Audio-visual content acquisition equipment, verification terminal and system

Abstract

The invention discloses audio-visual content acquisition equipment and a processing method of the acquisition equipment during audio-visual content acquisition, and also discloses a corresponding verification terminal and a corresponding verification method. Wherein a device digital certificate and a corresponding device private key are preset in the acquisition device, the device digital certificate is used for uniquely identifying the identity of the acquisition device, and a processing unit of the acquisition device can extract a basic unit in the audio-visual data, calculate a hash value corresponding to the basic unit and send the hash value corresponding to the basic unit to the acquisition device under the condition that the basic unit is a key unit. And calculating a signature corresponding to the key unit by using an equipment private key, wherein the calculated original data comprises the basic unit and a hash value corresponding to a previous basic unit of the basic unit. By adopting the scheme, the authenticity of the original acquisition equipment is guaranteed, the acquired audio-visual content has relatively strong tamper-proof capability and relatively good verifiability, the verification end can verify the authenticity of the complete audio-visual content or fragments thereof, and the authenticity and reliability of the audio-visual content are guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of audio-visual content processing, and in particular to an audio-visual content acquisition device and a corresponding verification terminal. In addition, the present application also relates to a system including the above-mentioned acquisition device and verification terminal. Background Art

[0002] With the rapid development of artificial intelligence (AI) technology, especially the application of deep learning and generative adversarial networks (GAN), current AI is able to generate highly realistic video, audio and other audio-visual content. These unreal audio-visual contents may be used to spread false information, mislead public opinion, or even be used for illegal activities such as fraud.

[0003] Traditional video and audio authenticity verification mainly uses file timestamp-based technology and basic digital watermarking technology. These technologies can prove the authenticity of audiovisual content to a certain extent, but there are still ways to circumvent or forge them, and these technologies are also difficult to deal with high-simulation content generated by AI, and users still find it difficult to identify and prevent them.

[0004] Currently, there are solutions that use cryptographic signature technology to protect audio-visual content such as video and audio. These solutions usually use the private key of the creator or copyright holder to sign the entire audio-visual content to ensure the integrity and authenticity of the entire audio-visual content. However, in some application scenarios, the verifier may only obtain a portion of the video or audio clip, but not the entire signed video or audio. In this case, it is difficult to verify whether this portion of the clip is authentic and reliable. Summary of the invention

[0005] One purpose of the present application is to provide an audio-visual content acquisition device and a processing method of the acquisition device when acquiring audio-visual content, and also provide a corresponding verification terminal and verification method. The acquisition device is used to acquire audio-visual content, and the audio-visual content has a strong anti-tampering capability and good verifiability. Whether it is the entire audio-visual content or a fragment, it can be verified by the verification terminal to ensure that the audio-visual content or its fragment is authentic and reliable.

[0006] The first aspect of the present application provides an audiovisual content acquisition device, in which a device digital certificate and a corresponding device private key are pre-set, and the device digital certificate is used to uniquely identify the identity of the acquisition device; the acquisition device includes a collection unit and a processing unit, wherein:

[0007] The acquisition unit is used to collect audio-visual data;

[0008] The processing unit is used to: extract the basic unit in the audio-visual data; calculate the hash value corresponding to the basic unit, wherein the original data used to calculate the hash value includes: the basic unit, and the hash value corresponding to the previous basic unit of the basic unit; when the basic unit is a key unit, use the device private key to calculate the signature corresponding to the key unit, wherein the original data used to calculate the signature includes the key unit and the hash value corresponding to the previous basic unit of the key unit; store the basic unit and the corresponding hash value, and, when the basic unit is the key unit, store the signature corresponding to the key unit.

[0009] In combination with the first aspect, in a possible implementation manner, the processing unit is further used to: embed a hash value corresponding to the basic unit into the basic unit or metadata of the basic unit.

[0010] In combination with the first aspect, in a possible implementation manner, the processing unit is further used to: embed the signature corresponding to the key unit into the basic unit or metadata of the basic unit.

