CN114359590A - NFT image work infringement detection method, device, and computer storage medium - Google Patents

Abstract

An embodiment of the present specification provides an NFT image work infringement detection method, an NFT image work infringement detection device, and a computer storage medium, where the method may include: comparing the NFT work with image characteristics of a plurality of images in an image database to obtain the similarity between the NFT work and each image in the image database; if the similarity between the NFT work and any image in the image database is greater than a preset similarity threshold, it may be determined that the NFT work is infringing.

Description

NFT image work infringement detection method, device, and computer storage medium

technical field

The embodiments of this specification relate to the field of computer technology, and in particular, to a method, device, and computer storage medium for detecting infringement of NFT image works.

Background technique

With the rapid development of Internet technology, a large number of original image works have been widely used in various fields of society, but the problem of infringement of corresponding original works has become more and more serious. (Non-Fungible Token, NFT) market for profit, some criminals often use image editing and other means to make similar digital image artworks to make huge profits.

Therefore, there are more and more legal disputes caused by the infringement of NFT works, and the rights and interests of the original authors have been violated. In view of this, how to detect the infringement of NFT works has become an urgent problem to be solved.

SUMMARY OF THE INVENTION

The embodiments of this specification provide a method, a device, and a computer storage medium for detecting infringement of an NFT image work, which can effectively detect whether an NFT work is infringing.

In the first aspect, the embodiments of this specification provide a method for detecting infringement of a non-fungible token NFT image work, the method comprising:

Comparing the image features of the NFT image work to be detected and multiple images in the image database to obtain the similarity between the NFT image work to be detected and the images in the image database;

If the similarity between the NFT image work to be detected and any one of the images in the image database is greater than a preset similarity threshold, it is determined that the NFT image work to be detected is infringing.

In the second aspect, this specification provides a non-fungible token NFT image work infringement detection device, the device includes:

A comparison module, configured to compare the image features of the NFT image work to be detected and the images in the image database to obtain the similarity between the NFT image work to be detected and the images in the image database;

A determining module, configured to determine that the NFT image work to be detected is infringing if the similarity between the NFT image work to be detected and any one of the images in the image database is greater than a preset similarity threshold.

In a third aspect, the embodiments of the specification provide a computer storage medium, where the computer storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the above-mentioned method steps.

In a fourth aspect, the embodiments of this specification provide an electronic device, which may include: a processor and a memory;

Wherein, the memory stores a computer program, and the computer program is adapted to be loaded by the processor and execute the above-mentioned method steps.

The beneficial effects brought by the technical solutions provided by some embodiments of this specification include at least:

This specification can obtain the similarity between the NFT image work to be detected and the images in the image database by comparing the image features of the NFT image work to be detected and the images in the image database; if the NFT image work to be detected is similar to the image If the similarity of any image in the database is greater than the preset similarity threshold, it is determined that the NFT image work to be detected is infringing. Therefore, on the one hand, this specification uses the method of image feature comparison to complete the infringement detection of NFT works, so as to avoid the inability to detect infringement due to slight modifications to the NFT works, such as scaling or editing a small number of pixels, and improve the accuracy of infringement detection. Rate. On the other hand, the creator does not need to provide a lot of evidence to prove the ownership information of the image, which saves a lot of manpower and material resources, avoids tedious work such as manual review and uploading review results, and improves the efficiency of infringement detection.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present specification, the accompanying drawings required in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present specification. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without any creative effort.

Fig. 1a is an NFT image work to be detected provided by the embodiment of this specification;

Fig. 1b is an image in a hot news provided by an embodiment of the present specification;

2 is a schematic flowchart of a method for detecting infringement of an NFT image work provided by an embodiment of the present specification;

3 is a schematic flowchart of another method for detecting infringement of NFT image works provided by the embodiment of the present specification;

Fig. 4a is another NFT image work to be detected provided by the embodiment of this specification;

Fig. 4b is a local feature map of another NFT image work to be detected provided by the embodiment of this specification;

Fig. 4c is another partial feature map of another NFT image work to be detected provided by the embodiment of this specification;

5 is a schematic flowchart of yet another method for detecting infringement of an NFT image work provided by an embodiment of the present specification;

6 is an application scenario diagram of a method for detecting infringement of an NFT image work provided by an embodiment of the present specification;

7 is a schematic structural diagram of an NFT image work infringement detection device provided by an embodiment of the present specification;

FIG. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present specification.

Detailed ways

Where the following description refers to the drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the illustrative examples below are not intended to represent all implementations consistent with this specification. Rather, they are merely examples of apparatus and methods consistent with some aspects of this specification, as recited in the appended claims.

In the description of this specification, it should be understood that the terms "first", "second" and the like are used for descriptive purposes only, and cannot be construed as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in this specification can be understood in specific situations. Also, in the description of this specification, unless otherwise specified, "a plurality" means two or more. "And/or", which describes the association relationship of the associated objects, means that there can be three kinds of relationships, for example, A and/or B, which can mean that A exists alone, A and B exist at the same time, and B exists alone. The character "/" generally indicates that the associated objects are an "or" relationship.

With the rapid development of blockchain technology, digital artwork NFT transactions based on blockchain infrastructure have also ushered in an opportunity for development. Specifically, NFT is a data unit called a blockchain digital ledger, and each NFT can represent a unique digital data. Since they are not interchangeable, non-fungible tokens can represent digital files such as paintings, sounds, videos, in-game items, or other forms of creative work. Further, NFTs can be used for verification of ownership and allow NFT-represented NFT image works to be traded and sold on digital marketplaces.

