A system for identity control and fraud detection with artificial intelligence and image processing technologies

Abstract

The present invention relates to a system (1) which is developed for carrying out processes of digital identity verification and detecting fake passports or IDs, by using artificial intelligence and image processing technologies.

Description

[0001] A SYSTEM FOR IDENTITY CONTROL AND FRAUD DETECTION WITH ARTIFICIAL INTELLIGENCE AND IMAGE PROCESSING TECHNOLOGIES

[0002] Technical Field

[0003] The present invention relates to a system for carrying out processes of digital identity verification and detecting fake passports or IDs, by using artificial intelligence and image processing technologies.

[0004] Background of the Invention

[0005] Today, there are many technical challenges and problems in terms of authentication and fraud detection. Current techniques are often be slow, may lack precision and be poor at fraud detection. Some of the current techniques may make authentication processes slow. Processing speed can be a significant issue, particularly when handling large data masses. Some techniques are not precise enough to detect fraud. Especially when confronted with sophisticated forgery methods, these techniques may be particularly inefficient.

[0006] Therefore, it is understood that there is need for a system for carrying out processes of digital identity verification and detecting fake passports or IDs, by using artificial intelligence and image processing technologies.

[0007] The International patent document no. WO2021054923, an application included in the state of the art, discloses a system for carrying out transactions of security document control and verification. The invention subject to the said international patent document relates to a system for performing security document control and verification transactions such as identification, verification, fraud control and detection by processing security documents such as identity cards, passports, driver’s licenses by using image processing technologies.

[0008] Summary of the Invention

[0009] An objective of the present invention is to realize a system which is developed for carrying out processes of digital identity verification and detecting fake passports or IDs, by using artificial intelligence and image processing technologies.

[0010] Another objective of the present invention is to realize a system which is developed for reducing financial losses by detecting fake IDs or fake passports.

[0011] Detailed Description of the Invention

[0012] “A System for Identity Control and Fraud Detection with Artificial Intelligence and Image Processing Technologies” realized to fulfil the objectives of the present invention is shown in the figure attached, in which:

[0013] Figure 1 is a schematic view of the inventive system.

[0014] The components illustrated in the figure are individually numbered, where the numbers refer to the following:

[0015] 1. System 2. Electronic Device

[0016] 3. Server

[0017] The inventive system (1) which is developed for carrying out processes of digital identity verification and detecting fake passports or IDs, by using artificial intelligence and image processing technologies; comprises at least one electronic device (2) which is configured to take high-resolution images of passports or identity documents; at least one server (3) which is configured to pre-process images for noise reduction and contrast enhancement; to extract various features from the images; to analyze the features for authentication and fraud detection by using artificial intelligence algorithms; to generate and report authentication and fraud analysis results; to transmit the results to users or related individuals; and to provide secure digital authentication.

[0018] The electronic device (2) included in the inventive system (1) is configured to establish communication with the server (3) by any communication protocol. The electronic device (2) is configured to transmit high-resolution images of passports or IDs to the server (3).

[0019] The server (3) included in the inventive system (1) is configured to establish communication with the electronic device (2) by any communication protocol. The server (3) is configured to receive high-resolution images of passports or IDs from the electronic device (2) and to split the data appropriately for training, verification and testing. The server (3) is configured to subject the image data to pre-processing transactions and then to turn them into a noise reduced and contrast enhanced form. The server (3) is configured to build a deep learning model by using a CNN (convolutional neural network) architecture. The server (3) is configured to train the deep learning model in order to sense and detect holograms. The server (3) is configured to ensure that the input layer of the model matches the dimensions of the prepared images. The server (3) is configured to ensure that the CNN layers are used in order to learn the features of the images. The server (3) is configured to enable the model to leam by using these layers in order to detect specific features of the holograms. The server (3) is configured to enable the final layers to be used in order to predict the presence or absence of holograms. The server (3) is configured to start training the model on the training data once it is ready. The server (3) is configured to ensure that the model correctly recognizes the holograms. The server (3) is configured to train the model by using loss functions and optimizing algorithms. The server (3) is configured to iteratively improve your model using the training data. The server (3) is configured to optimize the model in multiple iterations during training with the BFGS (Broyden-Fletcher-Goldfarb-Shanno) algorithm. The server (3) is configured to minimize the loss of the model in each iteration by the BFGS (Broyden-Fletcher- Goldfarb-Shanno) algorithm. The server (3) is configured to monitor the model parameters and loss updated by the BFGS algorithm during training. The server (3) is configured to terminate training if it detects that the model no longer improves or the loss value does not change in a given training iteration. The server (3) is configured to terminate training in order to avoid unnecessary computation time and to reduce the risk of overfitting. The server (3) is configured to evaluate the model on test data and to perform hologram detection. The server (3) is configured to ensure that the model correctly detects holograms in the test data. The server (3) is configured to measure the performance of the model by evaluating the precision and specificity values. The server (3) is configured to extract various features from the images such as face recognition, iris scanning. The server (3) is configured to analyze the extracted features for authentication and fraud detection by using artificial intelligence algorithms. The server (3) is configured to generate and report results of the authentication and fraud analysis. The server (3) is configured to transmit the results of the analysis to users or related persons and to provide secure digital authentication.

