Urban map video stream encryption method and system based on artificial intelligence

Abstract

The invention discloses a city map video stream encryption method and system based on artificial intelligence, relates to the technical field of video stream encryption, and solves the problem that an existing video stream encryption method lacks flexibility. The method comprises the steps that S1, a video acquisition module acquires video stream content of a city map and sends the video stream content to a video recognition module; the video identification module identifies and counts semantic elements in the video stream content, and video semantic data of the video stream content obtained through statistics is sent to the content rating module, and S3, the content rating module comprehensively evaluates the video encryption demand of the video stream content, s4, the encryption execution module executes corresponding encryption operation on the video stream content to obtain an output video stream, the output video stream is sent to the video terminal through the server, and the video terminal displays the output video stream. According to the invention, intelligent video hierarchical encryption based on video stream content features is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of video stream encryption, and in particular to an artificial intelligence-based city map video stream encryption method and system. Background Art

[0002] City map video streaming technology is increasingly becoming an important tool for smart city management. With the acceleration of urbanization and the increase in population density, the demand for real-time monitoring and data analysis has become more urgent. Through cameras deployed in every corner of the city, managers can obtain real-time information such as traffic conditions, public safety and environmental changes. With the help of deep learning and computer vision, the system can automatically identify key elements in the video, such as vehicles, pedestrians and potential safety hazards, thereby providing data basis for decision-making.

[0003] Under the current technical background, most encryption methods for city map video streams use unified encryption methods and encryption standards. However, in actual work, it is not necessary to perform high-intensity encryption on all images or all areas in the video stream. The uniform-intensity encryption results in unnecessary consumption of computing resources during the encryption process. At the same time, the lack of differentiation in the encryption degree is not conducive to providing customizable decryption permissions to users with different permissions, which is not conducive to improving the user experience of the video stream. To this end, the present invention proposes an artificial intelligence-based city map video stream encryption method and system. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention aims to provide an artificial intelligence-based city map video stream encryption method and system.

[0005] The technical problems to be solved by the present invention are: The problem of how to realize intelligent video graded encryption based on video stream content characteristics.

[0006] In the first aspect, in order to achieve the above-mentioned purpose, the present invention adopts the following technical solutions: A city map video stream encryption method based on artificial intelligence, the method steps include: Step S1, the video acquisition module acquires the video stream content of the city map, and the video acquisition module sends the video stream content of the city map to the video recognition module via the server; Step S2, the video recognition module recognizes and counts the semantic elements in the video stream content, and sends the video semantic data of the video stream content obtained by counting to the content rating module via the server; Step S3, the content rating module obtains and comprehensively evaluates the video encryption requirements of the video stream content according to the video semantic data, and sends the encryption requirement level of the video stream content obtained by the evaluation to the encryption execution module via the server; Step S4, the encryption execution module executes the corresponding encryption scheme on the video stream content according to the encryption requirement level. After the encryption scheme is executed, the output video stream is sent to the video terminal through the server, and the video terminal displays the output video stream.

[0007] Furthermore, the video semantic data includes the element level and element coverage of each semantic element in the video stream content.

[0008] Furthermore, in step S2, the recognition process of the semantic elements is specifically as follows: Semantic elements are semantically layered according to the privacy or sensitivity of the identified object; The result of semantic stratification is the element level of semantic elements, which includes high-sensitivity content, medium-sensitivity content and low-sensitivity content; Among them, the encryption requirement level of low-sensitivity content is lower than that of medium-sensitivity content, and the encryption requirement level of medium-sensitivity content is lower than that of high-sensitivity content.

[0009] Furthermore, in step S2, the statistical process of the semantic elements specifically includes: Step S201, obtaining video stream content, delineating semantic elements according to the recognition results of semantic elements, determining contours of the semantic elements according to the edge extraction results, and marking the area inside the contours as the element coverage of the semantic elements; Step S202, calculating the element coverage area of ​​each semantic element in the video stream content; Step S203: integrating the element level and element coverage area of ​​each semantic element to obtain video semantic data of the video stream content.