Possibly, the NFT image work trading platform may include the image work uploaded by the user and the NFT corresponding to the image work. For example, OpenSea is currently the world's largest comprehensive NFT image work trading platform. Users can mint, display, trade, and auction NFTs on the platform.

As shown in Figure 1a, the owner of the ownership of the NFT image work is Xiao Wang, and Xiao Wang can trade it through NFT on the NFT image work trading platform, that is, sell the ownership of the NFT image work to others. As shown in Figure 1b, the image in the hot news is owned by Xiaozhang. Looking at Figure 1a and Figure 1b, it can be found that the content in the two images is the same, and the only difference between them is that Figure 1a is in Figure 1b. A virtual border is added to the image works of , and accordingly, the NFTs generated in Figure 1a and Figure 1b are different. Therefore, if the real holder of the NFT image work is Xiao Zhang, then Xiao Wang infringes the ownership of Xiao Zhang when he trades the NFT work on the NFT image work trading platform. Therefore, this specification proposes a method for detecting infringement of NFT image works on the NFT image work trading platform, so as to reduce infringement disputes on the NFT image work trading platform.

Next, the method for detecting infringement of NFT image works provided by one or more embodiments of this specification will be introduced in conjunction with the NFT image works corresponding to the NFTs described in FIG. 1a and FIG. 1b. The NFT image work infringement detection method is used for the non-fungible token NFT image work trading platform.

In one or more embodiments of this specification, as shown in FIG. 2 , a schematic flowchart of a method for detecting infringement of an NFT image work is provided. As shown in Figure 2, the method for detecting infringement of NFT image works may include the following steps:

Step 202: Compare the image features of the NFT image work to be detected and the images in the image database to obtain the similarity between the NFT image work to be detected and the images in the image database.

Wherein, the NFT image work to be detected in the embodiment of this specification is an NFT work.

Optionally, the NFT image work to be detected is an NFT image work in the NFT image work trading platform. Due to the huge number of NFT image works on the NFT image works trading platform, in order to reduce the amount of data for infringement detection, further, the NFT image works to be detected may be those whose activity level is higher than the first preset threshold in the NFT image works trading platform. NFT image works. Among them, the activity can be represented by the number of transactions and the number of browsing of NFT image works. The first preset threshold can be, for example, 40%, that is to say, the NFT image works to be detected are the NFT image works with the top 40% transaction volume in the NFT image works trading platform, or the images to be detected are browsed on the NFT image works trading platform. The top 40% of NFT image works, etc.

Optionally, the NFT image work to be detected is an NFT image work to be uploaded to the NFT image work trading platform. If it is detected that the NFT image work to be detected does not infringe other image works, that is, the similarity between the NFT image work to be detected and any image in the image database is less than or equal to the preset similarity threshold, it is allowed to upload it to the NFT image work trading platform.

Specifically, the images in the image database are images in the hotspot event information, and the hotspot event information is event information whose pageviews are greater than the second preset threshold. The second preset threshold is used as a preset pageview amount. For example, the second preset threshold is 100,000. Specifically, the hot event information may be events whose pageviews on each website are greater than the second preset threshold. Specifically, the data on each website can be obtained by means of a web crawler. It can be known that hot events are usually time-sensitive, and the popularity of events may decrease over time. Therefore, the images in the image database can be updated every so often. For example, but not limited to, it is updated once a week, or once a month, or once every three months, etc. Then the number of pageviews of the hot event is the total number of pageviews in this period of time. Specifically, the hot event may include hot news expressed in the form of text and pictures, hot video delivered in the form of video, hot audio broadcast in the form of audio and pictures, and the like.

Specifically, the embodiments of this specification can use technical means such as structural similarity (Structural Similarity, SSIM), cosine similarity, histogram, and normalized mutual information to measure the relationship between the NFT image work to be detected and each image in the image database. similarity.

Specifically, structural similarity is a full-reference image quality evaluation index, which can measure image similarity from three aspects: brightness, contrast, and structure. In practical applications, a sliding window can be used to divide the image into blocks, so that the total number of blocks is N. Considering the influence of the shape of the window on the blocks, Gaussian weighting is used to calculate the mean, variance and covariance of each window, and then the corresponding block is calculated. The structural similarity SSIM of , and finally the average value is used as the structural similarity measure of the two images, that is, the average structural similarity SSIM. Cosine similarity is to represent the image as a vector, and to represent the similarity between the NFT image work to be detected and each image in the image database by calculating the cosine distance between the vectors. The histogram can describe the global distribution of colors in an image, and then the similarity between them can be determined by comparing the color distribution between the NFT image work to be detected and each image in the image database. The normalized mutual information can be understood as the information amount of the images in the image database contained in the NFT image work to be detected. The larger its value is, the higher the similarity between the images is. In the case of similar numbers, it has good registration accuracy and high reliability.

Step 204, if the similarity between the NFT image work to be detected and any image in the image database is greater than a preset similarity threshold, it is determined that the NFT image work to be detected is infringing.

The preset similarity threshold can be understood as a threshold used to determine whether the NFT image work to be detected has a risk of infringement. For example, the value range of the similarity is between 0 and 1, and the preset similarity threshold may be set to 0.7. In the case where the similarity between the NFT image work to be detected and any image in the image database is greater than 0.7, it can be determined that the NFT image work to be detected infringes the images in the image database.

