Video intelligent analysis method, system and device applied to unmanned aerial vehicle and medium

Abstract

The embodiment of the invention provides a video intelligent analysis method and system applied to an unmanned aerial vehicle, and the method comprises the steps: obtaining video stream data and an intelligent analysis instruction; calling an AI algorithm tool according to the intelligent analysis instruction, and performing intelligent analysis processing on the video stream data; and outputting and displaying the intelligently analyzed and processed video stream data. According to the embodiment of the invention, by calling the AI algorithm SDK and carrying out real-time analysis and target identification box labeling on the video stream according to the intelligent analysis instruction, the intelligent processing capability of the video content is remarkably improved, and the demand of a user for dynamic and accurate video analysis is met. Besides, the processed video stream data can be visually displayed, so that the user experience is enhanced, and the video analysis function is more practical and efficient in various application scenes.

Description

Technical Field

[0001] This invention relates to the field of video processing technology, and in particular to a video intelligent analysis method for unmanned aerial vehicles (UAVs), a video intelligent analysis system for UAVs, an electronic device, and a computer-readable storage medium. Background Technology

[0002] Existing streaming media service systems primarily rely on traditional video capture and transmission technologies to process video data, enabling the reception, distribution, and display of video streams. However, existing systems have significant technical limitations when dealing with video streams captured by drones.

[0003] First, streaming media service systems cannot directly receive and process video stream data from drones, requiring additional protocol conversion and data forwarding mechanisms to access specific network environments. Second, existing systems lack intelligent analysis capabilities for drone video streams, and cannot perform real-time processing and target identification of video content according to specific analytical needs. Summary of the Invention

[0004] In view of the above problems, embodiments of the present invention are proposed to provide a video intelligent analysis method for unmanned aerial vehicles (UAVs), a video intelligent analysis system for UAVs, an electronic device, and a computer-readable storage medium to overcome or at least partially solve the above problems.

[0005] To address the aforementioned problems, this invention discloses a video intelligent analysis method, the method comprising:

[0006] Acquire video stream data and intelligent analysis commands;

[0007] The AI ​​algorithm tool is invoked according to the intelligent analysis command to perform intelligent analysis and processing on the video stream data;

[0008] Output and display the video stream data after intelligent analysis and processing.

[0009] Optionally, the step of invoking AI algorithm tools according to the intelligent analysis instruction to perform intelligent analysis and processing on the video stream data includes:

[0010] Select the AI ​​algorithm model corresponding to the intelligent analysis command from the AI ​​algorithm tool;

[0011] The AI ​​algorithm model is used to detect target objects in the video stream data and generate analysis results containing target bounding boxes. The target objects include at least one of the following: people, vehicles, ships, crowds, or fireworks.

[0012] Optionally, the step of using the AI ​​algorithm model to detect target objects in the video stream data and generating analysis results containing target bounding boxes includes:

[0013] The AI ​​algorithm model is used to execute a target detection algorithm to identify the target object specified by the intelligent analysis command.

[0014] Draw the target identification box in the video stream data frame, and use the target identification box to mark the target object.

[0015] Optionally, the intelligent analysis command includes at least one of the following: a person / vehicle detection command, a person / boat detection command, a crowd gathering detection command, or a smoke and fire detection command.

[0016] Optionally, the output and display of the intelligently analyzed video stream data includes:

[0017] The video stream data containing the target identifier frame is sent to the terminal device;

[0018] The video stream data containing the target identifier frame is displayed on the terminal device.

[0019] Optionally, the method further includes:

[0020] Real-time monitoring of the transmission status of the video stream data;

[0021] When the transmission status indicates an abnormal data stream, an abnormal prompt message is generated and displayed.

[0022] Optionally, after monitoring the transmission status of the video stream data in real time, the method further includes:

[0023] When the transmission status indicates that the data stream has returned to normal, the intelligent analysis and processing of the video stream data will automatically resume.

[0024] Output and display the results of the intelligent analysis and processing after recovery.