[0010] Furthermore, in step S3, the evaluation process of the video encryption requirement is as follows: Obtaining the semantic layering of each semantic element in the video stream content, and selecting the corresponding encryption strength coefficient according to the semantic layering; Then, the element coverage area of ​​each semantic element in the video stream content is obtained, and the encryption requirement value of the video stream content is calculated; Comparing the encryption requirement value of the video stream content with the encryption requirement threshold; If the encryption requirement value is less than or equal to the first encryption requirement threshold, determining that the encryption requirement level of the video stream content is the first encryption requirement level; If the encryption requirement value is greater than the first encryption requirement threshold and less than or equal to the second encryption requirement threshold, determining that the encryption requirement level of the video stream content is the second encryption requirement level; If the encryption requirement value is greater than the second encryption requirement threshold, determining that the encryption requirement level of the video stream content is the third encryption requirement level; Among them, the values ​​of the encryption requirement thresholds are all greater than zero, the first encryption requirement threshold is less than the second encryption requirement threshold, the encryption strength corresponding to the first encryption requirement level is lower than the encryption strength corresponding to the second encryption requirement level, and the encryption strength corresponding to the second encryption requirement level is lower than the encryption strength corresponding to the third encryption requirement level.

[0011] Furthermore, in step S3, the corresponding relationship between the encryption strength coefficient and the semantic layer is as follows: If the semantic layer of the semantic element is low-sensitivity content, the encryption strength coefficient corresponding to the semantic element is q1; If the semantic layer of the semantic element is medium sensitive content, the encryption strength coefficient corresponding to the semantic element is q2; If the semantic layer of the semantic element is highly sensitive content, the encryption strength coefficient corresponding to the semantic element is q3; Among them, the encryption strength coefficient is used to reflect the encryption requirement degree of the semantic elements. The values ​​of the encryption strength coefficient are all greater than zero, q1<q2<q3.

[0012] Furthermore, in step S4, the corresponding relationship between the encryption scheme and the encryption requirement level specifically includes: If the encryption requirement level of the video stream content is the first encryption requirement level, the corresponding encryption scheme is low-level encryption; If the encryption requirement level of the video stream content is the second encryption requirement level, the corresponding encryption scheme is intermediate encryption; If the encryption requirement level of the video stream content is the third encryption requirement level, the corresponding encryption scheme is advanced encryption.

[0013] Furthermore, in step S4, the encrypted content of the video stream content is specifically as follows: Step S401, converting the video stream content to be encrypted into data plaintext, and dividing the data plaintext into plaintext data blocks of fixed size; Step S402, setting a key for encryption; Step S403, converting the plaintext data block and the key into binary form, and then performing an XOR operation on the plaintext data block in binary form and the key; Step S404, converting the calculation result of the XOR calculation back into hexadecimal to obtain an encrypted data block; Step S405, after replacing the plaintext data block with the encrypted data block, several rounds of encryption operations are performed on the encrypted data block; Step S406, after completing all encryption rounds, a final output video stream is generated. Specifically, the output video stream has the same video format as the video stream content, and the output video stream replaces the original video stream content.

[0014] Furthermore, in step S4, each round of encryption operations specifically includes: Step A1, byte replacement, each byte in the encrypted data block is replaced by a predefined replacement table; Step A2, row shift, the data rows of the encrypted data block are circularly shifted left, the first row of data rows remains unchanged, the second row of data rows is shifted left by 1 byte, the third row of data rows is shifted left by 2 bytes, and the fourth row of data rows is shifted left by 3 bytes; Step A3, column obfuscation, each data column of the encrypted data block is mathematically transformed with the transformation matrix of the AES algorithm; Among them, the 128-bit key corresponds to 10 encryption rounds, the 192-bit key corresponds to 12 encryption rounds, and the 256-bit key corresponds to 14 encryption rounds.