[0011] In combination with the first aspect, in a possible implementation, the processing unit is further used to: store the hash value corresponding to the basic unit and / or the signature corresponding to the key unit in an independent file.

[0012] In combination with the first aspect, in a possible implementation manner, the processing unit is further configured to:

[0013] In the case where the basic unit is the first basic unit in the audio-visual data, the original data used to calculate the hash value includes: the first basic unit and a preset initial value.

[0014] In combination with the first aspect, in a possible implementation, the processing unit is further used to: when the basic unit is the first basic unit in the audio-visual data, the original data used to calculate the hash value includes: the first basic unit and a preset initial value.

[0015] In combination with the first aspect, in a possible implementation, when the collected audio-visual data includes video data, the basic unit is a video clip or video frame extracted according to a first preset rule, and the key unit is a key video segment or key frame selected from the video according to a second preset rule.

[0016] In combination with the first aspect, in a possible implementation method, when the collected audio-visual data includes audio data, the basic unit is an audio segment or audio frame extracted according to a third preset rule; the key unit is a key audio segment or key frame selected from the audio according to a fourth preset rule.

[0017] In combination with the first aspect, in a possible implementation manner, the acquisition device includes a security chip, and the device private key is stored in the security chip.

[0018] In combination with the first aspect, in a possible implementation, the acquisition device is also used to: divide at least one first segment from the stored audiovisual content, each of the first segments includes at least two consecutive basic units, and the at least two basic units include at least one key unit; send the at least one first segment and corresponding first verification data, wherein the first verification data includes hash values ​​corresponding to each of the at least two basic units and a signature corresponding to at least one key unit.

[0019] A second aspect of the present application provides a verification end for audiovisual content, the verification end being used to: obtain a second segment to be verified in the audiovisual content and corresponding second verification data, the second segment including at least two basic units, the at least two basic units including at least one key unit, the second verification data including hash values ​​corresponding to the at least two basic units, and a signature corresponding to at least one key unit; verify the signature corresponding to at least one key unit using a device public key of a collection device; verify the hash value corresponding to each basic unit in the second segment except the first basic unit; and determine that the second segment is an audiovisual content segment originally collected by the collection device and not modified when all hash values ​​and signatures are verified.

[0020] In combination with the second aspect, in a possible implementation, the verification end is further used to: when at least one of all hash values ​​and the signature fails to pass verification, generate prompt information to indicate that the second segment is a non-original audio-visual content segment.

[0021] In combination with the second aspect, in a possible implementation manner, the verification end is a server or a client.

[0022] A third aspect of the present application provides an audiovisual content acquisition and verification system, comprising any possible acquisition device of the first aspect, and any possible verification terminal of the second aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 The following is a schematic diagram of the system architecture of an exemplary application scenario of the present application.

[0024] Figure 2 The present invention is a flowchart of an exemplary method executed by a processing unit in a collection device according to an embodiment of the present application.

[0025] Figure 3 A schematic diagram of exemplary audio-visual data and its corresponding hash value and signature according to an embodiment of the present application.

[0026] Figure 4 The present invention is a flowchart of an exemplary verification method performed by a verification terminal according to an embodiment of the present application.

[0027] Figure 5 The figure is a schematic diagram of an exemplary audio-visual content segment to be verified and corresponding verification data according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to clearly and completely describe the technical solution of the present application, further explanation will be given below in conjunction with embodiments and drawings.

[0029] In addition to the aforementioned scheme of signing the entire audio-visual content, in other schemes, the audio-visual content such as video and audio can also be signed segmented or frame by frame, so as to ensure that each segment or frame in the audio-visual content is authentic and reliable. However, this scheme still has some defects when applied. For example, when some segments or frames in the audio-visual content are deleted, the information expressed by the audio-visual content may have changed. If the signature of each segment or frame retained in the audio-visual content is verified, it can be verified, that is, such a method cannot distinguish whether some segments or frames of the audio-visual content have been deleted. In particular, in some cases, the verifier may only be able to obtain a part of the audio-visual content, and it can neither verify by signing the entire audio-visual content (since the complete audio-visual content is not obtained, the verification of the overall signature will inevitably fail to pass), nor can it detect whether there is a deletion problem by signing each segment or frame.

