CN114913534A - Block chain-based network security abnormal image big data detection method and system - Google Patents
Block chain-based network security abnormal image big data detection method and system Download PDFInfo
- Publication number
- CN114913534A CN114913534A CN202210844485.7A CN202210844485A CN114913534A CN 114913534 A CN114913534 A CN 114913534A CN 202210844485 A CN202210844485 A CN 202210844485A CN 114913534 A CN114913534 A CN 114913534A
- Authority
- CN
- China
- Prior art keywords
- image
- target image
- similarity
- target
- abnormal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/19007—Matching; Proximity measures
- G06V30/19013—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/19007—Matching; Proximity measures
- G06V30/19093—Proximity measures, i.e. similarity or distance measures
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Character Discrimination (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a block chain-based network security abnormal image big data detection method and system, and relates to the technical field of image recognition. The method comprises the following steps: acquiring a target image, acquiring corresponding viewer record information, and performing chain storage; carrying out character recognition on the target image, carrying out vocabulary matching, and if the matching is successful, determining the target image as a suspected abnormal image; otherwise, selecting a plurality of abnormal images as template images, and calculating the similarity between the target image and each template image; if the similarity is greater than a preset similarity threshold, determining the image is a suspected abnormal image; otherwise, carrying out image description processing on the target image, carrying out vocabulary matching, and if the matching is successful, determining the target image as a suspected abnormal image; otherwise, the corresponding target image is determined to be a non-abnormal image. The invention can carry out accurate image comparison and character recognition, ensures the real reliability of all data by combining the block chain technology and greatly improves the detection precision of abnormal images.
Description
Technical Field
The invention relates to the technical field of image identification, in particular to a block chain-based network security abnormal image big data detection method and system.
Background
With the rapid development of internet technology, network security draws more and more attention, and some abnormal images often appear in the network, which brings great negative effects to the network security.
Although the traditional image recognition and target detection method can recognize partial abnormal images, the abnormal images which harm the network security cannot be detected comprehensively and accurately. The block chain is an emerging technology in recent years and can play an important role in the aspect of network security; therefore, how to combine the blockchain technology to realize the comprehensive and accurate detection of the big data anomaly of the network security image becomes a new problem.
Disclosure of Invention
In order to overcome the above problems or at least partially solve the above problems, embodiments of the present invention provide a method and a system for detecting large data in a network security abnormal image based on a block chain, which can perform accurate image comparison and character recognition, ensure the real reliability of all data by combining with the block chain technology, and greatly improve the detection accuracy of the abnormal image.
The embodiment of the invention is realized by the following steps:
in a first aspect, an embodiment of the present invention provides a block chain-based method for detecting a network security abnormal image big data, including the following steps:
s1, acquiring a target image newly issued in the network and corresponding image issuer information, acquiring the record information of a viewer corresponding to the target image, and uploading the target image, the image issuer information and the record information of the corresponding viewer to a block chain for storage;
s2, recognizing characters in the target image by using a multi-OCR technology mutual-verification method to obtain a target character recognition result;
s3, matching the target character recognition result with the vocabulary in the preset abnormal vocabulary database, if the matching is successful, determining the corresponding target image as a suspected abnormal image, and sending the corresponding target image to the corresponding network manager for checking; otherwise, go to step S4;
s4, selecting a plurality of abnormal images as template images, and calculating the similarity of the target image and each template image by using a similarity calculation method based on a difference hole convolution kernel; if the similarity between the target image and any template image is greater than a preset similarity threshold, the corresponding target image is regarded as a suspected abnormal image, and the corresponding target image is sent to a corresponding network administrator for checking; otherwise, go to step S5;
s5, processing the target image by using an image description method based on super-resolution reconstruction to obtain a target description sentence corresponding to the target image;
s6, matching the target description sentence with the vocabulary in the preset abnormal vocabulary database, if the matching is successful, determining the corresponding target image as a suspected abnormal image, and sending the corresponding target image to the corresponding network administrator for checking; otherwise, the corresponding target image is determined to be a non-abnormal image.