[0015] In a second aspect, the present invention further proposes: an artificial intelligence-based city map video stream encryption system, the system comprising: Video acquisition module: used to obtain the video stream content of the city map; Video recognition module: used to identify the semantic elements appearing in the video stream content, and then perform range statistics on each semantic element to obtain the video semantic data of the video stream content; Content rating module: used to obtain video semantic data of video stream content, and conduct a comprehensive evaluation of the video encryption requirements of the video stream content based on the video semantic data, and obtain the encryption requirement level of the video stream content through evaluation; Encryption execution module: used to execute the corresponding encryption scheme on the video stream content according to the encryption requirement level, and obtain the output video stream after the encryption scheme is executed; Video terminal: used to display the output video stream.

[0016] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: The present invention firstly obtains the video stream content of the city map through the video acquisition module and sends it to the video recognition module, and then uses the video recognition module to identify and range-count the semantic elements in the video stream content, and obtains the video semantic data of the video stream content through statistics and sends it to the content rating module, and then uses the content rating module to comprehensively evaluate the video encryption requirements of the video stream content according to the video semantic data, and the encryption requirement level of the video stream content obtained through the evaluation is sent to the encryption execution module, and finally uses the encryption execution module to execute the corresponding encryption scheme on the video stream content, and the encrypted output video stream is sent to the video terminal for display. The present invention realizes the intelligent hierarchical encryption of the video stream content. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0018] Figure 1 is a flow chart of the method of the present invention; Figure 2 is the overall system block diagram of the present invention; Figure 3 It is a flowchart of the encryption execution module in the present invention. DETAILED DESCRIPTION

[0019] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Example 1: Please refer to Figure 1-Figure 3 As shown, the technical solution provided by the present invention is: a city map video stream encryption method based on artificial intelligence, comprising the following steps: Step S1, obtaining the video stream content of the city map through the video acquisition module, the video acquisition module sends the video stream content of the city map to the server, and the server sends the video stream content of the city map to the video recognition module; In this embodiment, the video acquisition module is specifically a video recording device for acquiring video stream content. In actual working process, the video recording device includes but is not limited to various monitoring probes, high-definition cameras, and aerial drones installed in urban areas. In this embodiment, monitoring probes installed around urban roads are preferably used as video recording devices; In step S1, the video stream content is specifically a video file captured by a video recording device through a lens. It should be specifically noted that the encryption state of the video stream content is an unencrypted file.

[0021] Step S2, the video recognition module recognizes the semantic elements appearing in the video stream content through the deep learning model, and then performs range statistics on each semantic element, and sends the video semantic data of the video stream content obtained by statistics to the server, and the server sends the video semantic data of the video stream content to the content rating module; The video semantic data includes the element level and element coverage of each semantic element in the video stream content, and the semantic element is specifically an identification object appearing in the video stream content; In step S2, the semantic element recognition process is as follows: Semantic elements are semantically layered according to the privacy or sensitivity of the identified object; The result of semantic stratification is the element level of the semantic element, which includes high-sensitivity content, medium-sensitivity content and low-sensitivity content. The encryption requirement of low-sensitivity content is lower than that of medium-sensitivity content, and the encryption requirement of medium-sensitivity content is lower than that of high-sensitivity content. For example, in the city map scene, vehicle license plates, pedestrian faces, and semantic elements of the same sensitivity level are layered into high-sensitivity content, key places of action (such as hospitals, schools, etc.), core buildings, and semantic elements of the same sensitivity level are layered into medium-sensitivity content, and urban roads, non-core buildings, public building complexes, and semantic elements of the same sensitivity level are layered into low-sensitivity content; In step S2, the statistical process of semantic elements specifically includes: Step S201, obtaining video stream content, delineating semantic elements according to the recognition results of semantic elements, determining contours of the semantic elements according to the edge extraction results, and marking the area inside the contours as the element coverage of the semantic elements; Specifically, the element coverage is the area occupied by the semantic element in the video stream content. In this embodiment, the video stream content is denoised and the semantic element is located and the edge is extracted by the Laplacian operator; Step S202, obtaining the element coverage area of ​​each semantic element in the video stream content by using OpenCV technology; In the actual work process, the OpenCV technology is used to calculate the area of ​​the corresponding element coverage of each semantic element, that is, the area of ​​the area delimited by the contour line of the semantic element is calculated as the element coverage area of ​​the semantic element; Step S203: integrating the element level and element coverage area of ​​each semantic element to obtain video semantic data of the video stream content.