[0030] To this end, the embodiments of the present application propose an audiovisual content acquisition device and a processing method for the acquisition device when acquiring audiovisual content, and also propose a corresponding verification terminal and verification method. With this solution, the authenticity of the original acquisition device is guaranteed, and the acquired audiovisual content has strong anti-tampering capabilities and good verifiability, which can ensure that the audiovisual content is authentic and reliable. However, highly simulated audiovisual content generated by AI, or modified audiovisual content, does not have such verifiability and is therefore easier to detect.

[0031] The following will first introduce the system architecture of an exemplary application scenario in which the present solution can be applied, then explain the acquisition device and its processing when acquiring audio-visual content, and then explain the verification end and its processing when verifying audio-visual content.

[0032] See also Figure 1 ,The system may include an audio-visual content acquisition device and a verification terminal, and in some cases may also include other physical and / or virtual devices, such as Figure 1 The terminal and server shown in the figure.

[0033] The acquisition device is an electronic device with acquisition and processing capabilities, such as a camera, a voice recorder, a smart monitor, or a mobile phone, tablet computer, wearable device, augmented reality (AR) / virtual reality (VR) device with acquisition and processing capabilities. This application does not limit the specific product form / type of the acquisition device.

[0034] The acquisition device is mainly used to acquire audio-visual content from the real environment, such as shooting video, recording audio, recording the changes in the device interface when the user operates the device (which can also be understood as a video), etc. Of course, in addition to acquiring video and audio, the acquisition device can also be used to acquire other forms of content, such as photos, etc., and this application does not limit this. It is understandable that the acquisition device can synchronously acquire multiple audio-visual content at the same time, or it can acquire a single audio-visual content, and this application does not limit this.

[0035] The collection device is pre-installed with a device digital certificate and a corresponding device private key. In some implementations, in order to identify the identity of each collection device and to enable verification, a pair of device keys can be generated for each collection device. The device key uses an asymmetric key, including a corresponding device private key SK and a device public key PK. Then a trusted certificate authority (CA) issues a digital certificate for the collection device based on the device public key PK, i.e., a device digital certificate, which can be used to uniquely identify the identity of the collection device. The device digital certificate and the corresponding device private key SK are embedded in the collection device for subsequent use. Exemplarily, the device digital certificate and the corresponding device private key can be embedded before the collection device leaves the factory.

[0036] In some implementations, the acquisition device may include a security chip, and the device private key may be stored in the security chip to prevent illegal access. The device digital certificate may be stored in the security chip or other storage space of the acquisition device, and may also be transmitted to other devices when necessary.

[0037] After the audiovisual data collected by the acquisition device is processed, it can form audiovisual content. The audiovisual content or part of it may be transmitted to other devices (such as Figure 1 The terminal or server shown in the figure may be stored in a storage medium). The terminal or server may directly or indirectly send the acquired audio-visual content or part thereof to the verification end, requesting the verification end to verify whether it is authentic and reliable. It is understandable that in some cases, the acquisition device may also send the audio-visual content or part thereof to the verification end by itself, and this application does not limit this.

[0038] The verification end in the embodiments of the present application can be developed as a service end and deployed in a server, or developed as a client and deployed on other possible electronic devices. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server or cloud computing service center that provides cloud services, cloud databases, cloud computing, cloud storage, network services, cloud communications, middleware services, security services, content distribution services, big data and artificial intelligence platforms and other cloud computing services, or a combination of physical servers and cloud servers. Other electronic devices can be terminal devices with verification capabilities, etc. This application does not limit the specific product forms of servers and other electronic devices.

[0039] The following will take the processing process of the acquisition device when acquiring a certain audio-visual content as an example to further explain the processing process of the acquisition device and the functions it can achieve.

