Image recognition system based on multi-factor authentication, block chain and network security visualization

Abstract

The invention discloses an image recognition and authentication system and method based on multi-factor authentication, a block chain and network security visualization. The system comprises a client, a server, a data processing and transmission device, a block chain technology, a multi-factor authentication module, an image processing module, a network security data visualization module and an image recognition module. The client side obtains a user image through the mobile device, and the server side carries out image processing and authentication. The image processing module carries out Hash processing, feature extraction and wavelet transformation on an image uploaded by a user to decompose the image. The block chain technology is used for storing and managing image data and ensuring the security and non-tampering of the data. And the multi-factor authentication module performs user authentication through user name, password and image recognition. The network security data visualization module integrates data of various data sources and displays network security conditions through a visualization technology. The image recognition module uses a convolutional neural network (CNN) model and an F-CDCGAN model to carry out image recognition, so that the accuracy of image recognition is improved. The system provides high-security and high-reliability image recognition and authentication services, effectively prevents image data from being tampered and resists various attacks.

Description

Technical Field

[0001] The present invention relates to the field of image recognition and authentication technology, and in particular to an image recognition and authentication system and method based on multi-factor authentication, blockchain and network security visualization. The system aims to ensure the integrity and confidentiality of image data through multiple security measures, thereby improving the security and reliability of image recognition and user authentication, and is widely used in security monitoring, identity authentication, smart devices and other fields. Background Art

[0002] With the rapid development of the Internet and information technology, image recognition technology has been widely used in various fields. Image recognition technology is mainly used to identify and verify the identity of an object by analyzing and processing image data, including face recognition, fingerprint recognition, iris recognition, etc. However, the existing image recognition system has the following major problems:

[0003] a) Low data security: In traditional image recognition systems, image data is usually stored on centralized servers, which makes the data easy to be attacked. Once the server is attacked, the image data may be stolen or tampered with, resulting in user privacy leakage and reduced system security.

[0004] b) Easy to be tampered: Image data may be maliciously tampered during transmission and storage, resulting in incorrect identification during the authentication process. This tampering behavior will not only affect the normal operation of the system, but may also cause serious security issues.

[0005] c) Weak anti-attack capability: Existing image recognition systems have weak anti-attack capabilities when facing complex and diverse attacks. For example, it is difficult for existing systems to effectively defend against adversarial attacks and forgery attacks on image recognition systems and they are easily breached.

[0006] d) Limitations of a single authentication method: Traditional image recognition systems usually rely on a single authentication method, such as identity authentication through face recognition alone. However, a single authentication method has certain limitations and it is difficult to ensure the absolute security of the authentication process. For example, when authentication is performed only through face recognition, it is easy to be deceived by attacks such as photos and videos.

[0007] e) Lack of data visualization analysis: The existing system lacks the means to visualize and analyze network security data, making it difficult to intuitively display and monitor the security status of the system, and unable to promptly detect and respond to potential security threats.

[0008] In order to solve the above problems, it is particularly important to provide an image recognition and authentication system based on multi-factor authentication, blockchain and network security visualization. By combining a variety of security technologies and means, the security and reliability of the system can be effectively improved, image data can be prevented from being tampered with or stolen, and the system's anti-attack ability can be enhanced. At the same time, the decentralization and immutability of blockchain technology can ensure the integrity and confidentiality of image data; multi-factor authentication can improve the security and accuracy of the authentication process; and through the visualization analysis of network security data, potential security threats can be discovered and responded to in a timely manner to ensure the overall security of the system. Summary of the invention

[0009] The present invention provides an image recognition and authentication system and method based on multi-factor authentication, blockchain and network security visualization. The system ensures the integrity and confidentiality of image data by combining multiple security technologies and means, thereby improving the security and reliability of image recognition and user authentication. The specific implementation plan is as follows:

[0010] This system includes client, server, data processing and transmission equipment, blockchain technology, multi-factor authentication module, image processing module, network security data visualization module and image recognition module. The following is a detailed description of each module and how they interact:

[0011] (1) Client obtains image: Users take or upload images for authentication through the client application on a mobile device (such as a smartphone or tablet). The client application provides a user-friendly interface that enables users to easily take and upload images. The user interface is intuitive, simple and easy to use, and supports uploading multiple image formats. In addition, the client application has a data encryption transmission function, using advanced encryption algorithms to ensure the security of image data during transmission and prevent data from being intercepted or tampered with during transmission.

[0012] (2) Data transmission to the server: The captured or uploaded image data is transmitted to the server through an encrypted channel. In order to ensure the security and stability of the transmission, the system adopts secure transmission protocols such as HTTPS and TLS. The Raspberry Pi device serves as an intermediate device for data processing and transmission to ensure the security and stability of data transmission. Raspberry Pi is not only responsible for data transmission, but also performs preliminary data processing such as compression and format conversion. Raspberry Pi is also responsible for running blockchain nodes, participating in the consensus and data storage of the blockchain network, and ensuring the distributed storage and immutability of data.