Further, in order to fix the evidence of infringement of the NFT image work to be detected, after determining the infringement of the NFT image work to be detected in the embodiment of this specification, the NFT image work to be detected and the infringing object (that is, the image database and the NFT image to be detected) can also be combined. The similarity of the work is greater than the preset similarity threshold), and the specific similarity between the two is uploaded to the blockchain, so as to use the non-tamperable feature of the blockchain to complete the evidence fixation and prevent others from modifying the detection results. Specifically, if the NFT image work to be detected is an NFT image work to be uploaded to the NFT image work trading platform, if the similarity between the NFT image work to be detected and any image in the image database is less than or equal to the preset similarity threshold, that is, If the NFT image work to be detected does not infringe the image work in the image database, the embodiment of this specification can upload the NFT image work to be detected to the NFT image work trading platform. For example, if there is no image in the image database and the similarity between the user's NFT image work is greater than the preset similarity threshold of 0.7, that is, each image in the image database is related to the image to be uploaded to the NFT image work trading platform. If the similarity between NFT image works is less than or equal to the preset similarity threshold, it can be determined that there is no infringement of the NFT image work to be uploaded to the NFT image work trading platform, and the NFT image work trading platform can receive the NFT image work, Users can use the NFT corresponding to the NFT image work on the NFT image work trading platform to trade or auction it on the blockchain.

In the embodiment of this specification, the similarity between the NFT image work to be detected and the images in the image database can be obtained by comparing the image features of the NFT image work to be detected and the images in the image database; If the similarity of any image in the image database is greater than the preset similarity threshold, it is determined that the NFT image work to be detected is infringing. Therefore, on the one hand, the embodiments of this specification use the method of image feature comparison to complete the infringement detection of NFT works, so as to avoid the failure of infringement detection due to slight modifications to the NFT works, such as scaling or editing a small number of pixels, and improve the infringement detection. 's accuracy. On the other hand, the creator does not need to provide a lot of evidence to prove the ownership information of the image, which saves a lot of manpower and material resources, avoids tedious work such as manual review and uploading review results, and improves the efficiency of infringement detection.

The image features in the embodiments of this specification may include global features and local features. As shown in FIG. 3 , a schematic flowchart of a method for detecting infringement of an NFT image work is provided. As shown in Figure 3, the method for detecting infringement of NFT image works may include the following steps:

Step 302, extracting global features and local features of the NFT image work to be detected.

It can be understood that, the global features of an image in the embodiments of the present specification refer to features that can represent the entire image, and are used to describe overall features such as color and shape of the image. Local features refer to some features that appear locally in the image, and this local refers to some features such as edges or corners that can appear stably and have better distinguishability.

Specifically, color features, texture features, and shape features in the NFT image work to be detected can be extracted as the global features of the NFT image work to be detected; edges, corners, lines, curves, and The preset attribute area determines the local features of the NFT image work to be detected. The preset attribute area is used to represent a preset specific area of each type of NFT image work, for example, a slowly changing sky in a landscape work.

Referring to the NFT image works to be detected shown in Figures 4a-4c, it can be understood that the global features of the NFT image works to be detected can be extracted through Figure 4a, for example, the color of pedestrians' clothes, the color of the sky, and the shape of traffic lights Wait. Figure 4b is an image for extracting local features such as edges and corners of the NFT image work to be detected. Further, the posture information of pedestrians can be observed through Figure 4b. Figure 4c is the image after the NFT image work to be detected is divided into grids. Specifically, feature extraction can be performed on certain areas in the grid image (for example, the edge area of clouds and the movement joint areas of pedestrians), and finally a plurality of The features are fused together as final local features.

Step 304: Compare the global features and local features of the NFT image work to be detected with the global features and local features of multiple images in the image database to obtain the similarity between the NFT image work to be detected and the images in the image database.

Specifically, this specification can use a deep convolutional neural network (Convolutional Neural Networks, CNN) to model the global context environment and local fine feature detection, that is, use different pre-training strategies to perform effective global operations on different types of NFT image works. Feature pre-training; and further extract refined features for different categories of NFT image works through a recurrent network. In this way, the NFT image work to be detected can be input into the deep convolutional neural network to output the similarity between the NFT image work to be detected and the image in the image database, so as to improve the accuracy of comparison.

Step 306 , if the similarity between the NFT image work to be detected and any image in the image database is greater than the preset similarity threshold, it is determined that the NFT image work to be detected is infringing.

Specifically, step 306 is the same as step 204, and details are not repeated here.

Therefore, the embodiments in this specification can accurately calculate the similarity between images by combining the global context information and local fine detection features of the NFT image work to be detected and each image in the image database; it can also be based on the deep convolutional neural network. The loop structure is repeatedly iterated to filter the influence of noise information, thereby improving the detection accuracy of the NFT image work to be detected.

In the embodiment of this specification, as shown in FIG. 5 , a schematic flowchart of a method for detecting infringement of an NFT image work is provided. As shown in Figure 5, the method for detecting infringement of NFT image works may include the following steps:

Step 502, acquiring hot event information.

Specifically, the hot event information is event information with a pageview greater than a second preset threshold. The second preset threshold is used as a preset pageview amount. For example, the second preset threshold is 100,000. Specifically, the hot event information may be events whose pageviews on each website are greater than the second preset threshold. Specifically, the data on each website can be obtained by means of a web crawler. It can be known that hot events are usually time-sensitive, and the popularity of events may decrease over time. Therefore, the images in the image database can be updated every so often. For example, but not limited to, it is updated once a week, or once a month, or once every three months, etc. Then the number of pageviews of the hot event is the total number of pageviews in this period of time.