[0040] In some implementations, the acquisition device may include an acquisition unit and a processing unit. The acquisition unit is mainly used to acquire audio-visual data, and may be, for example, a camera module, a microphone module, etc., and transmit the acquired audio-visual data to the processing unit for processing. It is understandable that the acquisition unit may transfer the data stream (such as a video stream or an audio stream) to the processing unit for processing while acquiring, or may acquire a relatively complete audio-visual data set or data packet and then process it by the processing unit, and this application does not limit this.

[0041] See also Figure 2 , Figure 2 This is a flowchart of an exemplary method executed by a processing unit of an embodiment of the present application. The method at least includes steps S101 to S104.

[0042] S101, extracting basic units from audio-visual data.

[0043] The basic unit is the basic processing unit for performing hash operations and signatures in the embodiments of the present application. For an audio-visual content, it can be divided into multiple basic units with order. Exemplarily, for video, each video frame can be used as a basic unit. Exemplarily again, for audio, each small segment of audio can be used as a basic unit. It can be understood that for an audio-visual content, the division of its basic units can be uniform or uneven, and can include one or more division methods. This application does not limit this, as long as it can eventually form a complete audio-visual content.

[0044] The audio-visual data here may be a data stream collected by the collection unit and transmitted to the processing unit, or may be data of a relatively complete audio-visual content, which is not limited in the present application.

[0045] The method of extracting basic units from audio-visual data may adopt different methods according to the type and format of the audio-visual data.

[0046] In some implementations, when the audiovisual content includes a video, a video frame or a video segment may be extracted from the video according to a first preset rule and used as a basic unit. Exemplarily, the first preset rule may be configured to extract frame by frame, or extract every preset number of frames (e.g., every 12 frames, etc.), or extract every preset time length (e.g., 1 s).

[0047] In some implementations, when the audiovisual content includes audio, an audio clip or audio frame extracted from the audio according to a third preset rule can be used as a basic unit. An audio frame usually divides the audio signal into a time window of a fixed size. The common audio frame size depends on the application requirements, and a time window of 20-40 milliseconds is usually used. The number of samples corresponding to each window is determined by the sampling rate of the audio. Exemplarily, the audio frame can be extracted using a sliding window method to divide the audio signal into frames by a fixed time length or number of samples. Exemplarily, the third preset rule can be configured to extract audio frames one by one, or to extract audio clips every other preset time length (e.g., 1s).

[0048] It is understandable that during extraction, information such as a number or a timestamp may be recorded for each basic unit extracted to record the sequence of different basic units, and to facilitate identification of adjacent basic units before and after different basic units when needed.

[0049] It can also be understood that when the audiovisual content contains multiple different types of data such as audio, video, etc., they can be processed separately, that is, the basic units are extracted to calculate the hash value, and the key units are signed and stored separately.

[0050] S102, calculating the hash value corresponding to the basic unit.

[0051] The original data used to calculate the hash value includes: the current basic unit, and the hash value corresponding to the previous basic unit of the basic unit.

[0052] For example, Figure 3 As shown, assuming that the acquisition unit acquires a segment of video data, the processing unit extracts the basic units in the segment of video data, for example, the currently extracted basic unit is D(i), and then the hash value H(i) corresponding to D(i) is calculated. The hash value H(i-1) corresponding to the previous basic unit D(i-1) of D(i) and D(i) are used as the original data, and a hash operation is performed to obtain the hash value H(i). That is, H(i) = Hash(D(i)||H(i-1)).

[0053] Then the processing unit can process the next extracted basic unit D(i+1) and calculate the hash value H(i+1)=Hash(D(i+1)||H(i)) corresponding to D(i+1). Similarly, the hash value corresponding to each basic unit can be calculated respectively.

[0054] It should be noted that, when the basic unit is the first basic unit D(0) in the audio-visual data, since there is no previous basic unit, a preset data can be used instead. That is, in this case, the original data used to calculate the hash value H(0) corresponding to the basic unit includes: the first basic unit D(0), and the preset initial value Initial_Value, H(0) = Hash(D(0)||Initial_Value).

