the server (4) uses a static image instead of a user-generated image. dynamic imaging operations performed via electronic device (2) 5 liveness detection and anti-counterfeiting protection for analysis. The server (4) is configured to use (anti-spoofing) methods. compression levels in the image and regional compression manipulation risk score by analyzing their differences It is configured to generate the score. The server (4) sets the score between 0 and 100. It can be determined between these two options, and if the score is above 85%, the image is displayed. It is configured to reject or halt the reconciliation process. Server (4) displays similar images uploaded by different users. graph-based relationship analysis to identify relationships It is configured to perform analysis. Server (4), common image 15 by identifying the relationships consisting of usage and similar abusive behaviors to identify possible links of organized abuse among users is configured. The server (4) displays the results obtained from image analyses. to evaluate and determine the final outcome of the reconciliation process is being configured. Server (4), 20 if the validation conditions are met. to approve the reconciliation process, otherwise reject the process or send it for review. It is configured to receive the generated results and related information. The server (4) to record technical evaluations into the database (3) It is being structured. 8 Industrial Application of the Invention The system in question (1) is used in the processes of assigning and reconciling corporate assets. Verification of the presented images, comparing the images with past records. Comparison, detection of image manipulations and asset information 5 in information processing infrastructures for comparison with corporate records It can be implemented. The system (1) is an electronic device (2) by the user. receiving asset images uploaded via, images from server (4) to analyze, the results obtained are in the database of assets and liabilities (3) compares it with the records and performs the reconciliation process according to the analysis results. 10 It approves, reviews, or rejects the invention in question. verification of corporate assets, embezzlement control and asset misuse It can be applied in information processing systems where detection is necessary. Within the framework of these fundamental concepts, the subject of the invention is "Corporate Entity Verification 15". Development of various regulations and equivalent practices related to the System(1)" It is possible, and the invention is not limited to the examples described here. Protection of the invention Its scope is essentially as defined in the claims.

Claims

9 REQUESTS 1. During corporate asset reconciliation, the user enters the following into the system: Verification of uploaded asset images, compared with past images. comparison, changes made to the image and detection of manipulations and the uploaded asset image, 5 An invention that can be verified by comparison with records; - the user obtaining images and videos of the asset. Entering the necessary permissions, and the analyses that have been performed. providing an interface to enable the results to be presented to the user at least one electronic device (2), 10 - user, asset, liability, historical video recordings, video signatures, Analysis results, asset characteristics, verification results, and abuse. to ensure the storage of information regarding their evaluations at least one database is structured (3) and, - processing images received via electronic device (2), images 15 creating visual signatures, comparing them with past images, Detecting image manipulations, identifying assets obtaining from the image, the information obtained is in the database (3) compare with registered asset information and according to the results obtained at least one 20-member group structured to evaluate the reconciliation process a system (1) characterized by containing server (4).

2. Asset information, liability information, and reconciliation by the user. Entering the necessary information regarding the transaction, information about the entity from the user Obtaining the necessary permissions to take the images, user presence 25 the input of the image, the image performed by the server (4) the status of the analyses, the reconciliation process, and the analysis results, acceptance of the image, review of the procedure performed or providing the user with information regarding the rejection of a request characterized by an electronic device (2) configured to provide an interface A system like the one in Request 1 (1).

3. User information, assets and liabilities associated with the user. information, asset identification information, product type, brand, model and color 5 information, asset images, image upload times, and uploaded user information, image signatures, past video recordings, visual similarity values, similarity distances, image manipulation analysis results, manipulation risk values, obtained from the image identity information, expected and detected entity characteristics, verification 10 and non-compliance results, abuse risk scores, reconciliation process their status, processing times, analysis results, and control performed. records and evaluation results generated by the system characterized by the database structured for storing (3) a system like any of the above requests (1). 15 4. Images of the presence from the user via the electronic device (2) to ensure that the necessary sharing permissions are requested for acquisition, electronic to receive the asset reconciliation request received via device (2) and the said The request includes user information, asset information, and 20 records relating to the entity requested to be verified from within the database (3) to receive, uploaded received by the user via electronic device (2) taking raw asset photographs, image taken from electronic device (2) to remove noise, resize, and visualize the image. perceptual summarization and / or convolutional neural network to create its signature 25 characterized by the server (4) configured to use its methods a system like any of the above-mentioned requests (1).

5. In another application of the invention, depth is used to create a visual signature. 30 to be able to use embedding and / or convolutional neural network models 11 a request like the one in 4 characterized by the configured server (4) system (1).

6. The generated visual signature is stored in the history database (3). comparing with video recordings, finding similarities between the images 5 to determine the Hamming distance and / or Euclidean distance methods characterized by the server (4) configured for use a system like any of the above requests (1).

7. In another application of the invention, vector 10 is used for the aforementioned search operation. to be able to use a search library and / or vector database a request like the one in 6 characterized by the configured server (4) system (1).

8. Set the calculated hamming distance to a predetermined threshold value of 15. to compare, the Hamming distance is below the threshold value in question. If so, the server (4) will display an image that has been previously uploaded. to identify duplicate usage and stop the reconciliation process, word By identifying the situation as a repeat use detection, an abuse alert has been issued. to create, based on the similarity result, whether the image has been used before 20 to determine if it is not being used and to identify instances of duplicate use and After determining that the image is not a duplicate, the image to determine whether or not there is any digital manipulation on it Server configured to implement error level analysis method (4) a 25 as in any of the above claims characterized by system (1).

9. Manipulation by analyzing regional compression differences in the image. To create a risk assessment, the asset's label, barcode, and identification field... Using the object detection method to determine, 30 from the determined area optical character to obtain serial number and other identification information 12 Using the recognition method, the serial number obtained from the image is retrieved from the data. compare with the asset records in the base (3) and serial number with the brand, model, color and other visual characteristics obtained from the image configured to compare with records in the database (3) any of the above requests characterized by server (4) 5 a system like one of them (1).

10. Perform the comparison in question using the semantic matching method. to perform, discrepancy between serial number and visual specifications If found, the physical label may be used on a different entity. 10 to detect forgery related to, another image taken from the display screen or a previously created one To determine instances of image reuse, forgery using methods for protection against and detection of reuse, After determining that the image had not been used before, image 15 to determine whether or not there is any digital manipulation on it Server configured to implement error level analysis method (4) as in any of the above claims characterized by system (1).

11. Instead of a static image, the user can use an electronic device (2) to analyze the dynamic imaging processes performed to use viability analysis and anti-counterfeiting methods a request like the one in 10 characterized by the configured server (4) system (1). 25 12. Compression levels and regional compression differences in the image. to create a manipulation risk score by analyzing, with that score ranging from 0 to... It can be determined between 100 and 100, and the score must be above 85%. In that case, reject the image or stop the reconciliation process, different 30 relationships between similar images uploaded by users 13 To identify common patterns, perform graphical relationship analysis. relationships consisting of use and similar abusive behaviors by identifying potential links to organized abuse among users characterized by the server (4) configured to detect a system like any of the above requests (1). 5 13. To evaluate and reconcile the results obtained from image analyses. determining the final outcome of the process, verification conditions If provided, approve the reconciliation process; otherwise, discontinue the process. reject or review the results generated and the related technical 10 configured to record the evaluations into the database (3) any of the above requests characterized by the server (4) a system like one of them (1). 20 30

Images