In order to solve the technical problem that abnormal images harmful to network safety cannot be detected comprehensively and accurately in the prior art, the method utilizes a multi-OCR (optical character recognition) technology mutual verification method to recognize characters in the images, obviously improves the character recognition accuracy, and further effectively ensures the accuracy of abnormal image recognition. Meanwhile, the similarity calculation method based on the difference hole convolution kernel is utilized, the calculation precision of the similarity between the newly issued target image and the template image is remarkably improved, and the suspected abnormal image is accurately identified. The invention also utilizes an image description method based on super-resolution reconstruction to process the image, thereby remarkably improving the generation quality of the image description sentence. The invention also combines the block chain technology to carry out the uplink storage of the recorded information of the viewer corresponding to the newly released image, thereby providing real and effective data for the subsequent responsibility confirmation and improving the reliability of abnormal image detection. The invention can carry out accurate image comparison and character recognition, ensures the real reliability of all data by combining the block chain technology and greatly improves the detection precision of abnormal images.
Based on the first aspect, in some embodiments of the present invention, the method for recognizing the characters in the target image by using the multi-OCR technology to obtain the target character recognition result includes the following steps:
respectively identifying characters in the target image by adopting a plurality of OCR (optical character recognition) models to obtain a plurality of character identification results;
and counting and determining a target character recognition result according to each character recognition result.
Based on the first aspect, in some embodiments of the present invention, the method for calculating the similarity between the target image and each template image by using the similarity calculation method based on the difference hole convolution kernel includes the following steps:
respectively filtering the target image and the template image by using a convolution kernel with a void ratio of N, respectively carrying out sparse coding on the filtered target image and template image, and calculating the similarity between the target image and the template image by using the Euclidean distance to obtain a first similarity result; n is a natural number greater than or equal to 1;
respectively filtering the target image and the template image by using a convolution kernel with a void ratio of N +1, respectively carrying out sparse coding on the filtered target image and template image, and calculating the similarity between the target image and the template image by using the Euclidean distance to obtain a second similarity result;
and respectively filtering the target image and the template image by using a convolution kernel with the void ratio of N +2, respectively carrying out sparse coding on the filtered target image and template image, and calculating the similarity between the target image and the template image by using the Euclidean distance to obtain a third similarity result.
Based on the first aspect, in some embodiments of the present invention, the block chain-based network security anomaly image big data detection method further includes the following steps:
and if at least one of the first similarity result, the second similarity result and the third similarity result is greater than a preset similarity threshold, determining the corresponding target image as a suspected abnormal image.
Based on the first aspect, in some embodiments of the present invention, the method for processing the target image by using the image description method based on super-resolution reconstruction includes the following steps:
performing super-resolution reconstruction on the target image;
and performing image description processing on the reconstructed target image by adopting an image description method.
Based on the first aspect, in some embodiments of the present invention, the image description method includes CNN + LSTM-based image description or transform + LSTM-based image description.
Based on the first aspect, in some embodiments of the present invention, the block chain-based network security anomaly image big data detection method further includes the following steps:
and acquiring and extracting corresponding image publisher information in the block chain according to the checking result of the network administrator.
In a second aspect, an embodiment of the present invention provides a block chain-based network security abnormal image big data detection system, including: image chain module, character recognition module, first vocabulary matching module, similarity judge module, image description module and second vocabulary matching module, wherein:
the image uplink module is used for acquiring a target image newly released in the network and corresponding image publisher information, acquiring the record information of a viewer corresponding to the target image, and uploading the target image, the image publisher information and the record information of the corresponding viewer to the block chain for storage;
the character recognition module is used for recognizing characters in the target image by utilizing a multi-OCR technology mutual-checking method so as to obtain a target character recognition result;
the first vocabulary matching module is used for matching the target character recognition result with vocabularies in a preset abnormal vocabulary database, if the matching is successful, the corresponding target image is regarded as a suspected abnormal image, and the corresponding target image is sent to a corresponding network manager for checking; otherwise, the similarity judging module works;
the similarity judging module is used for selecting a plurality of abnormal images as template images and calculating the similarity between the target image and each template image by using a similarity calculation method based on the difference cavity convolution kernel; if the similarity between the target image and any template image is greater than a preset similarity threshold, the corresponding target image is regarded as a suspected abnormal image, and the corresponding target image is sent to a corresponding network administrator for checking; otherwise, the image description module works;
the image description module is used for processing the target image by using an image description method based on super-resolution reconstruction so as to obtain a target description statement corresponding to the target image;
the second vocabulary matching module is used for matching the target description sentence with the vocabulary in the preset abnormal vocabulary database, if the matching is successful, the corresponding target image is regarded as a suspected abnormal image, and the corresponding target image is sent to the corresponding network manager for checking; otherwise, the corresponding target image is determined to be a non-abnormal image.
