Image authenticity verification method based on block chain

Abstract

The invention relates to an image authenticity verification method based on a block chain, and belongs to the technical field of block chains, and the method comprises the steps: 1, obtaining an image through a trusted data carrier; 2, preprocessing the image based on the smart contract; 3, performing authenticity classification on the preprocessed image through a convolutional neural network model; and step 4, associating the image authenticity result output by the convolutional neural network model with the image based on the smart contract, and performing real-time uploading and evidence storage on process data. According to the method, self-certification clearness of the image data serving as evidence is achieved, the model and service are automatically called in the AI image detection process by defining and using the intelligent contract, human intervention is reduced, authenticity verification is automatically carried out on the image data before uplink, it is guaranteed that the result is not tampered, and the authenticity and reliability of the image data under the link are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of blockchain, and in particular to an image authenticity verification method based on blockchain. Background Art

[0002] Blockchain is essentially a shared database that uses blocks to record transaction records and connects these blocks together through cryptographic methods to form an ever-growing chain. Each block contains a batch of transaction information, and cryptographic technology is used to ensure the immutability and security of the data; blockchain technology, as a distributed accounting book technology, can be used to record, store and verify transactions and data that occur on the chain, solidify off-chain data to form judicial evidence, and once the off-chain data is packaged into blocks and verified and added to the chain, it is difficult to tamper with or delete. However, blockchain technology cannot directly verify whether the off-chain data is authentic and reliable before it is generated and uploaded to the blockchain.

[0003] With the rapid development and popularization of artificial intelligence technology, AI-generated images have reached a very high degree of authenticity, and it is difficult to distinguish them from real photos and videos with the naked eye. The advancement of artificial intelligence technology has brought new challenges to the application of blockchain technology as judicial evidence. Summary of the invention

[0004] In order to solve the above-mentioned problems in the prior art, the blockchain itself does not have the ability to verify the authenticity of off-chain data. That is, if the off-chain data is not authentic, then even if the on-chain data is complete and has not been tampered with, its untrue nature cannot be changed. Therefore, a blockchain-based image authenticity verification method is provided.

[0005] The image authenticity verification method based on blockchain includes the following steps: Step 1: Obtain the image through a trusted data carrier; Step 2: Preprocess the image based on the smart contract; Step 3: Perform authenticity classification on the preprocessed images through the convolutional neural network model; Step 4: Based on the smart contract, the image authenticity result output by the convolutional neural network model is associated with the image, and the process data is stored on the chain in real time.

[0006] Furthermore, the step 1 also includes a user registration step, and the user registration is specifically as follows: the user generates a unique identity identifier when registering an account, and registers a unique device identifier for the image acquisition device, which is used to verify the identity information of the user and the device. When the user stores the image, a request to verify the authenticity of the image is automatically generated based on the smart contract, and the node receives the request and the image.

[0007] Furthermore, the step 2 specifically includes: defining the conditions for triggering image preprocessing in the smart contract, setting up a service for receiving and executing image preprocessing off-chain, automatically executing the image preprocessing task when the smart contract triggering conditions are met, and recording the image processing results, image hash value, user identity identifier, device identity identifier and timestamp.

[0008] Furthermore, the convolutional neural network model is registered with a unique model identifier, and when the convolutional neural network model is called, the model identifier is stored on the chain.

[0009] Furthermore, the step 3 specifically includes: the smart contract obtains the status of the image to be detected having completed preprocessing, automatically loads the image to be detected into one or more convolutional neural network models, the convolutional neural network model classifies the image authenticity according to the image features, and the image authenticity results include the original image and the AI ​​generated image.

[0010] Furthermore, the convolutional neural network model classifies image authenticity according to image features, specifically including: inputting training set data to train the convolutional neural network model to identify original images and AI-generated images, the convolutional neural network model extracts and integrates the features of the image to be detected, outputs the image authenticity classification results through the forward propagation algorithm, and iteratively updates the model parameters through the back propagation algorithm.

[0011] Furthermore, the step 4 specifically includes: obtaining the image authenticity result output by the neural network model; packaging the unique model identifier of the neural network model, the image authenticity result, the timestamp, the image hash value, the user identity identifier and the device identifier into the transaction based on the smart contract, and automatically storing them on the chain; if one convolutional neural network model is called, one transaction is stored; if multiple convolutional neural network models are called, multiple transactions are stored accordingly.

[0012] Furthermore, the data in the process of acquiring the image, preprocessing the image, and detecting the authenticity of the image are stored in real time through the blockchain.

[0013] Beneficial effects of the present invention: The present invention enables image data to be self-proven as evidence. By defining and using smart contracts, models and services are automatically called during the AI ​​image detection process to reduce human intervention. The authenticity of image data before being uploaded to the chain is automatically verified to ensure that the results are not tampered with, thereby improving the authenticity and reliability of off-chain image data. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] In this embodiment, refer to Figure 1 , aims to realize a blockchain-based image authenticity verification method, improve the authenticity and reliability of off-chain image data, protect intellectual property rights, and combat false information; the method includes the following steps: Step 1: When registering an account, the user generates a unique identity identifier and registers a unique device identifier for the image acquisition device to verify the identity information of the user and the device. When the user stores the image, a request to verify the authenticity of the image is automatically generated based on the smart contract, and the node receives the request and image; the image is obtained through a trusted data carrier; Step 2: Preprocess the image based on the smart contract; specifically: define the conditions that trigger image preprocessing in the smart contract, set up a service to receive and execute image preprocessing off-chain, automatically execute the image preprocessing task when the smart contract trigger conditions are met, and record the image processing results, image hash value, user identity identifier, device identity identifier, timestamp and other related status; Step 3: The pre-processed images are classified for authenticity through a convolutional neural network model. Specifically, the smart contract obtains the status of the pre-processed image to be detected, and automatically loads the image to be detected into one or more convolutional neural network models. The convolutional neural network model classifies the image authenticity according to the image features. The image authenticity results include the original image and the AI-generated image.