[0022] Step S3, the content rating module obtains video semantic data of the video stream content, performs a comprehensive evaluation on the video encryption requirement of the video stream content according to the video semantic data, and sends the encryption requirement level of the video stream content obtained by the evaluation to the server, and the server sends the encryption requirement level of the video stream content to the encryption execution module; In an embodiment of the present invention, in step S3, the evaluation process of the video encryption requirement is as follows: Obtain the semantic hierarchy of each semantic element in the video stream content, and select the corresponding encryption strength coefficient QDi according to the semantic hierarchy, where i is the sequence number of the semantic element, the upper limit value of i is n, and the upper limit value of n is equal to the number of semantic elements in the video stream content; It should be noted that the corresponding relationship between the encryption strength coefficient and the semantic layer is as follows: If the semantic layer of the semantic element is low-sensitivity content, the encryption strength coefficient corresponding to the semantic element is q1; If the semantic layer of the semantic element is medium sensitive content, the encryption strength coefficient corresponding to the semantic element is q2; If the semantic layer of the semantic element is highly sensitive content, the encryption strength coefficient corresponding to the semantic element is q3; Among them, the encryption strength coefficient is used to reflect the encryption requirement of the semantic elements, and the values ​​of the encryption strength coefficient are all greater than zero, q1<q2<q3; Then, the element coverage area FMi of each semantic element in the video stream content is obtained, and the encryption requirement value JM of the video stream content is calculated according to the formula. The specific formula is as follows: ; Comparing the encryption requirement value of the video stream content with the encryption requirement threshold; If the encryption requirement value is less than or equal to the first encryption requirement threshold, determining that the encryption requirement level of the video stream content is the first encryption requirement level; If the encryption requirement value is greater than the first encryption requirement threshold and less than or equal to the second encryption requirement threshold, determining that the encryption requirement level of the video stream content is the second encryption requirement level; If the encryption requirement value is greater than the second encryption requirement threshold, determining that the encryption requirement level of the video stream content is the third encryption requirement level; Among them, the values ​​of the encryption requirement thresholds are all greater than zero, the first encryption requirement threshold is less than the second encryption requirement threshold, the encryption strength corresponding to the first encryption requirement level is lower than the encryption strength corresponding to the second encryption requirement level, and the encryption strength corresponding to the second encryption requirement level is lower than the encryption strength corresponding to the third encryption requirement level.