[0055] Exemplarily, the preset initial value may be a fixed value, such as 0 or other fixed values, or may be assigned a variable value according to certain rules. The present application does not limit the specific value-taking method and value of the preset initial value.

[0056] S103: When the basic unit is a key unit, a signature corresponding to the key unit is calculated using a device private key.

[0057] It is understandable that the key unit itself is also a basic unit, which is selected from the basic units according to certain rules, and there are many ways to select it.

[0058] In some implementations, when the collected audiovisual data includes video data, a key video segment or key frame can be selected from the basic unit of the video data according to the second preset rule. Exemplarily, when the first preset rule is configured to extract the basic unit frame by frame, a key frame (Key Frame) of the video, such as an I frame (Intra Frame) can be used as a key unit.

[0059] In some implementations, when the collected audiovisual data includes audio data, a key audio segment or key frame may be selected from a basic unit of the audio data according to a fourth preset rule. Exemplarily, when the third preset rule is configured to extract an audio segment as a basic unit, one audio segment may be selected as a key audio segment every certain number (e.g., 5) of audio segments.

[0060] For a key unit, in addition to calculating its corresponding hash value, the device private key is also used to calculate its corresponding signature. The original data used for the signature includes the key unit and the hash value corresponding to the previous basic unit. For example, Figure 3 As shown, assuming that D(i) can be determined to be a key unit according to the second preset rule, the device private key SK is used to sign D(i) and H(i-1) to obtain the signature S(i). That is, S(i) = Sign_SK(D(i)||H(i-1)).

[0061] S104, storing the basic unit and the corresponding hash value, and storing the signature corresponding to the key unit when the basic unit is a key unit.

[0062] The hash value corresponding to the basic unit and the signature corresponding to the key unit can be stored in a variety of ways. This application does not limit the specific storage method. It only needs to be stored for use in subsequent verification.

[0063] In some implementations, the hash value corresponding to the basic unit may be embedded in the corresponding basic unit, or embedded in the metadata of the corresponding basic unit.

[0064] In some implementations, the hash values ​​corresponding to all basic units may also be stored in an independent file (referred to as the first file for easy distinction).

[0065] In some implementations, the signature corresponding to the key unit may be embedded in the corresponding key unit, or embedded in the metadata of the corresponding key unit.

[0066] In some implementations, the signatures corresponding to all key units may also be stored in an independent file (referred to as the second file for easy distinction). It is understandable that when all hash values ​​and all signatures are stored in independent files, the first file and the second file may be different files or the same file, and this application does not limit this.

[0067] It is understandable that the above steps S101 to S104 describe the processing process of the processing unit on one of the basic units, and the processing process on other basic units is similar. When the acquisition unit completes the acquisition of a whole segment of audio-visual data and the data is processed by the processing unit, a whole segment of complete audio-visual content is stored. For example, Figure 3 As shown, assuming that the entire video includes n basic units, m of which are key units, the entire audio-visual content includes n basic units, the hash values ​​corresponding to each of the n basic units, and the signatures corresponding to m of the key units.

[0068] Optionally, the collection device may store the audio-visual content collected in the above manner locally or in a possible location such as the cloud, or may directly or indirectly send the audio-visual content or a fragment thereof (for ease of distinction, the embodiment of the present application is referred to as the first fragment) and corresponding verification data (for ease of distinction, the embodiment of the present application is referred to as the first verification data). It is understandable that the first fragment includes at least two consecutive basic units, and these basic units include one or more key units.

[0069] It should be noted that when the first fragment includes multiple key units, the first verification data may include the signature of any one of the key units, or may include signatures corresponding to any several or all of the key units, and this application does not limit this.