[0025] This invention also discloses a video intelligent analysis system, the system comprising:

[0026] The video instruction acquisition module is used to acquire video stream data and intelligent analysis instructions;

[0027] The intelligent analysis and processing module is used to call AI algorithm tools according to the intelligent analysis instructions to perform intelligent analysis and processing on the video stream data;

[0028] The data output and display module is used to output and display the video stream data after intelligent analysis and processing.

[0029] Optionally, the intelligent analysis and processing module includes:

[0030] An algorithm model selection module is used to select an AI algorithm model from the AI ​​algorithm tool that corresponds to the intelligent analysis instruction;

[0031] The target object detection module is used to detect target objects in the video stream data using the AI ​​algorithm model and generate analysis results containing target bounding boxes. The target objects include at least one of the following: people, vehicles, ships, crowds, or fireworks.

[0032] Optionally, the target object detection module includes:

[0033] The detection algorithm execution module is used to execute the target detection algorithm through the AI ​​algorithm model to identify the target object specified by the intelligent analysis instruction;

[0034] The marker box drawing module is used to draw the target marker box in the video stream data and to mark the target object using the target marker box.

[0035] Optionally, the intelligent analysis command includes at least one of the following: a person / vehicle detection command, a person / boat detection command, a crowd gathering detection command, or a smoke and fire detection command.

[0036] Optionally, the data output display module includes:

[0037] The video stream sending module is used to send video stream data containing the target identifier frame to the terminal device;

[0038] The video stream display module is used to display video stream data containing the target identifier box on the terminal device.

[0039] Optionally, the system further includes:

[0040] A transmission status monitoring module is used to monitor the transmission status of the video stream data in real time;

[0041] An error message generation module is used to generate and display error message information when the transmission status indicates that the data stream is abnormal.

[0042] Optionally, the system further includes:

[0043] An automatic recovery module is used to automatically resume intelligent analysis and processing of the video stream data after the transmission status monitoring module monitors the transmission status of the video stream data in real time and the transmission status indicates that the data stream has returned to normal.

[0044] The data output and display module is also used to output and display the intelligent analysis and processing results after recovery.

[0045] This invention also discloses an electronic device, comprising: one or more processors; and one or more machine-readable media storing instructions thereon, which, when executed by the one or more processors, cause the electronic device to perform the video intelligent analysis method as described above.

[0046] This invention also discloses a computer-readable storage medium storing a computer program that causes a processor to execute the video intelligent analysis method described above.

[0047] The embodiments of the present invention have the following advantages:

[0048] The video intelligent analysis solution for drones provided in this embodiment of the invention acquires video stream data and intelligent analysis instructions; calls AI algorithm tools according to the intelligent analysis instructions to perform intelligent analysis and processing on the video stream data; and outputs and displays the intelligently analyzed and processed video stream data.

[0049] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0050] This invention significantly improves the intelligent processing capabilities of video content by invoking AI algorithm tools and performing real-time analysis and target bounding box annotation on video streams according to intelligent analysis instructions, thus meeting users' needs for dynamic and accurate video analysis. Furthermore, the processed video stream data can be displayed intuitively, enhancing the user experience and making the video analysis function more practical and efficient in various application scenarios. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating the steps of a video intelligent analysis method applied to unmanned aerial vehicles (UAVs) according to an embodiment of the present invention.

[0052] Figure 2 This is a flowchart illustrating a UAV video intelligent analysis scheme based on the visual network according to an embodiment of the present invention;

[0053] Figure 3 This is a structural block diagram of a video intelligent analysis system applied to unmanned aerial vehicles (UAVs) according to an embodiment of the present invention. Detailed Implementation

[0054] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0055] This invention provides a video intelligent analysis solution for unmanned aerial vehicles (UAVs). By acquiring video stream data and intelligent analysis commands, it invokes AI algorithm tools to select the corresponding algorithm model, performs real-time detection and bounding box annotation of target objects (such as people and vehicles) in the video stream data, and outputs the processed video stream data to a terminal for intuitive display. This video intelligent analysis solution includes a flight control platform, a cooperative transfer service system, and a streaming media service system. The flight control platform issues commands, the cooperative transfer service system encapsulates and forwards the data, and the streaming media service system performs intelligent analysis and outputs the results. This invention improves the intelligence level of video processing and user experience, and is suitable for various video application scenarios requiring dynamic analysis.