[0023] Step S4, the encryption execution module executes the corresponding encryption scheme on the video stream content according to the encryption requirement level, and after the encryption scheme is executed, the output video stream is sent to the server, and the server sends the output video stream to the video terminal, and the video terminal displays the output video stream; In step S4, the corresponding relationship between the encryption scheme and the encryption requirement level specifically includes: If the encryption requirement level of the video stream content is the first encryption requirement level, the corresponding encryption scheme is low-level encryption; If the encryption requirement level of the video stream content is the second encryption requirement level, the corresponding encryption scheme is intermediate encryption; If the encryption requirement level of the video stream content is the third encryption requirement level, the corresponding encryption scheme is advanced encryption; In this embodiment, low-level encryption preferably performs blurring on the video stream content, intermediate encryption preferably performs data encryption on the video stream content using the AES-128 algorithm, and high-level encryption preferably performs data encryption on the video stream content using the AES-256 algorithm; See also Figure 3 As shown, in step S4, the encrypted content of the video stream content is as follows: Step S401, converting the video stream content to be encrypted into data plaintext, and dividing the data plaintext into plaintext data blocks of fixed size. In this embodiment, the AES algorithm selects a plaintext data block with a data block size of 128 bits (16 bytes); Step S402, setting a key for encryption. In this embodiment, the key length corresponding to the AES-128 algorithm is 128 bits, and the key length corresponding to the AES-256 algorithm is 256 bits; Step S403, converting the plaintext data block and the key into binary form, and then performing an XOR operation on the plaintext data block in binary form and the key; For example, the value of the plaintext data block D0 is 0x32, which is represented as 00110010 in binary form, and the value of the key K0 is 0x2b, which is represented as 00101011 in binary form. The calculation process is as follows: 00110010⊕00101011=00011001; Step S404, converting the calculation result of the XOR calculation back into hexadecimal to obtain an encrypted data block; For example, 00011001 is converted to hexadecimal as 0x19, and the relationship equation is as follows: D0⊕K0=0x32⊕0x2b=00110010⊕00101011=00011001=0x19; Step S404, after replacing the plaintext data block with the encrypted data block, several rounds of encryption operations are performed on the encrypted data block; The corresponding relationship between encryption rounds and key length is as follows: A 128-bit key corresponds to 10 encryption rounds, a 192-bit key corresponds to 12 encryption rounds, and a 256-bit key corresponds to 14 encryption rounds. Each round consists of the following steps: Step A1, byte substitution (SubBytes), each byte in the encrypted data block is replaced by a predefined substitution table (S-Box). It should be specifically explained that the substitution table can be understood as a dictionary, and each byte in the dictionary is replaced by another byte; Step A2, row shift (ShiftRows): the data rows of the encrypted data block are circularly shifted left, wherein the first row of data rows remains unchanged, the second row of data rows is shifted left by 1 byte, the third row of data rows is shifted left by 2 bytes, and the fourth row of data rows is shifted left by 3 bytes; Step A3, Mix Columns: Each data column of the encrypted data block is mathematically transformed with the transformation matrix of the AES algorithm; For example, the data column of the encrypted data block is , the transformation matrix of the AES algorithm is ; The first line calculates: Output1=(0x02×0x01)⊕(0x03×0x01)⊕(0x01×0x01)⊕(0x01×0x01); 0x02×0x01=0x02; 0x03×0x01=0x03; 0x01×0x01=0x01; 0x01×0x01=0x01; Output1=0x02⊕0x03⊕0x01⊕0x01=0x02⊕0x03=0x00; Similarly, the calculation results of the second, third, and fourth rows are as follows: Output2=0x03, Output3=0x02, Output4=0x02; Finally, the output data is listed as ; Step S406, after completing all encryption rounds, a final output video stream is generated. Specifically, the output video stream has the same video format as the video stream content, and the output video stream replaces the original video stream content.

[0024] In this application, if corresponding calculation formulas appear, the above calculation formulas are all dimensionless and take their numerical calculations. The weight coefficients, proportional coefficients and other coefficients in the formulas are set to a result value obtained by quantifying each parameter. The size of the weight coefficient and the proportional coefficient can be determined as long as it does not affect the proportional relationship between the parameter and the result value.