[0070] In this way, the source of each key unit and its previous basic unit in the audio-visual content is guaranteed to be credible through the device private key signature, and the hash values ​​of the key unit and the previous basic unit are prevented from being forged or illegally tampered with. At the same time, by calculating the hash value corresponding to each basic unit, a chain relationship is established between adjacent basic units in the entire audio-visual content. Any tampering with a single basic unit will destroy the chain relationship, resulting in failure to pass the verification. Therefore, no matter whether the audio-visual content to be verified is a complete piece of content or a fragment of it, as long as it includes at least one key unit, it can be verified whether the source of the fragment is authentic, whether it has been tampered with or forged, and whether a frame or fragment has been deleted, and its authenticity and integrity can be better verified.

[0071] The following will take the verification process of a certain audio-visual content or a certain segment thereof as an example to further explain the processing process of the verification end and the functions that can be achieved. Figure 4 , Figure 4 This is a flow chart of an exemplary verification method performed by a verification terminal according to an embodiment of the present application. The method at least includes steps S201 to S204.

[0072] S201, obtaining a second segment to be verified in audio-visual content and corresponding second verification data.

[0073] The audiovisual content collected by the acquisition device may flow in different devices and / or networks. No matter how it flows, how many times it flows, and whether it is further edited during the flow (for example, new segments are cut out from it), it can be sent to the verification end to verify its authenticity and the integrity of the segment.

[0074] For the convenience of distinction and description, in the embodiment of the present application, the segment to be verified obtained by the verification end is referred to as the second segment, and the corresponding verification data is referred to as the second verification data. The second segment can be the entire audio-visual content, or a portion thereof. In some implementations, the second segment can be the aforementioned first segment or a portion thereof, or can be another segment in the audio-visual content that is different from the first segment.

[0075] The second segment includes at least two basic units, at least one of which is a key unit. The second verification data includes a hash value corresponding to each basic unit in the second segment, and a signature corresponding to at least one key unit in the second segment. For example, see Figure 5 , where the second fragment includes four basic units from D'(i-2) to D'(i+1), where D'(i) is the key unit, and its corresponding second verification data includes four hash values ​​from H(i-2) to H(i+1) and a signature S(i).

[0076] It should be noted that, similar to the first verification data, when the second segment includes multiple key units, the second verification data may include the signature of any one of the key units, or may include the signatures corresponding to any or all of the key units, and this application does not limit this. The second verification data may be embedded in the corresponding basic unit or its metadata, or may be stored separately in an independent file (such as the aforementioned first file and / or second file), and the aforementioned related description may be referred to, which will not be repeated here.

[0077] S202: Use the device public key of the acquisition device to verify the signature corresponding to at least one key unit.

[0078] When signatures corresponding to multiple key units are included, each of the signatures may be verified separately using the device public key of the acquisition device, or any one or several of the signatures may be verified.

[0079] Use the device public key of the acquisition device to verify the signature corresponding to a key unit, which corresponds to the signing process. Figure 5, take the verification of signature S(i) as an example, when signing, hash operation is performed with D(i) and H(i-1) as the original data, and encrypted with the device private key SK to obtain S(i). Correspondingly, when verifying the signature, the device private key PK can be used to decrypt S(i) to obtain the decrypted data; hash operation is performed with D'(i) and H(i-1) as the original data to obtain the operation data; the decrypted data and the operation data are compared, if the two are consistent, the signature S(i) is verified, if not, the signature S(i) fails to be verified. S(i) is verified, indicating that the key unit D'(i) in the second fragment to be verified is consistent with the corresponding key unit D(i) during the original collection and processing of the acquisition device, that is, D'(i) is indeed from the acquisition device and has not been modified during the circulation process, and is authentic and reliable, and the hash value H(i-1) in the second verification data is also authentic and reliable.

[0080] S203, verifying the hash value corresponding to each basic unit except the first basic unit in the second segment respectively.

[0081] For example, see Figure 5 , taking the verification of the hash value corresponding to D'(i) as an example, D'(i) and H(i-1) are used as the original data for hash operation to obtain H'(i). If H'(i) is consistent with H(i) in the second verification data, the hash value corresponding to the basic unit D'(i) is verified. This indicates that H(i) in the second verification data is authentic.