Further, in the embodiment of the present specification, the ownership information of the hot event information can also be extracted, and the ownership information of the hot event information can be stored in the ownership database. Among them, the ownership database can be regarded as a specific storage space, which is specially used to store the ownership information of hot events.

The ownership information is used to represent creator information or creation organization information corresponding to the hot event information, for example, the author information of the images in the hot news (such as the author's name, contact information, mailing address, etc.).

Step 504 , perform semantic analysis on the text information in the hot event to obtain the semantic information of the hot event.

It can be understood that, in the embodiment of this specification, technical means such as text classification and sentiment analysis can be performed on the text information in the hot event through semantic analysis, so as to determine the category, label, keyword, etc. of the hot event.

Specifically, this specification can perform text classification on hot events through text preprocessing, text feature extraction, and traditional machine learning methods (such as Bayesian, support vector machine, random forest model, etc.) to obtain the category and label of hot events. , keywords, etc.

Specifically, based on the preset sentiment dictionary method, this specification can first perform preprocessing such as word segmentation and stop word processing on the text information in hot events, and then use the sentiment dictionary to perform string matching on the text information, so as to mine positive and negative Negative information. For example, sentiment analysis can mine the evaluation information of hot events in various dimensions, so as to clarify the category, label, and keyword of the image. Such as environmental protection, environmental pollution, people's livelihood, war, epidemic prevention and control and other dimensions.

It is understandable that the sentiment dictionary can be constructed based on data sources such as Weibo, news, forums, HowNet, etc. Of course, the sentiment dictionary can also be trained by existing corpus. Possibly, the present specification can also perform semantic analysis on the audio information and/or video information in the hot event to obtain the semantic information of the hot event.

Specifically, this specification can convert the video information and/or audio information contained in the hot event into text information, and then perform semantic analysis on the text information to obtain the semantic information of the hot event, that is, the category, label, keyword and other information of the hot event. .

Step 506: Save the semantic information of the hot event in the semantic database, and save the image information of the hot event in the image database.

Specifically, in the embodiment of the present specification, information such as the category, label, and keyword of the extracted hot event can be saved in the semantic database, and the image information in the hot event can be saved in the image database, so as to form a one-to-one correspondence relationship . Among them, the semantic database can be regarded as a specific storage space, which is specially used to store the semantic information of hot events.

Correspondingly, the embodiments of the present specification can also filter the NFT image works to be detected from the NFT image works trading platform based on the similarity of semantic information, so as to narrow the scope of the NFT image works to be detected and reduce subsequent computing resources.

Possibly, the embodiment of this specification can determine the first image set whose activity is higher than the first preset threshold in the NFT image work trading platform; further, determine the similarity between each image in the first image set and the semantic information in the semantic database ; And from the first image set, the second image set whose similarity with the semantic information in the semantic database is greater than the preset semantic similarity threshold, the second image set can include at least one candidate image, and the NFT image work to be detected is the second image any candidate image in the set.

For example, images with an activity level lower than 40% (the first preset threshold) in the NFT image work trading platform can be filtered to obtain a first image set, and based on the category, label, or key corresponding to each NFT image work in the first image set Words and other information determine the similarity with the categories, labels, or keywords of each hot event in the semantic database, so that the similarity with the semantic information in the semantic database is greater than 60% (preset semantic similarity) from the first image set. degree threshold) of the second image set. In this way, detection can be performed based on the NFT image works in the second image set, so as to further narrow the scope of detecting the NFT image works and reduce subsequent computing resources.

It is not limited to the above-mentioned first image set whose activity degree of the NFT image work trading platform is higher than the first preset threshold, and then the first image set whose similarity with the semantic information in the semantic database is greater than the preset semantic similarity threshold is selected from the first image set. Second image set. In the specific implementation, it is also possible to first determine the second image set whose semantic information similarity between the NFT image work trading platform and the semantic database is greater than the preset semantic similarity threshold, and then filter out the second image set with high activity. The first image set with the first preset threshold is not limited in this embodiment of the present specification.

Possibly, the embodiment of the present specification may also upload the similarity between each image in the second image set and the semantic information in the semantic database to the blockchain, so as to complete the fixation of the process evidence.

Step 508: Compare the image features of the NFT image work to be detected and the images in the image database to obtain the similarity between the NFT image work to be detected and the images in the image database.

Specifically, step 508 is the same as step 202, and details are not repeated here.

Step 510: If the similarity between the NFT image work to be detected and any image in the image database is greater than the preset similarity threshold, it is determined that the NFT image work to be detected is infringing.

Specifically, step 510 is the same as step 204, and details are not repeated here.

Further, after it is determined that the image to be detected is infringed, the NFT image work infringement detection method provided by the embodiments of this specification can also notify the copyright owner (infringed party) to protect the rights and interests of the copyright owner. details as follows:

Step 512: Determine, according to the ownership database, ownership information corresponding to the image whose similarity of the NFT image work to be detected is greater than a preset similarity threshold.

Wherein, the ownership database in the embodiment of this specification is used to represent a database that stores the ownership information corresponding to the image information in the hot event.

It can be understood that in the case that the similarity corresponding to the NFT image work to be detected and the hot event in the image database is greater than the preset similarity threshold, the ownership information stored in the ownership database can be used to determine the corresponding hot event in the image database. Ownership information, that is, the author's name, contact information, mailing address and other information.