[0025] Embodiment 2: Figure 2 and Figure 3 As shown, based on another concept of the same invention, an artificial intelligence-based city map video stream encryption system is now proposed, including a video acquisition module, a video recognition module, a content rating module, an encryption execution module, a video terminal and a server; The video acquisition module is used to acquire the video stream content of the city map; The video recognition module is used to identify the semantic elements appearing in the video stream content, and then perform range statistics on each semantic element to obtain the video semantic data of the video stream content. The specific working process is as follows: Semantic elements are semantically layered according to the privacy or sensitivity of the identified object. The result of semantic layering is the element level of the semantic element. Then the video stream content is obtained, the semantic element is circled according to the recognition result of the semantic element, the contour of the semantic element is determined according to the edge extraction result, and the area inside the contour is marked as the element coverage of the semantic element. The element coverage area of ​​each semantic element in the video stream content is obtained through OpenCV technology, and the element level and element coverage area of ​​each semantic element are integrated to obtain the video semantic data of the video stream content: The content rating module is used to obtain video semantic data of video stream content, and to perform a comprehensive evaluation on the video encryption requirements of the video stream content according to the video semantic data, and to obtain the encryption requirement level of the video stream content through the evaluation. The specific working process is as follows: Obtaining the semantic layering of each semantic element in the video stream content, selecting the corresponding encryption strength coefficient according to the semantic layering, then obtaining the element coverage area of ​​each semantic element in the video stream content, and calculating the encryption requirement value of the video stream content, comparing the encryption requirement value of the video stream content with the encryption requirement threshold, and determining whether the encryption requirement level of the video stream content is the first encryption requirement level, the second encryption requirement level, or the third encryption requirement level; The encryption execution module is used to execute a corresponding encryption scheme on the video stream content according to the encryption requirement level, and obtain an output video stream after the encryption scheme is executed; The video terminal is used to display the output video stream.

[0026] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A city map video stream encryption method based on artificial intelligence, characterized in that: The method steps include: Step S1, the video acquisition module acquires the video stream content of the city map, and the video acquisition module sends the video stream content of the city map to the video recognition module via the server; Step S2, the video recognition module recognizes and counts the semantic elements in the video stream content, and sends the video semantic data of the video stream content obtained by counting to the content rating module via the server; Step S3, the content rating module obtains and comprehensively evaluates the video encryption requirements of the video stream content according to the video semantic data, and sends the encryption requirement level of the video stream content obtained by the evaluation to the encryption execution module via the server; Step S4, the encryption execution module executes the corresponding encryption scheme on the video stream content according to the encryption requirement level. After the encryption scheme is executed, the output video stream is sent to the video terminal through the server, and the video terminal displays the output video stream.

2. The method for encrypting city map video stream based on artificial intelligence according to claim 1 is characterized in that: The video semantic data includes the element level and element coverage of each semantic element in the video stream content.

3. The method for encrypting city map video stream based on artificial intelligence according to claim 2 is characterized in that: In step S2, the recognition process of the semantic elements is as follows: Semantic elements are semantically layered according to the privacy or sensitivity of the identified object; The result of semantic stratification is the element level of semantic elements, which includes high-sensitivity content, medium-sensitivity content and low-sensitivity content; Among them, the encryption requirement level of low-sensitivity content is lower than that of medium-sensitivity content, and the encryption requirement level of medium-sensitivity content is lower than that of high-sensitivity content.

4. The method for encrypting city map video stream based on artificial intelligence according to claim 3 is characterized in that: In step S2, the statistical process of the semantic elements specifically includes: Step S201, obtaining video stream content, delineating semantic elements according to the recognition results of semantic elements, determining contours of the semantic elements according to the edge extraction results, and marking the area inside the contours as the element coverage of the semantic elements; Step S202, calculating the element coverage area of ​​each semantic element in the video stream content; Step S203: integrating the element level and element coverage area of ​​each semantic element to obtain video semantic data of the video stream content.

5. The method for encrypting city map video stream based on artificial intelligence according to claim 4 is characterized in that: In step S3, the evaluation process of the video encryption requirement is as follows: Obtaining the semantic layering of each semantic element in the video stream content, and selecting the corresponding encryption strength coefficient according to the semantic layering; Then, the element coverage area of ​​each semantic element in the video stream content is obtained, and the encryption requirement value of the video stream content is calculated; Comparing the encryption requirement value of the video stream content with the encryption requirement threshold; If the encryption requirement value is less than or equal to the first encryption requirement threshold, determining that the encryption requirement level of the video stream content is the first encryption requirement level; If the encryption requirement value is greater than the first encryption requirement threshold and less than or equal to the second encryption requirement threshold, determining that the encryption requirement level of the video stream content is the second encryption requirement level; If the encryption requirement value is greater than the second encryption requirement threshold, determining that the encryption requirement level of the video stream content is the third encryption requirement level; Among them, the values ​​of the encryption requirement thresholds are all greater than zero, the first encryption requirement threshold is less than the second encryption requirement threshold, the encryption strength corresponding to the first encryption requirement level is lower than the encryption strength corresponding to the second encryption requirement level, and the encryption strength corresponding to the second encryption requirement level is lower than the encryption strength corresponding to the third encryption requirement level.