[0082] By analogy, we can perform hash operation on D'(i+1) and H(i) as the original data to obtain H'(i+1). If H'(i+1) is consistent with H(i+1) in the second verification data, the hash value corresponding to the basic unit D'(i) is verified. Since the previous verification can show that H(i) is authentic, it can be shown that D'(i+1) and H(i+1) are also authentic.

[0083] Similarly, we use D'(i-1) and H(i-2) as the original data for hash operation to obtain H'(i-1). If H'(i-1) is consistent with H(i-1) in the second verification data, the hash value H'(i-1) corresponding to the basic unit D'(i-1) passes the verification. Since the previous signature verification shows that the hash value H(i-1) is authentic, the verification shows that D'(i-1) and H(i-2) are authentic.

[0084] If the second segment includes other basic units before and after, the verification method can be deduced in the same manner as above.

[0085] In the application embodiment, there is no limitation on the order of verifying the signature and verifying each hash value.

[0086] S204: When the hash value and the signature are both verified, determine that the second segment is an audio-visual content segment originally captured by the capture device and has not been modified.

[0087] S205: If at least one of all the hash values ​​and the signatures fails to pass the verification, generate prompt information to prompt that the second segment is a non-original audio-visual content segment.

[0088] If a key unit of an audiovisual content segment has been modified, it cannot pass the corresponding signature verification. If a key unit is not collected by the original acquisition device, it cannot pass the corresponding signature verification. If a basic unit of a non-key unit has been modified, or an intermediate basic unit has been deleted, for example, the aforementioned D'(i-1) segments have been deleted, it cannot pass the corresponding hash value verification. It can be seen that the above-mentioned verification method can effectively detect the aforementioned non-authenticity or incompleteness of the audiovisual content or its segments, and the video segments generated by AI or modified by AI can naturally be detected. Therefore, the above-mentioned scheme can effectively help users distinguish the authenticity of audiovisual content.

[0089] The embodiment of the present application further provides a system, including any possible acquisition device as described above, and any possible verification terminal. The system can be applied in a variety of different scenarios.

[0090] For example, in the scenario of recording and publishing news media content, journalists can use the aforementioned camera (i.e., acquisition device) to record video and audio. After recording, the corresponding files can be directly uploaded to the content management system of the news organization. These contents can be published and disseminated in full or in fragments. The public and news organizations can confirm the authenticity and integrity of the video and audio through the verification terminal to ensure the credibility of the news content and prevent the spread of false information.

[0091] As another example, in the application scenario of intelligent traffic monitoring, traffic monitoring equipment (i.e., collection equipment) is deployed at traffic intersections to monitor and collect vehicle behavior and traffic violations. When traffic violation records need to be submitted to law enforcement agencies or for objection handling, they can be verified through the verification terminal to confirm the authenticity of the video and audio collected by the monitoring equipment to prevent the violation records from being maliciously tampered with.

[0092] In addition, the system can also be used in application scenarios such as remote examination invigilation, law enforcement or judicial on-site evidence collection, and remote meeting recording.

[0093] Those skilled in the art should understand that the methods performed by the processing unit and the verification end provided in the embodiments of the present application can also be provided as a computer-readable storage medium. It can be in the form of a complete software embodiment, or an embodiment combining software and hardware. That is, a computer-readable storage medium is also provided in the embodiments of the present application, storing a computer program, and when the computer program is executed by a processor, the processor executes some or all of the steps in any implementation of the aforementioned processing method or verification method.

[0094] The present application may also take the form of a computer program product embodied on one or more computer readable storage media having computer usable program code embodied therein.

[0095] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0096] It should be understood that in various embodiments of the present application, the execution order of each step should be determined by its function and internal logic, and the size of the step number does not mean the order of execution and does not constitute any limitation on the implementation process of the embodiment.