[0056] Reference Figure 1 This document illustrates a flowchart of a video intelligent analysis method for unmanned aerial vehicles (UAVs) according to an embodiment of the present invention. This video intelligent analysis method for UAVs can be applied to systems such as video intelligent analysis systems and video processing systems (hereinafter referred to as systems). Specifically, the video intelligent analysis method for UAVs may include the following steps:

[0057] Step 101: Obtain video stream data and intelligent analysis instructions.

[0058] Video stream data refers to continuous video data generated in real time by video acquisition devices (such as cameras), typically transmitted as an IP stream, containing dynamic image information. Intelligent analysis commands are instructions issued by users or the system through the flight control platform, specifying the analysis tasks to be performed, such as human / vehicle detection, human / ship detection, crowd gathering detection, or smoke / fire detection. These commands clearly define the analysis objectives and processing methods, ensuring targeted subsequent processing. In implementation, video stream data and intelligent analysis commands are received and processed through a protocol conversion service system. This system encapsulates the received video stream data and intelligent analysis commands into a video network protocol format to adapt to the transmission requirements of the video network. The encapsulation process includes format conversion of the video stream data and protocol packaging of the intelligent analysis commands, ensuring efficient transmission of data and commands within the video network. Furthermore, the protocol conversion service system performs preliminary verification of the data stream to ensure the integrity and undamaged nature of the received video stream data, while also checking the legality of the intelligent analysis commands (e.g., whether the command format conforms to predefined standards).

[0059] Step 102: Call the AI ​​algorithm tool according to the intelligent analysis instruction to perform intelligent analysis and processing on the video stream data.

[0060] The system utilizes an Artificial Intelligence Algorithm Software Development Kit (AI SDK) to intelligently analyze and process video stream data, enabling the detection and labeling of target objects. The AI ​​SDK is a software framework integrating multiple artificial intelligence algorithm models, including pre-trained models for object detection, classification, and tracking. It can dynamically select the appropriate algorithm model based on intelligent analysis instructions. For example, when the intelligent analysis instruction specifies human and vehicle detection, the AI ​​SDK will call a pre-trained object detection model to identify people and vehicles in the video stream data; if the instruction is for crowd detection, the corresponding crowd analysis model will be selected. The processing first parses the intelligent analysis instruction, extracting the analysis task type and target object category specified in the instruction. Then, it selects an algorithm model matching the task from the AI ​​SDK's model library. Next, the system inputs the video stream data into the selected algorithm model. The model performs feature extraction and object detection on each frame of video data, identifying the target object specified in the instruction (such as people, vehicles, ships, crowds, or fireworks). After detection, the model draws bounding boxes on the video frames to label the location and category of the target objects, and generates analysis results containing these bounding boxes. The analysis results are stored as a video stream, preserving the temporal sequence characteristics of the original video while embedding the target identification information.

[0061] Step 103: Output and display the video stream data after intelligent analysis and processing.

[0062] The processed video stream data is output and displayed on the terminal to provide users with an intuitive understanding of the analysis results. The processed video stream data includes the original video content and target bounding boxes generated by the AI ​​algorithm SDK. These bounding boxes visually label target objects (such as people, vehicles, etc.) in the video and include category information. The output process first transmits the processed video stream data to the video network terminal via the video network protocol. During transmission, the system ensures the integrity and real-time performance of the video stream data, using a dedicated transmission channel of the video network to reduce latency and guarantee data security. After receiving the video stream data, the video network terminal converts the data into a displayable video image through the decoding and rendering modules and displays it on the terminal screen. In the displayed image, target bounding boxes are marked around the target object with a prominent color and shape (such as a rectangle), and the box may contain labels (such as "people" or "vehicles") to clearly indicate the category. Users can intuitively observe the dynamic analysis results in the video through the terminal interface, such as the distribution of crowd gathering areas or the location of fireworks events.