6. The method for encrypting city map video stream based on artificial intelligence according to claim 5 is characterized in that: In step S3, the corresponding relationship between the encryption strength coefficient and the semantic layer is as follows: If the semantic layer of the semantic element is low-sensitivity content, the encryption strength coefficient corresponding to the semantic element is q1; If the semantic layer of the semantic element is medium sensitive content, the encryption strength coefficient corresponding to the semantic element is q2; If the semantic layer of the semantic element is highly sensitive content, the encryption strength coefficient corresponding to the semantic element is q3; Among them, the encryption strength coefficient is used to reflect the encryption requirement degree of the semantic elements. The values ​​of the encryption strength coefficient are all greater than zero, q1<q2<q3.

7. The method for encrypting city map video stream based on artificial intelligence according to claim 5 is characterized in that: In step S4, the corresponding relationship between the encryption scheme and the encryption requirement level specifically includes: If the encryption requirement level of the video stream content is the first encryption requirement level, the corresponding encryption scheme is low-level encryption; If the encryption requirement level of the video stream content is the second encryption requirement level, the corresponding encryption scheme is intermediate encryption; If the encryption requirement level of the video stream content is the third encryption requirement level, the corresponding encryption scheme is advanced encryption.

8. The method for encrypting city map video stream based on artificial intelligence according to claim 7 is characterized in that: In step S4, the encrypted content of the video stream content is specifically as follows: Step S401, converting the video stream content to be encrypted into data plaintext, and dividing the data plaintext into plaintext data blocks of fixed size; Step S402, setting a key for encryption; Step S403, converting the plaintext data block and the key into binary form, and then performing an XOR operation on the plaintext data block in binary form and the key; Step S404, converting the calculation result of the XOR calculation back into hexadecimal to obtain an encrypted data block; Step S405, after replacing the plaintext data block with the encrypted data block, several rounds of encryption operations are performed on the encrypted data block; Step S406, after completing all encryption rounds, a final output video stream is generated. Specifically, the output video stream has the same video format as the video stream content, and the output video stream replaces the original video stream content.

9. The method for encrypting city map video stream based on artificial intelligence according to claim 8, characterized in that: In step S4, the encryption operation of each round specifically includes: Step A1, byte replacement, each byte in the encrypted data block is replaced by a predefined replacement table; Step A2, row shift, the data rows of the encrypted data block are circularly shifted left, the first row of data rows remains unchanged, the second row of data rows is shifted left by 1 byte, the third row of data rows is shifted left by 2 bytes, and the fourth row of data rows is shifted left by 3 bytes; Step A3, column obfuscation, each data column of the encrypted data block is mathematically transformed with the transformation matrix of the AES algorithm; Among them, the 128-bit key corresponds to 10 encryption rounds, the 192-bit key corresponds to 12 encryption rounds, and the 256-bit key corresponds to 14 encryption rounds.

10. The city map video stream encryption system based on artificial intelligence is characterized by: Using the artificial intelligence-based city map video stream encryption method according to any one of claims 1 to 9, the system includes: Video acquisition module: used to obtain the video stream content of the city map; Video recognition module: used to identify the semantic elements appearing in the video stream content, and then perform range statistics on each semantic element to obtain the video semantic data of the video stream content; Content rating module: used to obtain video semantic data of video stream content, and conduct a comprehensive evaluation of the video encryption requirements of the video stream content based on the video semantic data, and obtain the encryption requirement level of the video stream content through evaluation; Encryption execution module: used to execute the corresponding encryption scheme on the video stream content according to the encryption requirement level, and obtain the output video stream after the encryption scheme is executed; Video terminal: used to display the output video stream.

Images