Step 514 , based on the ownership information, send a notification message to the copyright owner of the image whose similarity with the NFT image work to be detected is greater than a preset similarity threshold.

Specifically, in the embodiment of this specification, when the similarity corresponding to the NFT image work to be detected and the hot event in the image database is greater than the preset similarity threshold, the ownership information of the NFT image work corresponding to the hot event in the image database can be sent. To the copyright owner of the NFT image work to be detected in the NFT image work trading platform to remind the copyright owner that there may be infringement problems in his work.

Possibly, the embodiments of this specification may also send the infringement details of the NFT image work to be detected, that is, similar global image features and/or local image features, to the copyright owner of the NFT image work to be detected.

Further, the NFT image work trading platform in the embodiment of this specification can also prompt the copyright owner of the NFT image work to be detected that has been stored in the blockchain to confirm whether the NFT image work is infringing, and prompt its corresponding legal Risk, it is recommended to remove from the shelf. For the NFT image works to be detected that have not been uploaded to the blockchain, the NFT image work trading platform in the embodiment of this specification may prompt the copyright owner of the NFT image work to be detected that the NFT image work has a risk of infringement, and reject the NFT to be detected. The image works are uploaded to the NFT image works trading platform.

Possibly, in the embodiment of this specification, the ownership information corresponding to the image whose similarity of the NFT image work to be detected is greater than the preset similarity threshold can be uploaded to the blockchain, so as to complete the fixation of the process evidence and prevent the comparison of the detection process. The tampering of the information makes the detection results of the NFT image work trading platform wrong, which increases the risk of infringement disputes on the NFT image work trading platform.

Referring to Figure 6, in a specific example, the text information obtained from the portal website under the title of "Little Red Book Filter Scenic Spots and Events Party Responses" obtains semantic information: Xiaohongshu, Scenic Spots, Sanya Blue House After the words such as , pink beach, etc., the semantic information of each NFT image work on the NFT image work trading platform in the semantic database is retrieved, and multiple image sets (that is, the NFT image works to be detected) that are greater than 60% of the semantic similarity threshold are obtained. Further, the image under the hot news title can be extracted, and based on the global features and local features of each NFT image work to be detected in the image database, it is determined whether there is a similarity with the image under the hot news title greater than the similarity threshold of 0.7. The NFT image works to be detected, if there are multiple NFT image works to be detected and the image under the hot news title, the similarity is greater than the image similarity threshold of 0.7, then further, the corresponding author related images under the hot news title can be extracted. Information (ownership information), and send infringement notices to the copyright owners of multiple NFT image works to be detected that are at risk of infringement, so as to avoid possible infringement disputes.

Therefore, the embodiments of this specification can obtain hot events and NFT image works with high activity in the NFT image work trading platform, and based on the images in hot events and NFT image works with high activity in the NFT image work trading platform. The semantic similarity of the first image set is obtained by filtering the first image set to narrow the detection range of the NFT image works to be detected to achieve the purpose of saving computing resources. It can be determined whether there is an NFT image work to be detected with the risk of infringement in the first image set, so as to avoid problems such as inaccurate detection results caused by image editing or transformation when directly comparing NFTs. Further, the relevant ownership information and infringement details (similar image features) can also be sent to the copyright owner of the image work that may have the risk of infringement in the NFT image work trading platform through the ownership database, and all comparisons can be made. The results are uploaded to the blockchain, which not only reduces the risk of possible infringement disputes on the NFT image work trading platform, but also effectively fixes the process evidence.

The foregoing describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that in the embodiments and still achieve desirable results. Additionally, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

FIG. 7 is a schematic structural diagram of a device for detecting infringement of an NFT image work provided by an embodiment of this specification. The device is used in a non-fungible token NFT image work trading platform, and executes the NFT image work infringement detection method according to any of the above-mentioned embodiments of this specification. As shown in Figure 7, the NFT image work infringement detection device may include:

The comparison module 71 is used to compare the image features of the NFT image work to be detected and the images in the image database to obtain the similarity between the NFT image work to be detected and the images in the image database;

The determining module 72 is configured to determine that the NFT image work to be detected is infringing if the similarity between the NFT image work to be detected and any one of the images in the image database is greater than a preset similarity threshold.

In the embodiment of this specification, the similarity between the NFT image work to be detected and the images in the image database can be obtained by comparing the image features of the NFT image work to be detected and the images in the image database; If the similarity of any image in the image database is greater than the preset similarity threshold, it is determined that the NFT image work to be detected is infringing. Therefore, on the one hand, the embodiments of this specification use the method of image feature comparison to complete the infringement detection of NFT works, so as to avoid the failure to detect infringement due to slight modifications to the NFT works, such as scaling or editing a small number of pixels, and improve the infringement detection. 's accuracy. On the other hand, the creator does not need to provide a lot of evidence to prove the ownership information of the image, which saves a lot of manpower and material resources, avoids tedious work such as manual review and uploading review results, and improves the efficiency of infringement detection.

In some embodiments, the comparison module 71 includes:

an extraction unit for extracting the global features and local features of the NFT image work to be detected;

The comparison unit is used to compare the global features and local features of the NFT image work to be detected with the global features and local features of a plurality of images in the image database to obtain the NFT image work to be detected and the Similarity of images in an image database.

In some embodiments, the extraction unit is specifically used for:

Determine the global feature of the NFT image work to be detected based on the color feature, texture feature, and shape feature of the NFT image work to be detected;

Based on the edges, corners, lines, curves, and preset attribute areas of the NFT image work to be detected, local features of the NFT image work to be detected are determined.