[0063] This invention significantly improves the intelligent processing capabilities of video content by calling an AI algorithm SDK and performing real-time analysis and target bounding box annotation on the video stream according to intelligent analysis instructions, thus meeting users' needs for dynamic and accurate video analysis. Furthermore, the processed video stream data can be displayed intuitively, enhancing the user experience and making the video analysis function more practical and efficient in various application scenarios.

[0064] In one exemplary embodiment of the present invention, one implementation method for intelligently analyzing and processing video stream data by calling an AI algorithm tool according to an intelligent analysis instruction is as follows: selecting an AI algorithm model corresponding to the intelligent analysis instruction from the AI ​​algorithm tool; using the AI ​​algorithm model to detect target objects in the video stream data and generating an analysis result containing target bounding boxes, wherein the target objects include at least one of the following: people, vehicles, ships, crowds, or fireworks.

[0065] The system parses the received intelligent analysis instructions, extracting the analysis task type and target object category. Based on this, the system dynamically selects an AI algorithm model matching the instruction from AI algorithm tools (e.g., an AI algorithm SDK). For example, if the instruction is "human and vehicle detection," a pre-trained target detection model is selected; if the instruction is "crowd gathering detection," a specialized group behavior analysis model is invoked. The AI ​​algorithm SDK, as an integrated framework, contains various pre-trained models and calling interfaces, enabling flexible adaptation to different tasks based on instructions. After selecting a model, the system inputs video stream data into the model, extracts features frame by frame, and identifies the target object specified in the instruction. The detection process uses deep learning algorithms to perform convolution operations and classification on video frames, determining the location and category of the target object, and generating target bounding boxes at the corresponding locations. Labels can be attached to the boxes for clear identification. The analysis results are retained in the form of a video stream, embedding target bounding box information to ensure the continuity of subsequent outputs and visualization effects.

[0066] This implementation method can accurately select AI algorithm models and efficiently detect target objects, solving the problem of traditional systems lacking intelligent analysis capabilities. It significantly improves the targeting and accuracy of video processing, and the generated identification boxes intuitively present target information, enhancing users' understanding of video content and its application value.

[0067] In one exemplary embodiment of the present invention, an implementation method for detecting target objects in video stream data using an AI algorithm model and generating analysis results containing target bounding boxes is as follows: executing a target detection algorithm through an AI algorithm model to identify the target object specified by the intelligent analysis instruction; drawing target bounding boxes in the video stream data frame and labeling the target objects using the target bounding boxes.

[0068] During execution, the system first runs an object detection algorithm using a selected AI algorithm model to identify the target object specified by the intelligent analysis command. The object detection algorithm extracts and classifies features from each frame of the video stream data, analyzing pixel distribution, edge features, and texture information to determine the location and category of the target object. After recognition, the system draws target bounding boxes on the video frames. These boxes surround the target object in rectangular form, and usually include a label to clarify the category. The color and style of the boxes can be configured according to preset rules to ensure visual distinguishability. The drawing process uses graphics rendering technology to embed the bounding boxes into the video stream data, maintaining the real-time nature and continuity of the video, while ensuring that the bounding boxes are updated synchronously with the dynamic movement of the target object.

[0069] By employing deep learning algorithms and dynamic bounding box drawing, this system overcomes the limitations of traditional video processing in real-time analysis and annotation of targets, significantly improving the semantic expression capabilities of video content, providing users with clear and intuitive analysis results, and enhancing the system's practicality and interactivity.

[0070] In one exemplary embodiment of the present invention, the intelligent analysis command includes at least one of the following: a person and vehicle detection command, a person and boat detection command, a crowd gathering detection command, or a fire detection command.

[0071] In practice, intelligent analysis commands are generated and issued by the flight control platform, encapsulated into a video network protocol format by the cooperative transfer service system, and then transmitted to the streaming media service system. For example, the personnel and vessel detection command detects vessels and related personnel, suitable for waterway monitoring scenarios; the crowd gathering detection command analyzes whether there are densely populated areas in the video, commonly used in public safety monitoring; and the smoke and fire detection command focuses on detecting fire or smoke events, suitable for fire early warning scenarios. Each command includes information on the task type, target object category, and expected output format.