In order to solve the technical problem that abnormal images harmful to network safety cannot be detected comprehensively and accurately in the prior art, the system identifies the characters in the images by utilizing a method of mutual testing of multiple OCR technologies through the cooperation of a plurality of modules such as an image chain module, a character identification module, a first vocabulary matching module, a similarity judging module, an image description module and a second vocabulary matching module, so that the accuracy of character identification is remarkably improved, and the accuracy of abnormal image identification is effectively ensured. Meanwhile, the similarity calculation method based on the difference hole convolution kernel is utilized, the calculation precision of the similarity between the newly issued target image and the template image is remarkably improved, and the suspected abnormal image is accurately identified. The invention also utilizes an image description method based on super-resolution reconstruction to process the image, thereby obviously improving the generation quality of the image description sentence. The invention also combines the block chain technology to carry out uplink storage on the recorded information of the viewer corresponding to the newly released image, thereby providing real and effective data for subsequent responsibility confirmation and improving the reliability of abnormal image detection. The invention can carry out accurate image comparison and character recognition, ensures the real reliability of all data by combining the block chain technology and greatly improves the detection precision of abnormal images.
In a third aspect, an embodiment of the present application provides an electronic device, which includes a memory for storing one or more programs; a processor. The program or programs, when executed by a processor, implement the method of any of the first aspects as described above.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the method according to any one of the first aspect described above.
The embodiment of the invention at least has the following advantages or beneficial effects:
the embodiment of the invention provides a block chain-based network security abnormal image big data detection method and system, which solve the technical problem that abnormal images harmful to network security cannot be detected comprehensively and accurately in the prior art. Meanwhile, the invention also utilizes a similarity calculation method based on the convolution kernel of the difference cavity to obviously improve the calculation precision of the similarity between the newly issued target image and the template image, thereby accurately identifying the suspected abnormal image. The invention also utilizes an image description method based on super-resolution reconstruction to process the image, thereby obviously improving the generation quality of the image description sentence. The invention also combines the block chain technology to carry out the uplink storage of the recorded information of the viewer corresponding to the newly released image, thereby providing real and effective data for the subsequent responsibility confirmation and improving the reliability of abnormal image detection.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
Fig. 1 is a flowchart of a block chain-based network security abnormal image big data detection method according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating the identification of characters in a target image by a multi-OCR mutual verification method in a block chain-based network security abnormal image big data detection method according to an embodiment of the present invention;
fig. 3 is a flowchart of processing a target image by using an image description method based on super-resolution reconstruction in a block chain-based network security abnormal image big data detection method according to an embodiment of the present invention;
FIG. 4 is a schematic block diagram of a block chain-based network security anomaly image big data detection system according to an embodiment of the present invention;
fig. 5 is a block diagram of an electronic device according to an embodiment of the present invention.
Description of reference numerals: 100. an image uplink module; 200. a character recognition module; 300. a first vocabulary matching module; 400. a similarity judging module; 500. an image description module; 600. a second vocabulary matching module; 101. a memory; 102. a processor; 103. a communication interface.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the same element.
Example (b):
as shown in fig. 1 to fig. 3, in a first aspect, an embodiment of the present invention provides a method for detecting a network security abnormal image big data based on a block chain, including the following steps:
s1, acquiring a target image newly issued in the network and corresponding image issuer information, acquiring the record information of a viewer corresponding to the target image, and uploading the target image, the image issuer information and the record information of the corresponding viewer to a block chain for storage; for a newly released image in the network, all image viewers at the time of release record that the image is released by Zhang III for the first time (for example, 300 people watch the image at the same time, and 300 people record that the image is released by Zhang III for the first time), and record information of the viewers is also uploaded to the block chain for storage at the same time, so that the authenticity of the image owner is better ensured.
S2, recognizing characters in the target image by using a multi-OCR technology mutual-verification method to obtain a target character recognition result; the OCR recognition models include EAST models, CRNN + CTC models, CNN + LSTM + CTC architectures, etc., and the OCR recognition models all use the existing conventional recognition models, which are not described herein again.
Further, as shown in fig. 2, the method includes:
s21, respectively recognizing characters in the target image by adopting a plurality of OCR recognition models to obtain a plurality of character recognition results;
and S22, counting and determining a target character recognition result according to each character recognition result.