[0013] (3) Image processing: After receiving the image data, the server first performs hash processing on the image and uses the SHA256 algorithm to generate a unique hash value to ensure the integrity and tamper-proofness of the data. The hash value generated by the hash processing serves as the unique identifier of the image data for subsequent verification and authentication. Then, the server uses the Haar cascade classifier to extract the facial features of the image to ensure the accuracy and robustness of recognition. The feature extraction process includes detecting facial areas, extracting key points, and calculating feature vectors. These feature vectors are used for subsequent image comparison and recognition. Subsequently, the image is decomposed through wavelet transform to hide the details of the image, improve the security of the image data, and prevent it from being tampered with or abused. Wavelet transform can retain the overall structural information of the image while hiding the details to protect the privacy of the image.

[0014] (4) Data storage in blockchain: The processed image data, hash value, and feature file are passed to Raspberry Pi, new blocks are created in the blockchain, and these files are stored in the distributed ledger. The application of blockchain technology ensures the immutability and security of data. The image data and authentication records of each authentication process are packaged into blocks and added to the blockchain, using the distributed ledger technology of blockchain to ensure the transparency and immutability of data. The blockchain network consists of multiple nodes, and the consensus mechanism between nodes ensures the consistency and security of data.

[0015] (5) Multi-factor authentication process: The user first enters the username and password, and the system randomly displays a set of mixed images (including images uploaded by the user and general images generated by the system). The user needs to select his own image from them and correctly establish the relationship between the images to complete the multi-factor authentication. This process effectively prevents photo and video attacks and improves the security of the system. In addition, the multi-factor authentication module also supports the integration of biometric recognition (such as fingerprints, irises) and physical security keys to further improve the security of authentication. After the user completes the above steps, the system will verify the user's selection. If the verification is successful, the user will successfully complete the authentication.

[0016] (6) Image recognition: After completing multi-factor authentication, the system uses a convolutional neural network (CNN) model and a generative adversarial network (F-CDCGAN) model to recognize images. The CNN model is used to extract deep features of images, which contain key information about the image and can effectively distinguish different users. The F-CDCGAN model further improves the accuracy and robustness of recognition through an adversarial training mechanism. The image recognition module continuously updates and trains the model to improve recognition performance and adaptability, ensuring that the system can accurately distinguish different users. The recognition process includes feature extraction, feature comparison, and final identity confirmation.

[0017] (7) Network security data visualization: The system integrates multiple data sources and displays network security status through visualization technology. The network security data visualization module uses tools such as Echarts, Jtopo, and JfreeChart to generate real-time security situation diagrams to help security analysts intuitively monitor and analyze the security status of the system and promptly identify potential security threats. Visual reports include network traffic analysis, attack event detection, and security incident response, providing comprehensive security situation awareness. Through visualization, security analysts can quickly identify abnormal behavior and take timely measures to ensure the overall security of the system.

[0018] Through the mutual cooperation of the above modules, the image recognition and authentication system of the present invention provides a highly secure and reliable solution, which is widely applicable to the fields of security monitoring, identity authentication, smart devices, etc. The system ensures the security of image data, the reliability of the authentication process, and the visualization of network security status through the combination of multi-factor authentication, blockchain technology, and network security data visualization technology, and has important practical application value and broad market prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is the overall system architecture diagram of the present invention;

[0020] Figure 2 It is the image processing flow chart; DETAILED DESCRIPTION

[0021] This invention introduces an image recognition and authentication system and method that integrates multi-factor authentication, blockchain technology, and network security visualization. The system uses a variety of advanced security technologies to ensure the integrity and confidentiality of image data, thereby improving the security and reliability of image recognition and user authentication. The specific implementation plan is as follows:

[0022] The system includes client, server, data processing and transmission equipment, blockchain technology, multi-factor authentication module, image processing module, network security data visualization module and image recognition module. The following is a detailed description of each module and how they interact with each other:

[0023] (1) Client obtains image: Users take or upload images for authentication through the client application on a mobile device (such as a smartphone or tablet). The client application has a simple and easy-to-use user interface that allows users to easily capture and upload images. The application supports multiple image formats and has a built-in data encryption transmission function. It uses advanced encryption algorithms to ensure the security of image data during transmission and prevent data from being intercepted or tampered with.

[0024] (2) Data transmission to the server: Image data is transmitted from the client to the server through an encrypted channel. The system uses secure transmission protocols such as HTTPS and TLS to ensure the security and stability of transmission. The Raspberry Pi device acts as an intermediate device in the data transmission and processing process, responsible for preliminary data processing (such as compression and format conversion) and ensuring the security and stability of data transmission. In addition, Raspberry Pi also runs a blockchain node, participates in the consensus and data storage of the blockchain network, and ensures distributed data storage and immutability.

[0025] (3) Image processing: After receiving the image data, the server first uses the SHA 256 algorithm to hash the image and generate a unique hash value to ensure the integrity and tamper-proofness of the data. The generated hash value serves as the unique identifier of the image data for subsequent verification and authentication. Next, the server uses the Haar cascade classifier to extract the facial features of the image to ensure the accuracy and robustness of recognition. The feature extraction process includes facial region detection, key point extraction, and feature vector calculation, which are used for subsequent image comparison and recognition. Subsequently, the server performs a wavelet transform on the image to hide the details of the image, improve data security, and prevent the image from being tampered with or abused.