[0072] This implementation provides diverse intelligent analysis commands, enabling flexible selection of analysis tasks to meet different scenario requirements. It overcomes the limitations of traditional systems with limited functionality, significantly improves the adaptability and application scope of video intelligent analysis, provides users with diverse monitoring and analysis capabilities, and enhances the practical value of the system.

[0073] In one exemplary embodiment of the present invention, one way to output and display the intelligently analyzed video stream data is as follows: sending the video stream data containing the target identifier box to the terminal device; and displaying the video stream data containing the target identifier box on the terminal device.

[0074] In this process, video stream data processed by the AI ​​algorithm SDK is transmitted to the video network terminal via the video network protocol. These video stream data embed target bounding boxes generated by the AI ​​algorithm model to label target objects specified by intelligent analysis commands. The bounding boxes are presented in rectangular form and accompanied by labels to clearly identify the object category. During transmission, the system utilizes the dedicated channel of the video network to ensure high efficiency and low latency, and employs data compression and encryption technologies to guarantee the integrity and security of transmission. After receiving the video stream data, the video network terminal decodes the data into displayable video frames, which are then displayed on the terminal screen via the rendering module. In the displayed image, the target bounding boxes are presented with striking colors and dynamically follow the target objects, ensuring that users can intuitively identify key information in the video.

[0075] This implementation overcomes the limitations of traditional video processing systems that only provide raw images and lack semantic annotation by efficiently transmitting and intuitively displaying video stream data containing target bounding boxes. It significantly improves the efficiency of users' understanding of video content and interactive experience, and provides intuitive and practical analysis results for scenarios such as monitoring and security, thereby enhancing the application value of the system.

[0076] In one exemplary embodiment of the present invention, the transmission status of video stream data is monitored in real time; when the transmission status indicates that the data stream is abnormal, an abnormal prompt message is generated and displayed.

[0077] Specifically, a data stream monitoring module integrated into the streaming media service system continuously tracks the transmission status of video stream data from the transmission service system to the video network terminal. The monitoring process involves real-time analysis of the integrity, continuity, and transmission latency of the video stream data encapsulated by the video network video protocol. For example, it checks for packet loss, video frame interruptions, or latency exceeding preset thresholds. Furthermore, monitoring also covers the transmission status of intelligent analysis commands, ensuring that commands are transmitted synchronously with the video stream data and preventing AI algorithm SDK processing errors due to command loss. When a data stream anomaly is detected (such as network interruption, packet loss, or frame rate anomaly), an anomaly alert is immediately generated, including the anomaly type, occurrence time, and suggested handling measures. These alerts are displayed on the video network terminal in the form of pop-ups, text, or icons, ensuring users are promptly informed of the problem.

[0078] This implementation effectively solves the problem of data loss or interruption caused by unstable transmission in traditional video processing systems through real-time monitoring and anomaly alert mechanisms. It significantly improves the reliability of video stream data transmission and user experience, ensures that intelligent analysis results can be continuously and stably output and displayed, and enhances the robustness and practicality of the system in complex network environments.

[0079] In one exemplary embodiment of the present invention, one implementation method after real-time monitoring of the transmission status of video stream data is as follows: when the transmission status indicates that the data stream has returned to normal, the intelligent analysis and processing of the video stream data is automatically resumed; the intelligent analysis and processing results after the recovery are output and displayed.

[0080] Specifically, once the system detects that the video stream data transmission status has returned to normal, it automatically triggers a recovery mechanism. First, the system verifies the integrity and continuity of the video stream data, such as checking whether the received video frame sequence is complete and whether there are any missing or distorted frames, to ensure the data is suitable for subsequent processing. After successful verification, the system reactivates the AI ​​algorithm SDK and calls the corresponding AI algorithm model based on the previously received intelligent analysis instructions to perform intelligent analysis and processing on the recovered video stream data. The analysis process is consistent with the normal workflow, involving the detection of target objects and the drawing of target bounding boxes, generating analysis results containing the bounding boxes. Subsequently, the system transmits the processed video stream data to the video network terminal via the video network protocol and displays it on the terminal screen through the decoding and rendering modules, presenting a video image containing the target bounding boxes.