S3, matching the target character recognition result with the vocabulary in the preset abnormal vocabulary database, if the matching is successful, determining the corresponding target image as a suspected abnormal image, and sending the corresponding target image to the corresponding network administrator for checking; otherwise, go to step S4;
in some embodiments of the present invention, a method of multi-OCR mutual inspection is used to identify characters in a newly issued image, if an abnormal vocabulary (such as a virus, a violent vocabulary, etc.) is identified, i.e. the abnormal vocabulary is successfully matched with a vocabulary in a preset abnormal vocabulary database, the image is determined to be a suspected abnormal image, and the image is sent to a network administrator for inspection; otherwise, the next step is continued. If the recognition results of a plurality of OCR technologies are inconsistent, a few principles subject to majority are utilized, for example, most methods recognize a certain character as a character A, less methods recognize a certain character as a character B, and finally the character is recognized as a character A. And if the characters are not detected in the image, continuing to perform the next step.
S4, selecting a plurality of abnormal images as template images, and calculating the similarity of the target image and each template image by using a similarity calculation method based on a difference hole convolution kernel; if the similarity between the target image and any template image is greater than a preset similarity threshold, the corresponding target image is regarded as a suspected abnormal image, and the corresponding target image is sent to a corresponding network administrator for checking; otherwise, go to step S5;
further, comprising: respectively filtering the target image and the template image by using a convolution kernel with a void ratio of N, respectively carrying out sparse coding on the filtered target image and template image, and calculating the similarity between the target image and the template image by using the Euclidean distance to obtain a first similarity result; n is a natural number more than or equal to 1; respectively filtering the target image and the template image by using a convolution kernel with a void ratio of N +1, respectively carrying out sparse coding on the filtered target image and template image, and calculating the similarity between the target image and the template image by using the Euclidean distance to obtain a second similarity result; and respectively filtering the target image and the template image by using a convolution kernel with the void ratio of N +2, respectively carrying out sparse coding on the filtered target image and template image, and calculating the similarity between the target image and the template image by using the Euclidean distance to obtain a third similarity result.
Further, comprising: and if at least one of the first similarity result, the second similarity result and the third similarity result is greater than a preset similarity threshold, determining the corresponding target image as a suspected abnormal image.
In some embodiments of the present invention, a certain number of abnormal images are manually selected as template images, and the similarity between a newly-published image and the template images is calculated by using a similarity calculation method based on a difference hole convolution kernel. If the similarity between the newly released image and any template image is higher, the newly released image is determined to be a suspected abnormal image, and the suspected abnormal image is sent to a network administrator for checking; otherwise, the next step is continued.
The similarity calculation method based on the difference hole convolution kernel comprises the following steps:
(a) and (4) utilizing the convolution kernel with the void rate of 1 to check the newly released image and a certain specific template image for processing to obtain the filtered newly released image and the specific template image. Respectively carrying out sparse coding on the filtered newly released image and the specific template image, and calculating the similarity between the newly released image and the specific template image by using Euclidean distance;
(b) and (4) utilizing the convolution kernel with the void rate of 2 to check the newly released image and a certain specific template image for processing to obtain the filtered newly released image and the specific template image. Respectively carrying out sparse coding on the filtered newly released image and the specific template image, and calculating the similarity between the newly released image and the specific template image by using the Euclidean distance;
(b) and (4) processing the newly released image and a certain specific template image by using a convolution kernel with a void ratio of 3 to obtain the filtered newly released image and the specific template image. Respectively carrying out sparse coding on the filtered newly released image and the specific template image, and calculating the similarity between the newly released image and the specific template image by using the Euclidean distance;
if any one of the steps (a), (b) and (c) obtains a result with high similarity, the newly issued image and a specific template image are considered to have high similarity.
S5, processing the target image by using an image description method based on super-resolution reconstruction to obtain a target description sentence corresponding to the target image;
further, as shown in fig. 3, the method includes:
s51, performing super-resolution reconstruction on the target image;
and S52, performing image description processing on the reconstructed target image by adopting an image description method. The image description method comprises image description based on CNN + LSTM or image description based on Transformer + LSTM. The image description methods are all image description methods disclosed in the prior art, and are not described herein again.
S6, matching the target description sentence with the vocabulary in the preset abnormal vocabulary database, if the matching is successful, determining the corresponding target image as a suspected abnormal image, and sending the corresponding target image to the corresponding network administrator for checking; otherwise, the corresponding target image is determined to be a non-abnormal image.
In some embodiments of the invention, the new layout image is processed using an image description method (imaging) based on super-resolution reconstruction, resulting in a corresponding description sentence of the new layout image. The super-resolution reconstruction is firstly carried out on the image, and then the image description processing is carried out on the image on the basis of the reconstruction. If the description sentence contains abnormal words (such as viruses, violence and other words), the words are successfully matched with words in a preset abnormal word database, the words are determined to be suspected abnormal images, and the images are sent to a network administrator for checking; otherwise, the image is not considered as an abnormal image.