[0097] It should also be understood that, in order to facilitate the clear description of the technical solutions of the embodiments of the present application, the words "first", "second", etc. are used in the embodiments of the present application to distinguish between the same items or similar items with substantially the same functions and effects, or concepts with certain differences. Those skilled in the art can understand that the words "first", "second", etc. do not limit the quantity and execution order, and the words "first", "second", etc. do not necessarily limit the difference.

[0098] The same or similar parts between the various embodiments in this specification can be referred to each other. As long as there is no logical contradiction, different embodiments can be combined with each other.

Claims

1. An audio-visual content acquisition device, characterized in that: The acquisition device is pre-set with a device digital certificate and a corresponding device private key, and the device digital certificate is used to uniquely identify the identity of the acquisition device; the acquisition device includes an acquisition unit and a processing unit, wherein: The acquisition unit is used to collect audio-visual data; The processing unit is used for: Extracting basic units from the audio-visual data; Calculating a hash value corresponding to the basic unit, wherein the original data used to calculate the hash value includes: the basic unit, and the hash value corresponding to the previous basic unit of the basic unit; In the case where the basic unit is a key unit, using the device private key to calculate a signature corresponding to the key unit, wherein the original data used to calculate the signature includes the key unit and a hash value corresponding to a previous basic unit of the key unit; The basic unit and the corresponding hash value are stored, and when the basic unit is the key unit, the signature corresponding to the key unit is stored.

2. The acquisition device according to claim 1, characterized in that: The processing unit is also used for: embedding the hash value corresponding to the basic unit into the basic unit or the metadata of the basic unit; or, Embed the signature corresponding to the key unit into the basic unit or the metadata of the basic unit; or, The hash value corresponding to the basic unit and / or the signature corresponding to the key unit are stored in an independent file.

3. The acquisition device according to any one of claims 1 to 2, characterized in that: The processing unit is further configured to: when the basic unit is the first basic unit in the audio-visual data, the original data used to calculate the hash value includes: the first basic unit and a preset initial value.

4. The acquisition device according to any one of claims 1 to 3, characterized in that: In the case where the collected audio-visual data includes video data, the basic unit is a video segment or a video frame extracted according to a first preset rule, and the key unit is a key video segment or a key frame selected from the video according to a second preset rule; and / or, When the collected audio-visual data includes audio data, the basic unit is an audio segment or an audio frame extracted according to a third preset rule; and the key unit is a key audio segment or a key frame selected from the audio according to a fourth preset rule.

5. The acquisition device according to any one of claims 1 to 4, characterized in that: The acquisition device includes a security chip, and the device private key is stored in the security chip.

6. The acquisition device according to any one of claims 1 to 5, characterized in that: The acquisition device is also used for: Dividing at least one first segment from the stored audiovisual content, each of the first segments includes at least two consecutive basic units, and the at least two basic units include at least one key unit; The at least one first fragment and corresponding first verification data are sent, wherein the first verification data includes hash values ​​corresponding to each of the at least two basic units and a signature corresponding to at least one key unit.

7. A verification terminal for audio-visual content, characterized in that: The verification end is used for: Acquire a second segment to be verified in the audiovisual content and corresponding second verification data, wherein the second segment includes at least two basic units, the at least two basic units include at least one key unit, and the second verification data includes hash values ​​corresponding to the at least two basic units, and a signature corresponding to the at least one key unit; Verify the signature corresponding to at least one key unit using the device public key of the acquisition device; Verify the hash value corresponding to each basic unit except the first basic unit in the second fragment respectively; When the verification of all hash values ​​and signatures is passed, it is determined that the second segment is an audio-visual content segment originally captured by the capture device and has not been modified.

8. The verification terminal according to claim 7, characterized in that: The verification end is also used for: When at least one of all hash values ​​and the signature fails to pass verification, prompt information is generated to indicate that the second segment is a non-original audio-visual content segment.

9. The verification terminal according to claim 7, characterized in that: The verification end is a server or a client.

10. An audio-visual content collection and verification system, comprising the collection device according to any one of claims 1 to 6, and the verification terminal according to any one of claims 7 to 9.

Images