[0081] This implementation method effectively solves the limitation of difficulty in quickly recovering intelligent analysis after data stream anomalies through automatic recovery mechanism and result display. It significantly improves the continuity and robustness of the system, ensures that users can seamlessly obtain accurate analysis results after the anomaly is resolved, and enhances the reliability and application value of video intelligent analysis in dynamic network environments.

[0082] Based on the above description of an embodiment of a video intelligent analysis method, a UAV video intelligent analysis scheme based on the visual network is introduced below. (Refer to...) Figure 2 This diagram illustrates a flowchart of a UAV video intelligent analysis scheme based on a visual network, according to an embodiment of the present invention. The scheme involves a flight control platform, a UAV, an assistance service system, a streaming media service system, and a terminal.

[0083] In practical applications, operators issue drone operation commands to the drone via the flight control platform. Upon receiving the commands, the drone takes off smoothly and adjusts its flight path according to the in-flight instructions. Simultaneously, it uses cameras to capture real-time video footage, generating a video IP stream. The flight control platform also sends intelligent analysis commands to the assistance service system, such as "person and vehicle detection commands," requiring the detection of people and vehicles in the video IP stream. The video IP stream is transmitted to the forwarding service system, which forwards it to the streaming media service system, along with the intelligent analysis commands. The forwarding service system encapsulates the video IP stream and intelligent analysis commands into a video network video protocol format (video V2V stream) to ensure data compatibility within the video network.

[0084] Next, after receiving the video V2V stream and intelligent analysis instructions, the streaming media service system invokes the integrated AI algorithm SDK to activate the intelligent analysis function. The AI ​​algorithm SDK selects a suitable AI model based on the intelligent analysis instructions, performs real-time analysis of the video stream data, identifies people and vehicles in the scene, and draws target bounding boxes around the detected objects. The analysis results are output as a video stream containing the target bounding boxes. The streaming media service system then transmits the intelligently analyzed video data to the video network terminal via the video network. After receiving the data, the terminal displays the analysis results on the screen, showing people and vehicles marked with rectangular boxes, with labels inside the boxes to clearly indicate the category.

[0085] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0086] Reference Figure 3 The diagram illustrates a structural block diagram of a video intelligent analysis system for unmanned aerial vehicles (UAVs) according to an embodiment of the present invention. This video intelligent analysis system for UAVs may specifically include the following modules.

[0087] The video instruction acquisition module 31 is used to acquire video stream data and intelligent analysis instructions;

[0088] The intelligent analysis and processing module 32 is used to call AI algorithm tools according to the intelligent analysis instructions to perform intelligent analysis and processing on the video stream data;

[0089] The data output and display module 33 is used to output and display the video stream data after intelligent analysis and processing.

[0090] In an exemplary embodiment of the present invention, the intelligent analysis and processing module 32 includes:

[0091] An algorithm model selection module is used to select an AI algorithm model from the AI ​​algorithm tool that corresponds to the intelligent analysis instruction;

[0092] The target object detection module is used to detect target objects in the video stream data using the AI ​​algorithm model and generate analysis results containing target bounding boxes. The target objects include at least one of the following: people, vehicles, ships, crowds, or fireworks.

[0093] In an exemplary embodiment of the present invention, the target object detection module includes:

[0094] The detection algorithm execution module is used to execute the target detection algorithm through the AI ​​algorithm model to identify the target object specified by the intelligent analysis instruction;

[0095] The marker box drawing module is used to draw the target marker box in the video stream data and to mark the target object using the target marker box.

[0096] In one exemplary embodiment of the present invention, the intelligent analysis command includes at least one of the following: a person and vehicle detection command, a person and boat detection command, a crowd gathering detection command, or a fire detection command.