In order to solve the technical problem that abnormal images harmful to network safety cannot be detected comprehensively and accurately in the prior art, the method utilizes a multi-OCR technology mutual inspection method to identify characters in the images, obviously improves the accuracy of character identification, and further effectively ensures the accuracy of abnormal image identification. Meanwhile, the similarity calculation method based on the difference hole convolution kernel is utilized, the calculation precision of the similarity between the newly issued target image and the template image is remarkably improved, and the suspected abnormal image is accurately identified. The invention also utilizes an image description method based on super-resolution reconstruction to process the image, thereby obviously improving the generation quality of the image description sentence. The invention also combines the block chain technology to carry out the uplink storage of the recorded information of the viewer corresponding to the newly released image, thereby providing real and effective data for the subsequent responsibility confirmation and improving the reliability of abnormal image detection. The invention can carry out accurate image comparison and character recognition, ensures the real reliability of all data by combining the block chain technology and greatly improves the detection precision of abnormal images.
Based on the first aspect, in some embodiments of the present invention, the block chain-based network security anomaly image big data detection method further includes the following steps:
and acquiring and extracting corresponding image publisher information in the block chain according to the checking result of the network administrator.
In order to further improve the effectiveness of follow-up responsibility, if the newly issued image is identified as a suspected abnormal image and is confirmed as an abnormal image by a network administrator, the related information stored in the block chain is extracted, so that accurate and real data reference can be provided for the legal responsibility of the image issuer.
As shown in fig. 4, in a second aspect, an embodiment of the present invention provides a system for detecting a network security abnormal image big data based on a block chain, including: an image winding module 100, a character recognition module 200, a first vocabulary matching module 300, a similarity determination module 400, an image description module 500, and a second vocabulary matching module 600, wherein:
the image chaining module 100 is configured to acquire a target image newly published in a network and corresponding image publisher information, acquire record information of a viewer corresponding to the target image, and upload the target image, the image publisher information, and the record information of the corresponding viewer to a block chain for storage;
the character recognition module 200 is configured to recognize characters in the target image by using a multi-OCR technology mutual verification method to obtain a target character recognition result;
the first vocabulary matching module 300 is used for matching the target character recognition result with the vocabulary in the preset abnormal vocabulary database, if the matching is successful, the corresponding target image is regarded as a suspected abnormal image, and the corresponding target image is sent to the corresponding network manager for checking; otherwise, the similarity judging module 400 works;
the similarity judging module 400 is used for selecting a plurality of abnormal images as template images and calculating the similarity between the target image and each template image by using a similarity calculation method based on a difference cavity convolution kernel; if the similarity between the target image and any template image is greater than a preset similarity threshold, the corresponding target image is regarded as a suspected abnormal image, and the corresponding target image is sent to a corresponding network administrator for checking; otherwise, the image description module 500 works;
the image description module 500 is configured to process the target image by using an image description method based on super-resolution reconstruction to obtain a target description statement corresponding to the target image;
the second vocabulary matching module 600 is configured to match the target description sentence with vocabularies in a preset abnormal vocabulary database, and if the matching is successful, regard the corresponding target image as a suspected abnormal image, and send the corresponding target image to a corresponding network administrator for verification; otherwise, the corresponding target image is determined to be a non-abnormal image.
In order to solve the technical problem that abnormal images harmful to network security cannot be detected comprehensively and accurately in the prior art, the system identifies the characters in the images by using a method of mutual inspection of multiple OCR technologies through the cooperation of a plurality of modules such as the image chaining module 100, the character recognition module 200, the first vocabulary matching module 300, the similarity judging module 400, the image description module 500, the second vocabulary matching module 600 and the like, so that the accuracy of character recognition is remarkably improved, and the accuracy of abnormal image recognition is further effectively ensured. Meanwhile, the similarity calculation method based on the difference hole convolution kernel is utilized, the calculation precision of the similarity between the newly issued target image and the template image is remarkably improved, and the suspected abnormal image is accurately identified. The invention also utilizes an image description method based on super-resolution reconstruction to process the image, thereby obviously improving the generation quality of the image description sentence. The invention also combines the block chain technology to carry out the uplink storage of the recorded information of the viewer corresponding to the newly released image, thereby providing real and effective data for the subsequent responsibility confirmation and improving the reliability of abnormal image detection. The invention can carry out accurate image comparison and character recognition, ensures the real reliability of all data by combining the block chain technology and greatly improves the detection precision of abnormal images.