[0020] Industrial Application of the Invention

[0021] With the inventive system (1), it is ensured to carry out processes of digital identity verification and to detect fake passports or IDs, by using artificial intelligence and image processing technologies. Within these basic concepts; it is possible to develop various embodiments of the inventive “System (1) for Identity Control and Fraud Detection with Artificial Intelligence and Image Processing Technologies”; the invention cannot be limited to examples disclosed herein and it is essentially according to claims.

Claims

CLAIMS1. A system (1) which is developed for carrying out processes of digital identity verification and detecting fake passports or IDs, by using artificial intelligence and image processing technologies; comprising at least one electronic device (2) which is configured to take high- resolution images of passports or identity documents; and characterized by at least one server (3) which is configured to pre-process images for noise reduction and contrast enhancement; to extract various features from the images; to analyze the features for authentication and fraud detection by using artificial intelligence algorithms; to generate and report authentication and fraud analysis results; to transmit the results to users or related individuals; and to provide secure digital authentication.

2. A system (1) according to Claim 1; characterized by the electronic device (2) which is configured to establish communication with the server (3) by any communication protocol.

3. A system (1) according to Claim 1 or 2; characterized by the electronic device (2) which is configured to transmit high-resolution images of passports or IDs to the server (3).

4. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to establish communication with the electronic device (2) by any communication protocol.

5. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to receive high-resolution images of passports or IDs from the electronic device (2) and to split the data appropriately for training, verification and testing.

6. A system (1) according to any of the preceding claims; characterized by the server (3) which subject the image data to pre-processing transactions and then to turn them into a noise reduced and contrast enhanced form.

7. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to build a deep learning model by using a CNN (convolutional neural network) architecture.

8. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to train the deep learning model in order to sense and detect holograms.

9. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to ensure that the input layer of the model matches the dimensions of the prepared images.

10. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to ensure that the CNN (convolutional neural networks) layers are used in order to learn the features of the images.

11. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to enable the model to learn by using these layers in order to detect specific features of the holograms.

12. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to enable the final layers to be used in order to predict the presence or absence of holograms.

13. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to start training the model on the training data once it is ready.

14. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to ensure that the model correctly recognizes the holograms.

15. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to train the model by using loss functions and optimizing algorithms.

16. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to iteratively improve your model using the training data.

17. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to optimize the model in multiple iterations during training with the BFGS (Broyden-Fletcher-Goldfarb-Shanno) algorithm.

18. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to minimize the loss of the model in each iteration by the BFGS (Broyden-Fletcher-Goldfarb-Shanno) algorithm.

19. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to monitor the model parameters and loss updated by the BFGS algorithm during training.

20. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to terminate training if it detects that the model no longer improves or the loss value does not change in a given training iteration.

21. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to terminate training in order to avoid unnecessary computation time and to reduce the risk of overfitting.

22. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to evaluate the model on test data and to perform hologram detection.

23. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to ensure that the model correctly detects holograms in the test data.

24. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to measure the performance of the model by evaluating the precision and specificity values.

25. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to extract various features from the images such as face recognition, iris scanning.

26. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to analyze the extracted features for authentication and fraud detection by using artificial intelligence algorithms.

27. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to generate and report results of the authentication and fraud analysis.

28. A system (1) according to any of the preceding claims; characterized by the server (3) which is configured to transmit the results of the analysis to users or related persons and to provide secure digital authentication.

Images