In some embodiments, the NFT image work to be detected is an image whose activity level is higher than a first preset threshold in the NFT image work trading platform.

In some embodiments, the image to be detected is an image to be uploaded to the NFT trading platform;

The device also includes:

The uploading module of the NFT image work to be detected is used to upload the NFT image work to be detected if the similarity between the NFT image work to be detected and any one of the images in the image database is less than or equal to the preset similarity threshold. The NFT image works are uploaded to the NFT image works trading platform.

In some embodiments, the images in the image database are images in hot event information, and the hot event information is event information whose pageviews are greater than a second preset threshold.

In some embodiments, the apparatus further includes:

an acquisition module for acquiring the hot event information;

a semantic information comparison module, configured to perform semantic analysis on the text information in the hot event to obtain the semantic information of the hot event;

a semantic information saving module, used for saving the semantic information of the hot event in the semantic database, and saving the image information of the hot event in the image database;

The device also includes:

a first image set determination module, configured to determine a first image set whose activity level is higher than a first preset threshold in the NFT image work trading platform;

A screening module, configured to screen a second image set whose similarity with the semantic information in the semantic database is greater than a preset semantic similarity threshold from the first image set, where the second image set includes at least one candidate image, and the The NFT image work to be detected is any candidate image in the second image set.

In some embodiments, after the acquiring module, the apparatus further includes:

a title information extraction module, configured to extract the title information of the hot event information, and save the title information of the hot event information in the title database;

The device also includes:

A title information determination module, configured to determine, according to the title database, the title information corresponding to the image whose similarity of the NFT image work to be detected is greater than a preset similarity threshold;

A message sending module, configured to send a notification message to the copyright owner of an image whose similarity with the NFT image work to be detected is greater than a preset similarity threshold based on the ownership information.

In some embodiments, the apparatus further includes: an image similarity uploading module, configured to upload the similarity between the NFT image work to be detected and the NFT image work to be detected in the image database that is greater than a preset similarity threshold The similarity of the images is uploaded to the blockchain.

In some embodiments, the apparatus further includes:

a semantic information similarity determination module, configured to determine the similarity between each image in the first image set and the semantic information in the semantic database;

The device also includes:

The semantic information similarity uploading module is configured to upload the similarity between each image in the second image set and the semantic information in the semantic database to the blockchain.

In some embodiments, the apparatus further includes: an ownership information uploading module, configured to upload the ownership information corresponding to the image whose similarity of the NFT image work to be detected is greater than a preset similarity threshold to the blockchain .

In some embodiments, the apparatus further includes: a hot-spot event semantic information comparison module, configured to perform semantic analysis on the audio information and/or video information in the hot-spot event to obtain semantic information of the hot-spot event.

It should be noted that, when the NFT image work infringement detection device provided in the above embodiment executes the NFT image work infringement detection method, only the division of the above functional modules is used as an example for illustration. In practical applications, the above functions can be allocated as required. It is completed by different functional modules, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the NFT image work infringement detection device provided in the above embodiment and the NFT image work infringement detection method embodiment belong to the same concept, and the embodiment and implementation process thereof are detailed in the method embodiment, which will not be repeated here.

The above-mentioned serial numbers of one or more embodiments in this specification are only for description, and do not represent the advantages or disadvantages of the embodiments.

Referring to FIG. 8 , a schematic structural diagram of an electronic device is provided in one or more embodiments of the present specification. As shown in FIG. 8 , the electronic device 80 may include: at least one processor 81 , at least one network interface 84 , user interface 83 , memory 85 , and at least one communication bus 82 .

Among them, the communication bus 82 is used to realize the connection and communication between these components.

The user interface 83 may include a display screen (Display) and a camera (Camera), and the optional user interface 83 may also include a standard wired interface and a wireless interface.

Wherein, the network interface 84 may optionally include a standard wired interface and a wireless interface (eg, a WI-FI interface).

The processor 81 may include one or more processing cores. The processor 81 uses various excuses and lines to connect various parts of the entire electronic device 80, and executes by running or executing the instructions, programs, code sets or instruction sets stored in the memory 85, and calling the data stored in the memory 85. Various functions of the electronic device 80 and processing data. Optionally, the processor 81 may use at least one of digital signal processing (Digital Signal Processing, DSP), field-programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). implemented in hardware. The processor 801 may integrate one or a combination of a processor (Central Processing Unit, CPU), a graphics processor (Graphics Processing Unit, GPU), a modem, and the like. Among them, the CPU mainly handles the operating system, user interface, and application programs; the GPU is used to render and draw the content that needs to be displayed on the display screen; the modem is used to handle wireless communication. It can be understood that, the above-mentioned modem may not be integrated into the processor 81, and is implemented by a single chip.

The memory 85 may include random access memory (Random Access Memory, RAM), or may include read-only memory (Read-Only Memory). Optionally, the memory 85 includes a non-transitory computer-readable storage medium. Memory 85 may be used to store instructions, programs, codes, sets of codes, or sets of instructions. The memory 805 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), Instructions and the like used to implement the above method embodiments; the storage data area may store the data and the like involved in the above method embodiments. The memory 85 can optionally also be at least one storage device located remote from the aforementioned processor 81 . As shown in FIG. 8 , the memory 85 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an NFT image work infringement detection application program.