As shown in fig. 5, in a third aspect, an embodiment of the present application provides an electronic device, which includes a memory 101 for storing one or more programs; a processor 102. The one or more programs, when executed by the processor 102, implement the method of any of the first aspects as described above.
Also included is a communication interface 103, and the memory 101, processor 102 and communication interface 103 are electrically connected to each other, directly or indirectly, to enable transfer or interaction of data. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The memory 101 may be used to store software programs and modules, and the processor 102 executes the software programs and modules stored in the memory 101 to thereby execute various functional applications and data processing. The communication interface 103 may be used for communicating signaling or data with other node devices.
The Memory 101 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like.
The processor 102 may be an integrated circuit chip having signal processing capabilities. The Processor 102 may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.
In the embodiments provided in the present application, it should be understood that the disclosed method and system and method can be implemented in other ways. The method and system embodiments described above are merely illustrative and, for example, the flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of methods and systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, which, when executed by the processor 102, implements the method according to any one of the first aspect described above. The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Claims (10)
1. A block chain-based network security abnormal image big data detection method is characterized by comprising the following steps:
s1, acquiring a target image newly issued in the network and corresponding image issuer information, acquiring the record information of a viewer corresponding to the target image, and uploading the target image, the image issuer information and the record information of the corresponding viewer to a block chain for storage;
s2, recognizing characters in the target image by using a multi-OCR technology mutual-verification method to obtain a target character recognition result;
s3, matching the target character recognition result with the vocabulary in the preset abnormal vocabulary database, if the matching is successful, determining the corresponding target image as a suspected abnormal image, and sending the corresponding target image to the corresponding network administrator for checking; otherwise, go to step S4;
s4, selecting a plurality of abnormal images as template images, and calculating the similarity of the target image and each template image by using a similarity calculation method based on a difference hole convolution kernel; if the similarity between the target image and any template image is greater than a preset similarity threshold, the corresponding target image is regarded as a suspected abnormal image, and the corresponding target image is sent to a corresponding network administrator for checking; otherwise, go to step S5;
s5, processing the target image by using an image description method based on super-resolution reconstruction to obtain a target description sentence corresponding to the target image;
s6, matching the target description sentence with the vocabulary in the preset abnormal vocabulary database, if the matching is successful, determining the corresponding target image as a suspected abnormal image, and sending the corresponding target image to the corresponding network administrator for checking; otherwise, the corresponding target image is determined to be a non-abnormal image.
2. The method for detecting the abnormal image big data of the network security based on the block chain as claimed in claim 1, wherein the method for recognizing the characters in the target image by using the multi-OCR technology mutual verification method to obtain the target character recognition result comprises the following steps:
respectively identifying characters in the target image by adopting a plurality of OCR (optical character recognition) models to obtain a plurality of character identification results;
and counting and determining a target character recognition result according to each character recognition result.
3. The method for detecting the large data of the network security abnormal image based on the block chain according to claim 1, wherein the method for calculating the similarity between the target image and each template image by using the similarity calculation method based on the difference hole convolution kernel comprises the following steps:
respectively filtering the target image and the template image by using a convolution kernel with a void ratio of N, respectively carrying out sparse coding on the filtered target image and template image, and calculating the similarity between the target image and the template image by using the Euclidean distance to obtain a first similarity result; n is a natural number more than or equal to 1;
respectively filtering the target image and the template image by using a convolution kernel with a void ratio of N +1, respectively carrying out sparse coding on the filtered target image and template image, and calculating the similarity between the target image and the template image by using the Euclidean distance to obtain a second similarity result;
and respectively filtering the target image and the template image by using a convolution kernel with the void ratio of N +2, respectively carrying out sparse coding on the filtered target image and template image, and calculating the similarity between the target image and the template image by using the Euclidean distance to obtain a third similarity result.
4. The method for detecting the network security abnormal image big data based on the block chain according to claim 3, characterized by further comprising the following steps:
and if at least one of the first similarity result, the second similarity result and the third similarity result is greater than a preset similarity threshold, determining the corresponding target image as a suspected abnormal image.
5. The method for detecting the image big data of the network security anomaly based on the block chain according to the claim 1, wherein the method for processing the target image by using the image description method based on the super-resolution reconstruction comprises the following steps:
performing super-resolution reconstruction on the target image;
and performing image description processing on the reconstructed target image by adopting an image description method.