[0097] In an exemplary embodiment of the present invention, the data output display module 33 includes:

[0098] The video stream sending module is used to send video stream data containing the target identifier frame to the terminal device;

[0099] The video stream display module is used to display video stream data containing the target identifier box on the terminal device.

[0100] In one exemplary embodiment of the present invention, the system further includes:

[0101] A transmission status monitoring module is used to monitor the transmission status of the video stream data in real time;

[0102] An error message generation module is used to generate and display error message information when the transmission status indicates that the data stream is abnormal.

[0103] In one exemplary embodiment of the present invention, the system further includes:

[0104] An automatic recovery module is used to automatically resume intelligent analysis and processing of the video stream data after the transmission status monitoring module monitors the transmission status of the video stream data in real time and the transmission status indicates that the data stream has returned to normal.

[0105] The data output and display module 33 is also used to output and display the intelligent analysis and processing results after recovery.

[0106] As the system implementation is basically similar to the method implementation, it is described in a relatively simple way. For relevant details, please refer to the description of the method implementation.

[0107] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0108] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, embodiments of the present invention can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0109] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0110] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0111] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0112] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0113] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0114] The foregoing has provided a detailed description of a video intelligent analysis method and a video intelligent analysis system for unmanned aerial vehicles (UAVs) provided by this invention. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and core ideas of this invention. Furthermore, those skilled in the art will recognize that, based on the ideas of this invention, there will be changes in the specific implementation methods and application scope.

[0115] In conclusion, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A video intelligent analysis method, characterized in that, The method includes: Acquire video stream data and intelligent analysis commands; The AI ​​algorithm tool is invoked according to the intelligent analysis command to perform intelligent analysis and processing on the video stream data; Output and display the video stream data after intelligent analysis and processing.

2. The method according to claim 1, characterized in that, The step of calling AI algorithm tools according to the intelligent analysis instructions to perform intelligent analysis and processing on the video stream data includes: Select the AI ​​algorithm model corresponding to the intelligent analysis command from the AI ​​algorithm tool; The AI ​​algorithm model is used to detect target objects in the video stream data and generate analysis results containing target bounding boxes. The target objects include at least one of the following: people, vehicles, ships, crowds, or fireworks.

3. The method according to claim 2, characterized in that, The step of using the AI ​​algorithm model to detect target objects in the video stream data and generating analysis results containing target bounding boxes includes: The AI ​​algorithm model is used to execute a target detection algorithm to identify the target object specified by the intelligent analysis command. Draw the target identification box in the video stream data frame, and use the target identification box to mark the target object.

4. The method according to claim 1, characterized in that, The intelligent analysis commands include at least one of the following: human and vehicle detection commands, human and vessel detection commands, crowd gathering detection commands, or smoke and fire detection commands.

5. The method according to claim 2, characterized in that, The output and display of the intelligently analyzed and processed video stream data includes: The video stream data containing the target identifier frame is sent to the terminal device; The video stream data containing the target identifier frame is displayed on the terminal device.

6. The method according to claim 1, characterized in that, The method further includes: Real-time monitoring of the transmission status of the video stream data; When the transmission status indicates an abnormal data stream, an abnormal prompt message is generated and displayed.

7. The method according to claim 6, characterized in that, After monitoring the transmission status of the video stream data in real time, the method further includes: When the transmission status indicates that the data stream has returned to normal, the intelligent analysis and processing of the video stream data will automatically resume. Output and display the results of the intelligent analysis and processing after recovery.

8. A video intelligent analysis system, characterized in that, The system includes: The video instruction acquisition module is used to acquire video stream data and intelligent analysis instructions; The intelligent analysis and processing module is used to call AI algorithm tools according to the intelligent analysis instructions to perform intelligent analysis and processing on the video stream data; The data output and display module is used to output and display the video stream data after intelligent analysis and processing.

9. An electronic device, characterized in that, include: One or more processors; and One or more machine-readable media having instructions stored thereon, which, when executed by the one or more processors, cause the electronic device to perform the video intelligent analysis method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The stored computer program causes the processor to execute the video intelligent analysis method as described in any one of claims 1 to 7.

Images