In the electronic device 80 shown in FIG. 8 , the user interface 83 is mainly used to provide an input interface for the user to obtain data input by the user; and the processor 81 can be used to call the NFT image work infringement detection application stored in the memory 85 , and specifically do the following:

Comparing the image features of the NFT image work to be detected and multiple images in the image database to obtain the similarity between the NFT image work to be detected and the images in the image database;

If the similarity between the NFT image work to be detected and any one of the images in the image database is greater than a preset similarity threshold, it is determined that the NFT image work to be detected is infringing.

In a possible embodiment, the processor 81 is performing the step of comparing the image features of the NFT image work to be detected and the images in the image database to obtain the NFT image work to be detected and the image database. When the similarity of the images in the specific implementation:

Extracting the global features and local features of the NFT image work to be detected;

Compare the global features and local features of the NFT image work to be detected with the global features and local features of multiple images in the image database to obtain the NFT image work to be detected and the image in the image database. similarity.

In a possible embodiment, when the processor 81 executes the extraction of global features and local features of the NFT image work to be detected, it specifically executes:

Determine the global feature of the NFT image work to be detected based on the color feature, texture feature, and shape feature of the NFT image work to be detected;

Based on the edges, corners, lines, curves, and preset attribute areas of the NFT image work to be detected, local features of the NFT image work to be detected are determined.

In a possible embodiment, the NFT image work to be detected is an image whose activity level is higher than a first preset threshold in the NFT image work trading platform.

In a possible embodiment, the image to be detected is an image to be uploaded to the NFT trading platform;

After the processor 81 executes the obtaining of the similarity between the NFT image work to be detected and the image in the image database, it also executes:

If the similarity between the NFT image work to be detected and any one of the images in the image database is less than or equal to the preset similarity threshold, upload the NFT image work to be detected to the NFT image work trading platform.

In a possible embodiment, the images in the image database are images in hotspot event information, and the hotspot event information is event information whose pageview volume is greater than a second preset threshold.

In a possible embodiment, before the processor 81 executes the comparing the image feature of the NFT image work to be detected with the multiple images in the image database, the processor 81 further executes:

obtain the hot event information;

Perform semantic analysis on the text information in the hot event to obtain the semantic information of the hot event;

saving the semantic information of the hot event in the semantic database, and saving the image information of the hot event in the image database;

Before the processor 81 performs the comparison of the NFT image work to be detected with the image features of the multiple images in the image database, it also executes:

Determine the first image set whose activity is higher than the first preset threshold in the NFT image work trading platform;

A second image set whose similarity with the semantic information in the semantic database is greater than a preset semantic similarity threshold is selected from the first image set, where the second image set includes at least one candidate image, the NFT image to be detected The work is any candidate image in the second image set.

In a possible embodiment, after the processor 81 executes the obtaining of the hotspot event information, the processor 81 further executes:

extracting the ownership information of the hot event information, and saving the ownership information of the hot event information in the ownership database;

After determining that the to-be-detected NFT image work is infringed, the method further includes:

Determine the ownership information corresponding to the image whose similarity of the NFT image work to be detected is greater than the preset similarity threshold according to the ownership database;

Based on the ownership information, a notification message is sent to the copyright owner of the image whose similarity with the NFT image work to be detected is greater than a preset similarity threshold.

In a possible embodiment, after performing the determining of the infringement of the NFT image work to be detected, the processor 81 further performs: combining the NFT image work to be detected and the image database with the to-be-detected image work Detect the similarity of NFT image works whose similarity is greater than the preset similarity threshold and upload it to the blockchain.

In a possible embodiment, after the processor 81 executes the determining of the first image set whose activity level is higher than the first preset threshold in the NFT image work trading platform, the processor 81 executes Before the screening from the first image set to the second image set whose semantic information similarity in the semantic database is greater than the preset semantic similarity threshold, further execute:

determining the similarity between each image in the first image set and the semantic information in the semantic database;

After the screening of the second image set with the semantic information similarity in the semantic database greater than the preset semantic similarity threshold from the first image set, the method further includes:

Upload the similarity between each image in the second image set and the semantic information in the semantic database to the blockchain.

In a possible embodiment, after the processor 81 executes the determining, according to the ownership database, the ownership information corresponding to the image whose similarity of the NFT image work to be detected is greater than a preset similarity threshold , and also perform: uploading the ownership information corresponding to the image whose similarity of the NFT image work to be detected is greater than the preset similarity threshold to the blockchain.

In a possible embodiment, after performing the acquiring of the hot event information, the processor 81 further performs: performing semantic analysis on the audio information and/or video information in the hot event to obtain the Semantic information of hot events.

One or more embodiments of the present specification also provide a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, when executed on a computer or a processor, the computer or the processor causes the computer or processor to execute the above-mentioned FIG. 2 - Figure 3, and one or more steps in the embodiment shown in Figure 5. If each component module of the above-mentioned NFT image work infringement detection device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in the computer-readable storage medium.

In the above-mentioned embodiments, it may be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented in software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. The computer program instructions, when loaded and executed on a computer, produce, in whole or in part, the procedures or functions described in accordance with one or more embodiments of this specification. The computer may be a general purpose computer, special purpose computer, computer network, or other programmable device. The computer instructions may be stored in or transmitted over a computer-readable storage medium. The computer instructions can be sent from a website site, computer, server, or data center via wired (eg, coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (eg, infrared, wireless, microwave, etc.) another website site, computer, server or data center for transmission. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, a data center, or the like that includes an integration of one or more available media. The available media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, Digital Versatile Disc (DVD)), or semiconductor media (eg, Solid State Disk (SSD) ))Wait.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, The processes of the embodiments of the various methods described above may be included. The aforementioned storage medium includes: a system memory (Read Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk and other media that can store program codes. The technical features in this embodiment and the implementation can be combined arbitrarily if there is no conflict.

The above-mentioned embodiments are only the preferred embodiments of this specification to describe, and do not limit the scope of this specification. Without departing from the design spirit of this specification, those of ordinary skill in the art can make technical solutions of this specification. Various modifications and improvements shall fall within the protection scope determined by the claims of this specification.

Claims (15)

1. A method for detecting infringement of a non-fungible token NFT image work, the method comprising: Comparing the image features of the NFT image work to be detected and multiple images in the image database to obtain the similarity between the NFT image work to be detected and the images in the image database; If the similarity between the NFT image work to be detected and any one of the images in the image database is greater than a preset similarity threshold, it is determined that the NFT image work to be detected is infringing. 2. The method according to claim 1, wherein the NFT image work to be detected is compared with the image features of a plurality of images in the image database, and the similarity between the NFT image work to be detected and the image in the image database is obtained. degrees, including: Extracting the global features and local features of the NFT image work to be detected; Compare the global features and local features of the NFT image work to be detected with the global features and local features of multiple images in the image database to obtain the NFT image work to be detected and the image in the image database. similarity. 3. method according to claim 2, described extracting the global feature and local feature of described NFT image work to be detected, comprising: Determine the global feature of the NFT image work to be detected based on the color feature, texture feature, and shape feature of the NFT image work to be detected; Based on the edges, corners, lines, curves, and preset attribute areas of the NFT image work to be detected, local features of the NFT image work to be detected are determined. 4 . The method according to claim 1 , wherein the NFT image work to be detected is an image whose activity level is higher than a first preset threshold in the NFT image work trading platform. 5 . 5. The method according to any one of claims 1-4, wherein the images in the image database are images in hot event information, and the hot event information is event information whose pageviews are greater than a second preset threshold. 6. The method according to claim 5, before the NFT image work to be detected is compared with the image features of a plurality of images in the image database, the method further comprises: obtain the hot event information; Perform semantic analysis on the text information in the hot event to obtain the semantic information of the hot event; saving the semantic information of the hot event in the semantic database, and saving the image information of the hot event in the image database; Before the NFT image work to be detected is compared with the image features of multiple images in the image database, the method further includes: Determine the first image set whose activity is higher than the first preset threshold in the NFT image work trading platform; A second image set whose similarity with the semantic information in the semantic database is greater than a preset semantic similarity threshold is selected from the first image set, where the second image set includes at least one candidate image, the NFT image to be detected The work is any candidate image in the second image set. 7. The method according to claim 6, after the acquisition of the hotspot event information, the method further comprises: extracting the ownership information of the hot event information, and saving the ownership information of the hot event information in the ownership database; After determining that the to-be-detected NFT image work is infringed, the method further includes: Determine the ownership information corresponding to the image whose similarity of the NFT image work to be detected is greater than the preset similarity threshold according to the ownership database; Based on the ownership information, a notification message is sent to the copyright owner of the image whose similarity with the NFT image work to be detected is greater than a preset similarity threshold. 8. The method according to claim 1, the NFT image work to be detected is an image to be uploaded to the NFT image work trading platform; After obtaining the similarity between the NFT image work to be detected and the image in the image database, the method further includes: If the similarity between the NFT image work to be detected and any one of the images in the image database is less than or equal to the preset similarity threshold, upload the NFT image work to be detected to the NFT image work trading platform . 9. The method according to claim 1, after said determining that the NFT image work to be detected is infringed, the method further comprises: combining the NFT image work to be detected with the NFT to be detected in the image database with the NFT to be detected. The similarity of the image works whose similarity is greater than the preset similarity threshold is uploaded to the blockchain. 10. The method according to claim 6, after said determining a first image set whose activity level is higher than a first preset threshold in said NFT image work trading platform, said filtering from said first image set with all relevant images. Before the second image set in which the semantic information similarity in the semantic database is greater than the preset semantic similarity threshold, the method further includes: determining the similarity between each image in the first image set and the semantic information in the semantic database; After the screening of the second image set with the semantic information similarity in the semantic database greater than the preset semantic similarity threshold from the first image set, the method further includes: Upload the similarity between each image in the second image set and the semantic information in the semantic database to the blockchain. 11. The method according to claim 7, after determining the ownership information corresponding to the image whose similarity of the NFT image work to be detected is greater than a preset similarity threshold according to the ownership database, the method further Including: uploading the ownership information corresponding to the image whose similarity of the NFT image work to be detected is greater than the preset similarity threshold to the blockchain. 12. The method according to claim 6, after the acquisition of the hotspot event information, the method further comprises: performing semantic analysis on the audio information and/or video information in the hotspot event to obtain the hotspot event semantic information. 13. A non-fungible token NFT image work infringement detection device, the device comprising: A comparison module, configured to compare the image features of the NFT image work to be detected and the images in the image database to obtain the similarity between the NFT image work to be detected and the images in the image database; A determining module, configured to determine that the NFT image work to be detected is infringing if the similarity between the NFT image work to be detected and any one of the images in the image database is greater than a preset similarity threshold. 14. A computer storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform the method steps of any of claims 1-12. 15. An electronic device comprising: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and perform the method steps of any of claims 1-12.

Images