6. The method for detecting network security anomaly image big data based on the blockchain as claimed in claim 5, wherein the image description method comprises image description based on CNN + LSTM or image description based on transform + LSTM.
7. The method for detecting the network security abnormal image big data based on the block chain according to claim 1, characterized by further comprising the following steps:
and acquiring and extracting corresponding image publisher information in the block chain according to the checking result of the network administrator.
8. A block chain-based network security abnormal image big data detection system is characterized by comprising: image chain module, character recognition module, first vocabulary matching module, similarity judge module, image description module and second vocabulary matching module, wherein:
the image chaining module is used for acquiring a target image newly released in the network and corresponding image publisher information, acquiring the record information of a viewer corresponding to the target image, and uploading the target image, the image publisher information and the record information of the corresponding viewer to the block chain for storage;
the character recognition module is used for recognizing characters in the target image by utilizing a multi-OCR technology mutual-checking method so as to obtain a target character recognition result;
the first vocabulary matching module is used for matching the target character recognition result with vocabularies in a preset abnormal vocabulary database, if the matching is successful, the corresponding target image is regarded as a suspected abnormal image, and the corresponding target image is sent to a corresponding network manager for checking; otherwise, the similarity judging module works;
the similarity judging module is used for selecting a plurality of abnormal images as template images and calculating the similarity between the target image and each template image by using a similarity calculation method based on the difference cavity convolution kernel; if the similarity between the target image and any template image is greater than a preset similarity threshold, the corresponding target image is regarded as a suspected abnormal image, and the corresponding target image is sent to a corresponding network administrator for checking; otherwise, the image description module works;
the image description module is used for processing the target image by using an image description method based on super-resolution reconstruction so as to obtain a target description statement corresponding to the target image;
the second vocabulary matching module is used for matching the target description sentence with the vocabulary in the preset abnormal vocabulary database, if the matching is successful, the corresponding target image is regarded as a suspected abnormal image, and the corresponding target image is sent to the corresponding network manager for checking; otherwise, the corresponding target image is determined to be a non-abnormal image.
9. An electronic device, comprising:
a memory for storing one or more programs;
a processor;
the one or more programs, when executed by the processor, implement the method of any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210844485.7A CN114913534A (en) | 2022-07-19 | 2022-07-19 | Block chain-based network security abnormal image big data detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210844485.7A CN114913534A (en) | 2022-07-19 | 2022-07-19 | Block chain-based network security abnormal image big data detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114913534A true CN114913534A (en) | 2022-08-16 |
Family
ID=82772517
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210844485.7A Pending CN114913534A (en) | 2022-07-19 | 2022-07-19 | Block chain-based network security abnormal image big data detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114913534A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115171144A (en) * | 2022-09-05 | 2022-10-11 | 北京吉道尔科技有限公司 | Intelligent education test paper big data evaluation method and system based on block chain |
| CN115205866A (en) * | 2022-09-16 | 2022-10-18 | 北京吉道尔科技有限公司 | Block chain-based scientific and technological paper big data plagiarism detection method and system |
| CN115205766A (en) * | 2022-09-16 | 2022-10-18 | 北京吉道尔科技有限公司 | Block chain-based network security abnormal video big data detection method and system |
| CN116204594A (en) * | 2023-05-05 | 2023-06-02 | 中国民航信息网络股份有限公司 | Data processing method, device and equipment based on block chain |
| CN116612476A (en) * | 2023-07-20 | 2023-08-18 | 合肥图迅电子科技有限公司 | Chip character detection method and device, computer equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111405360A (en) * | 2020-03-25 | 2020-07-10 | 腾讯科技（深圳）有限公司 | Video processing method and device, electronic equipment and storage medium |
| CN111709414A (en) * | 2020-06-29 | 2020-09-25 | 济南浪潮高新科技投资发展有限公司 | AR device, character recognition method and device thereof, and computer-readable storage medium |
| CN112036273A (en) * | 2020-08-19 | 2020-12-04 | 泰康保险集团股份有限公司 | Image identification method and device |