[0017] Among them, the convolutional neural network model is registered with a unique model identifier. When the convolutional neural network model is called, the model identifier is stored on the chain.

[0018] Step 4: Based on the smart contract, the image authenticity result output by the convolutional neural network model is associated with the image, and the process data is stored on the chain in real time. Specifically, it includes: obtaining the image authenticity result output by the neural network model; based on the smart contract, the unique model identifier of the neural network model, the image authenticity result, the timestamp, the image hash value, the user identity identifier, the device identifier and other related information are packaged into the transaction, and automatically stored on the chain; if a convolutional neural network model is called, one transaction is stored, and if multiple convolutional neural network models are called, multiple transactions are stored accordingly.

[0019] In this embodiment, the convolutional neural network model is trained by inputting training set data to recognize original images and AI-generated images. The convolutional neural network model extracts and integrates the features of the image to be detected, outputs the image authenticity classification results through the forward propagation algorithm, and iteratively updates the model parameters through the back propagation algorithm.

[0020] In this embodiment, data in the process of image acquisition, image preprocessing, and image authenticity detection is stored in real time through blockchain.

[0021] A single blockchain evidence cannot guarantee the authenticity of off-chain data, and a single AI image detection is difficult to avoid human intervention. This solution enables image data to prove its innocence as evidence. By combining blockchain smart contracts and AI image detection, the authenticity of image data before being uploaded to the chain is automatically verified and the results are guaranteed not to be tampered with, so as to improve the authenticity and reliability of off-chain image data.

[0022] In the description of the embodiments of the present invention, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "center", "top", "bottom", "top", "bottom", "inside", "outside", "inner side", "outer side" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Among them, "inside" refers to an internal or enclosed area or space. "Periphery" refers to the area surrounding a specific component or a specific area.

[0023] In the description of the embodiments of the present invention, the terms "first", "second", "third", and "fourth" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", and "fourth" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0024] In the description of the embodiments of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "install", "connect", "connect", and "assemble" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0025] In the description of the embodiments of the present invention, specific features, structures, materials or characteristics may be combined in a suitable manner in any one or more embodiments or examples.

[0026] In the description of the embodiments of the present invention, it should be understood that "-" and "~" represent a range between two values, and the range includes the endpoints. For example: "AB" represents a range greater than or equal to A and less than or equal to B. "A~B" represents a range greater than or equal to A and less than or equal to B.

[0027] In the description of the embodiments of the present invention, the term "and / or" herein is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " herein generally indicates that the associated objects before and after are in an "or" relationship.

[0028] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The image authenticity verification method based on blockchain is characterized by: The following steps are involved: Step 1: Obtain the image through a trusted data carrier; Step 2: Preprocess the image based on the smart contract; Step 3: Perform authenticity classification on the preprocessed images through the convolutional neural network model; Step 4: Based on the smart contract, the image authenticity result output by the convolutional neural network model is associated with the image, and the process data is stored on the chain in real time.

2. The image authenticity verification method based on blockchain according to claim 1 is characterized in that: The step 1 also includes a user registration step. Specifically, the user registration includes: when registering an account, the user generates a unique identity identifier and registers a unique device identifier for the image acquisition device to verify the identity information of the user and the device. When the user stores the image, a request for verifying the authenticity of the image is automatically generated based on the smart contract, and the node receives the request and the image.

3. The image authenticity verification method based on blockchain according to claim 1 is characterized in that: The step 2 specifically includes: defining the conditions for triggering image preprocessing in the smart contract, setting up a service for receiving and executing image preprocessing off-chain, automatically executing the image preprocessing task when the smart contract triggering conditions are met, and recording the image processing results, image hash value, user identity identifier, device identity identifier and timestamp.

4. The image authenticity verification method based on blockchain according to claim 1 is characterized in that: The convolutional neural network model is registered with a unique model identifier. When the convolutional neural network model is called, the model identifier is stored on the chain.

5. The image authenticity verification method based on blockchain according to claim 1 is characterized in that: The step 3 specifically includes: the smart contract obtains the status of the image to be detected having completed preprocessing, automatically loads the image to be detected into one or more convolutional neural network models, and the convolutional neural network model classifies the image authenticity according to the image features. The image authenticity results include the original image and the AI ​​generated image.

6. The image authenticity verification method based on blockchain according to claim 5 is characterized in that: The convolutional neural network model classifies the authenticity of images according to image features, specifically including: inputting training set data to train the convolutional neural network model to identify original images and AI-generated images, the convolutional neural network model extracts and integrates the features of the images to be detected, outputs the image authenticity classification results through the forward propagation algorithm, and iteratively updates the model parameters through the back propagation algorithm.

7. The image authenticity verification method based on blockchain according to claim 1 is characterized in that: The step 4 specifically includes: obtaining the image authenticity result output by the neural network model; packaging the unique model identifier of the neural network model, the image authenticity result, the timestamp, the image hash value, the user identity identifier and the device identifier into the transaction based on the smart contract, and automatically storing them on the chain; if one convolutional neural network model is called, one transaction is stored; if multiple convolutional neural network models are called, multiple transactions are stored accordingly.

8. The image authenticity verification method based on blockchain according to claim 1 is characterized in that: The data in the process of acquiring the image, preprocessing the image, and detecting the authenticity of the image are stored in real time through the blockchain.

Images