[0026] (4) Data storage in blockchain: The processed image data, hash value and feature files are transmitted through the RaspberryPi device, and new blocks are created in the blockchain to store these files in the distributed ledger. Blockchain technology ensures the immutability and security of data. The image data and authentication records of each authentication process are packaged into blocks and added to the blockchain, using distributed ledger technology to ensure data transparency and immutability. The blockchain network consists of multiple nodes, and the nodes ensure data consistency and security through a consensus mechanism.

[0027] (5) Multi-factor authentication process: The user first enters the username and password, and the system randomly displays a set of mixed images (including images uploaded by the user and general images generated by the system). The user needs to select his own image from them and correctly establish the relationship between the images to complete the multi-factor authentication. This process effectively prevents photo and video attacks and enhances system security. In addition, the multi-factor authentication module also supports the integration of biometric recognition (such as fingerprints, irises) and physical security keys to further enhance the security of authentication. After the user completes the above steps, the system will verify the user's selection. If the verification is successful, the user has successfully completed the authentication.

[0028] (6) Image recognition: After completing multi-factor authentication, the system uses a convolutional neural network (CNN) model and a generative adversarial network (F-CDCGAN) model to recognize images. The CNN model extracts deep features of the image, which contain key information about the image and can effectively distinguish different users. The F-CDCGAN model further improves the accuracy and robustness of recognition through an adversarial training mechanism. The image recognition module continuously updates and trains the model to improve recognition performance and adaptability, ensuring that the system can accurately distinguish different users. The recognition process includes feature extraction, feature comparison, and final identity confirmation.

[0029] (7) Network security data visualization: The system integrates multiple data sources and displays network security status through visualization technology. The network security data visualization module uses tools such as Echarts, Jtopo, and JfreeChart to generate real-time security situation diagrams to help security analysts intuitively monitor and analyze the security status of the system and promptly identify potential security threats. Visual reports include network traffic analysis, attack event detection, and security incident response, providing comprehensive security situation awareness. Through visualization, security analysts can quickly identify abnormal behavior and take timely measures to ensure the overall security of the system.

[0030] The image recognition and authentication system of the present invention provides a highly secure and reliable solution through the collaborative work of multiple modules, and is widely used in the fields of security monitoring, identity authentication, and smart devices. The system combines multi-factor authentication, blockchain technology, and network security data visualization technology to ensure the security of image data, the reliability of the authentication process, and visualize the network security status, which has important practical application value and broad market prospects.

Claims

1. An image recognition and authentication system based on multi-factor authentication, blockchain and network security visualization, characterized in that: include: a) Client, obtains user images through mobile devices; b) Server side, image processing and authentication; c) Data processing and transmission equipment, using Raspberry Pi as data processing and transmission equipment; d) Blockchain technology for storing and managing image data; e) Multi-factor authentication module for user authentication through username, password and image recognition.

2. The system according to claim 1, characterized in that The multi-factor authentication module includes: a) an authentication unit for a user to input a username and password; b) A unit for randomly selecting images uploaded by users and displaying them mixed with common images, and requiring users to select the correct image; c) A unit for establishing and verifying image relationships.

3. The system according to claim 1, characterized in that The image processing module comprises: a) A unit for hashing images uploaded by users, using the SHA 256 algorithm to generate hash values; b) a unit for extracting facial features of an image, using a Haar cascade classifier; c) A unit for decomposing an image by wavelet transform to hide image details.

4. The system according to claim 1, characterized in that The blockchain technology includes: a) used to pass the processed image, hash value and signature file to the Raspberry Pi, and the unit that creates new blocks in the blockchain; b) A unit for storing these documents in the ledger.

5. The system according to claim 1, characterized in that The network security data visualization module includes: a) A unit used to aggregate data from multiple data sources, including packets, network traffic, and log files; b) A unit used to display the network security status through visualization technology; c) Unit for data visualization using tools such as Echarts, Jtopo and JfreeChart.

6. The system according to claim 1, characterized in that The image recognition module comprises: a) a unit for image recognition using a convolutional neural network (CNN) model; b) Unit used to improve image recognition accuracy in combination with the F-CDCGAN model.

7. The system according to claim 6, characterized in that The F-CDCGAN model is used to train the MNIST dataset and improves the accuracy of image recognition to 99.46%.

8. The system according to claim 1, characterized in that The high-pass filter fusion technology is used to extract high-frequency components of the high-resolution image and fuse them with the low-resolution image to enhance the edge features of the fused image.

9. An image recognition method based on the system according to any one of claims 1 to 8, comprising the following steps: a) Obtain user images through mobile devices; b) Perform hashing, feature extraction and wavelet transform to decompose the image; c) storing processed images, hash values ​​and feature files in the blockchain; d) User authentication through a multi-factor authentication module; e) Image recognition using convolutional neural network and F-CDCGAN model.

10. The image recognition method according to claim 9, characterized in that: The high-pass filter fusion technology is used to extract the high-frequency components of the image and perform image fusion to enhance the accuracy of image recognition.

Images