| CN112989921A (en) * | 2020-12-31 | 2021-06-18 | 上海智臻智能网络科技股份有限公司 | Target image information identification method and device |
| CN113935774A (en) * | 2021-10-15 | 2022-01-14 | 北京百度网讯科技有限公司 | Image processing method, image processing device, electronic equipment and computer storage medium |
| CN114626024A (en) * | 2022-05-12 | 2022-06-14 | 北京吉道尔科技有限公司 | Internet infringement video low-consumption detection method and system based on block chain |
-
2022
- 2022-07-19 CN CN202210844485.7A patent/CN114913534A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111405360A (en) * | 2020-03-25 | 2020-07-10 | 腾讯科技（深圳）有限公司 | Video processing method and device, electronic equipment and storage medium |
| CN111709414A (en) * | 2020-06-29 | 2020-09-25 | 济南浪潮高新科技投资发展有限公司 | AR device, character recognition method and device thereof, and computer-readable storage medium |
| CN112036273A (en) * | 2020-08-19 | 2020-12-04 | 泰康保险集团股份有限公司 | Image identification method and device |
| CN112989921A (en) * | 2020-12-31 | 2021-06-18 | 上海智臻智能网络科技股份有限公司 | Target image information identification method and device |
| CN113935774A (en) * | 2021-10-15 | 2022-01-14 | 北京百度网讯科技有限公司 | Image processing method, image processing device, electronic equipment and computer storage medium |
| CN114626024A (en) * | 2022-05-12 | 2022-06-14 | 北京吉道尔科技有限公司 | Internet infringement video low-consumption detection method and system based on block chain |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115171144A (en) * | 2022-09-05 | 2022-10-11 | 北京吉道尔科技有限公司 | Intelligent education test paper big data evaluation method and system based on block chain |
| CN115205866A (en) * | 2022-09-16 | 2022-10-18 | 北京吉道尔科技有限公司 | Block chain-based scientific and technological paper big data plagiarism detection method and system |
| CN115205766A (en) * | 2022-09-16 | 2022-10-18 | 北京吉道尔科技有限公司 | Block chain-based network security abnormal video big data detection method and system |
| CN116204594A (en) * | 2023-05-05 | 2023-06-02 | 中国民航信息网络股份有限公司 | Data processing method, device and equipment based on block chain |
| CN116612476A (en) * | 2023-07-20 | 2023-08-18 | 合肥图迅电子科技有限公司 | Chip character detection method and device, computer equipment and storage medium |
| CN116612476B (en) * | 2023-07-20 | 2023-09-22 | 合肥图迅电子科技有限公司 | Chip character detection method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114913534A (en) | Block chain-based network security abnormal image big data detection method and system | |
| CA2738480C (en) | Detection of confidential information | |
| CN115205766A (en) | Block chain-based network security abnormal video big data detection method and system | |
| CN112257613B (en) | Physical examination report information structured extraction method and device and computer equipment | |
| CN111639648B (en) | Certificate identification method, device, computing equipment and storage medium | |
| CN113381962B (en) | Data processing method, device and storage medium | |
| CN110968689A (en) | Training method of criminal name and law bar prediction model and criminal name and law bar prediction method | |
| CN114626024A (en) | Internet infringement video low-consumption detection method and system based on block chain | |
| CN113836128A (en) | Abnormal data identification method, system, equipment and storage medium | |
| CN115205866A (en) | Block chain-based scientific and technological paper big data plagiarism detection method and system | |
| CN114579523A (en) | Double-recording file quality inspection method and device, computer equipment and storage medium | |
| CN111107074B (en) | Method, training method and device for preventing network crawler from stealing private data | |
| EP3355240B1 (en) | A method and a system for generating a multi-level classifier for image processing | |
| CN115862638A (en) | Financial transaction operation and big data secure storage method and system based on block chain | |
| CN112597997A (en) | Region-of-interest determining method, image content identifying method and device | |
| CN114996707B (en) | Static detection method and device for picture Trojan horse, electronic equipment and storage medium | |
| CN115223022B (en) | Image processing method, device, storage medium and equipment | |
| CN116127106A (en) | Block chain-based intelligent storage method and system for Internet image big data | |
| CN115982388A (en) | Case quality control map establishing method, case document quality testing method, case quality control map establishing equipment and storage medium | |
| CN111277465A (en) | Abnormal data message detection method and device and electronic equipment | |
| US20200226162A1 (en) | Automated Reporting System | |
| CN115759964A (en) | Method, system and equipment for verifying universal qualification of bid document | |
| CN111612157B (en) | Training method, character recognition device, storage medium and electronic equipment | |
| CN111899833A (en) | Medical data processing method and device, server and storage medium | |
| CN115114588B (en) | Intelligent education academic achievement discussion anti-plagiarism method and system based on block chain |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |