Digital human service intelligent processing system for intelligent police service

Abstract

The embodiment of the invention discloses a digital human service intelligent processing system for intelligent police affairs. According to the technical scheme provided by the embodiment of the invention, the on-site alarm condition is accurately identified by combining the audio and video data of the monitoring area site, the alarm condition report is realized by interacting with the associated camera and the associated police officer terminal of the monitoring site, the police officer is timely notified to process the alarm condition service, the police affair processing efficiency is improved, and the police officer experience is improved. While the patrol deployment and control pressure of the police officers is relieved, the police officers are efficiently assisted in handling alarm business.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of data processing technology, and in particular to a digital human business intelligent processing system for smart policing. Background Art

[0002] At present, with the rapid development of artificial intelligence science, digital people using artificial intelligence technology have also been widely used. For example, in the police system, digital robots are set up to perform patrol and control operations to assist in maintaining social order and facilitate the processing of corresponding police operations. For example, digital robots can perform operations such as identity recognition of corresponding targets to assist patrols based on the relevant information identified.

[0003] However, when patrolling, digital robots generally only identify the target, and their functions are relatively simple. They lack sufficient recognition capabilities for complex police scenarios (such as abnormal behavior), which provides limited help to police officers and makes it difficult to improve police processing efficiency. Summary of the invention

[0004] The embodiment of the present application provides a digital human business intelligent processing system for smart policing, which can accurately identify on-site police situations by combining audio and video data at the scene of the control area, and realize police situation reporting by interacting with the associated cameras and associated police officer terminals at the control site, so as to promptly notify police officers to handle police situation business, improve police processing efficiency, and effectively assist police officers in handling police situation business while reducing the patrol and control pressure of police officers.

[0005] In a first aspect, an embodiment of the present application provides a digital human business intelligent processing system for smart policing, including: A collection module is used to collect audio and video data of the control site corresponding to the set control area, and extract target behavior data, target audio data and target face data based on the audio and video data of the control site; A warning recognition module is used to traverse the set warning feature library based on the target behavior data, target audio data and target face data, and when the corresponding warning feature is traversed, output a warning report of the current control area according to the traversed warning feature, and report the warning report to the warning system; The alarm business processing module is used to call the associated camera of the controlled area based on the alarm report, track the alarm based on the associated camera, and obtain real-time tracking video; obtain police officer terminals within a set range, select at least one target terminal from the police officer terminals, report the alarm report and the tracking video to the target terminal, so as to provide alarm prompts to the corresponding police officer; after a set time period, query the location distance between the target terminal and the controlled area, and forward the alarm report according to the alarm type of the alarm report when it is detected that the location distance exceeds the set distance threshold.

[0006] Further, the selecting at least one target terminal from the police officer terminals includes: Determine the alarm level of the alarm report, and determine the corresponding number of processing personnel according to the alarm level; Taking the control area as the center, a corresponding number of target terminals are selected from the police officer terminals from near to far according to the number of the processing personnel.

[0007] Further, the determining the alarm level of the alarm report includes: A preset alarm score is queried according to the traversed alarm characteristics, and the alarm level of the alarm report is determined according to the alarm score.

[0008] Further, forwarding the alarm report according to the alarm type of the alarm report includes: According to the alarm type of the alarm report, a background dispatch center or the nearest police officer terminal is selected to forward the alarm report.

[0009] Furthermore, the extracting of target behavior data, target audio data and target face data based on the control site audio and video data includes: The control site audio and video data are input into a pre-constructed composite model, and target behavior data, target audio data and target face data are output based on the composite model. The composite model pre-constructs training samples based on historical control site audio and video, and model training is performed based on the training samples.

[0010] Furthermore, when the corresponding alarm feature is traversed, it also includes: The audio and video data of the control site are written into an optimized sample set, and when the number of samples in the optimized sample set reaches a set threshold, optimization training of the composite model is performed based on the optimized sample set.

[0011] Furthermore, after outputting the alarm report of the current control area according to the traversed alarm features, it also includes: The historical police processing data is queried according to the traversed police situation characteristics, the processing experience information is extracted from the historical police processing data, and the processing experience information is written into the police situation report.

[0012] In a second aspect, an embodiment of the present application provides a method for intelligent processing of digital human services for smart policing, which is applied to the intelligent processing system for digital human services for smart policing as described in the first aspect, including: Collecting the audio and video data of the control scene corresponding to the set control area, and extracting the target behavior data, target audio data and target face data based on the audio and video data of the control scene; Based on the target behavior data, the target audio data and the target face data, the set alarm feature library is traversed, and when the corresponding alarm feature is traversed, the alarm report of the current control area is output according to the traversed alarm feature, and the alarm report is reported to the alarm system; Based on the alarm report, the associated camera of the controlled area is called, and the alarm tracking is performed based on the associated camera to obtain a real-time tracking video; the police officer terminals within a set range are obtained, at least one target terminal is selected from the police officer terminals, and the alarm report and the tracking video are reported to the target terminal to provide an alarm prompt to the corresponding police officer; after a set period of time, the position distance between the target terminal and the controlled area is queried, and when it is detected that the position distance exceeds a set distance threshold, the alarm report is forwarded according to the alarm type of the alarm report.

[0013] In a third aspect, an embodiment of the present application provides an electronic device, including: memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the digital human business intelligent processing method for smart policing as described in the first aspect.

[0014] In a fourth aspect, an embodiment of the present application provides a storage medium comprising computer executable instructions, which, when executed by a computer processor, are used to execute the digital human business intelligent processing method for smart policing as described in the first aspect.

[0015] The embodiment of the present application collects audio and video data of the control site through the corresponding set control area, and extracts target behavior data, target audio data and target face data based on the audio and video data of the control site; traverses the set alarm feature library based on the target behavior data, target audio data and target face data, and when the corresponding alarm feature is traversed, outputs the alarm report of the current control area according to the traversed alarm feature, and reports the alarm report to the alarm system; calls the associated camera of the control area based on the alarm report, tracks the alarm based on the associated camera, and obtains real-time tracking video; obtains the police officer terminals within the set range, selects at least one target terminal from the police officer terminals, and uploads the alarm report and tracking video to the target terminal to provide the corresponding police officer with an alarm prompt; after the set time period, queries the location distance between the target terminal and the control area, and when it is detected that the location distance exceeds the set distance threshold, forwards the alarm report according to the alarm type of the alarm report. By adopting the above-mentioned technical means, it is possible to accurately identify on-site police situations by combining the audio and video data of the controlled area, and to realize police situation reports by interacting with the associated cameras and associated police officer terminals at the controlled area, so as to promptly notify police officers to handle police situations and improve police handling efficiency. While reducing the pressure of patrol and control on police officers, it also effectively assists police officers in handling police situations. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of a method for intelligent processing of digital human services for smart policing provided in Example 1 of the present application; Figure 2 This is a flow chart for identifying the audio and video data of the control site in Example 1 of the present application; Figure 3 This is a flow chart of sending a police report in Embodiment 1 of the present application; Figure 4 is a flow chart for selecting a target terminal in Embodiment 1 of the present application; Figure 5 This is a structural diagram of a digital human business intelligent processing system for smart policing provided in Example 2 of the present application; Figure 6 It is a structural schematic diagram of an electronic device provided in Example 3 of the present application. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical scheme and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for the convenience of description, only part of the present application is shown in the accompanying drawings, but not all of the content. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow chart describes each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The process can correspond to a method, a function, a procedure, a subroutine, a subprogram, etc.

[0018] Embodiment 1: Figure 1 A flow chart of a method for intelligent processing of digital human business for smart policing provided in the first embodiment of the present application is given. The method for intelligent processing of digital human business for smart policing provided in this embodiment can be executed by a digital human business intelligent processing system for smart policing. The digital human business intelligent processing system for smart policing can be implemented by software and / or hardware. The digital human business intelligent processing system for smart policing can be composed of two or more physical entities, or it can be composed of one physical entity.

[0019] The following description is made by taking the digital human device of smart policing as the main body of the digital human business intelligent processing method for smart policing as an example. Figure 1 The digital human business intelligent processing method of smart policing specifically includes: S110, collecting audio and video data of the control site corresponding to the set control area, and extracting target behavior data, target audio data and target face data based on the audio and video data of the control site.

[0020] The digital human business intelligent processing system and method for smart policing proposed in this application aims to significantly improve the intelligence level and processing efficiency of police work by deeply integrating artificial intelligence technology with police workflows to efficiently identify and process complex police scenarios.

[0021] Among them, when handling the security monitoring task of the controlled area, a digital robot based on smart policing is set up at the controlled site, and the digital robot collects the audio and video data of the controlled site in real time, and extracts key information based on these data (i.e. target behavior data, target audio data and target face data). Use a high-definition camera or video surveillance system to record real-time video in the set controlled area. The camera is configured with sufficient resolution and viewing angle coverage to ensure that people, vehicles and other important targets in the area can be clearly captured. The audio information of the controlled site is synchronously collected through the microphone or the audio channel in the video surveillance system.

[0022] Furthermore, algorithms such as background difference, frame difference or optical flow are used to detect moving targets (such as people and cars) from video data. Then, through machine learning or deep learning models (such as convolutional neural network CNN, recurrent neural network RNN, etc.), the detected targets are analyzed for behavior, and abnormal behaviors such as running, wandering, fighting, and abnormal gathering are identified to obtain target behavior data; For target audio data extraction, audio signal processing technology is used to identify specific sound events from audio data, such as gunshots, alarms, etc. In addition, the voice signals detected on site are converted and text information is extracted for further semantic analysis to see if there are any potential dangerous situations.

[0023] For target face data extraction, face detection algorithms (such as Haar features, HOG features combined with SVM classifiers, or deep learning models such as YOLO, SSD, etc.) are used to quickly locate the face area in the video frame. By extracting facial features (such as LBP, SIFT, SURF features, or using deep learning features), subsequent face comparison can be performed to achieve identity recognition.

[0024] Optionally, extracting target behavior data, target audio data and target face data based on the audio and video data of the surveillance site includes: The audio and video data of the control site are input into the pre-built composite model, and the target behavior data, target audio data and target face data are output based on the composite model. The composite model pre-builds training samples based on the historical control site audio and video, and the model is trained based on the training samples.

[0025] In the process of extracting target behavior data, target audio data and target face data based on the audio and video data of the control site, a composite model is used to simultaneously process multiple types of information. The composite model combines multiple deep learning algorithms to cope with the complexity and diversity of video and audio data.

[0026] Prior to this, a large amount of audio and video data was collected from the surveillance site, including different scenes, lighting conditions, weather conditions, human behaviors, sound events, etc. The collected audio and video data were annotated in detail, including the categories of target behaviors (such as walking, running, fighting, etc.), labels of audio events (such as gunshots, alarms, cries for help, etc.), and frame selection and annotation of facial features. The annotated data was divided into training sets, validation sets, and test sets for model training, validation, and testing.

[0027] A composite model is further designed to process both video and audio data. The composite model includes multiple sub-networks, each of which is specialized for performing a specific task (i.e., face detection network, behavior recognition network, and speech detection network). The different networks of the composite model are trained separately using the training set data. During the training process, a variety of techniques are used to optimize the performance of the model, such as gradient descent, regularization, data augmentation, transfer learning, etc. The performance of the model is evaluated through the validation set, and the model architecture or training parameters are adjusted as needed. This process may require multiple iterations until the model reaches the specified performance on the validation set.

[0028] Deploy the trained composite model to the video surveillance system at the control site. Figure 2 , by inputting the audio and video data of the control site into the model in real time, the model processes the audio and video images of the control site respectively, and outputs the target behavior data, target audio data and target face data. This composite model-based method can make full use of the rich information in the audio and video data to improve the security monitoring level of the control site. At the same time, by continuously optimizing and updating the model, it can adapt to the ever-changing security threats and monitoring needs.

[0029] S120. Traverse the set alarm feature library based on the target behavior data, target audio data and target face data. When the corresponding alarm feature is traversed, output the alarm report of the current controlled area according to the traversed alarm feature, and report the alarm report to the alarm system.

[0030] In the process of traversing the set alarm feature library based on the target behavior data, target audio data and target face data, and outputting the alarm report of the current control area based on it, the target behavior data, target audio data and target face data extracted from the audio and video data of the control site are first integrated and pre-processed, such as data cleaning (removing noise, outliers, etc.), normalization, feature extraction, etc., to ensure the consistency and comparability of the data.

[0031] Then, the police situation feature library is queried based on the preprocessed data. The police situation feature library is a predefined data set that contains various possible police situation features and their corresponding judgment criteria and rules. These features include specific behavior patterns (such as fast running, gathering and fighting, etc.), audio events (such as gunshots, cries for help), facial features (such as blacklisted people, suspicious persons), etc. Each feature has a clear definition, threshold and / or matching algorithm.

[0032] Use the integrated and preprocessed data to traverse each feature in the alarm feature library. For each feature, use the corresponding matching algorithm to compare the current data with the feature definition. For example, calculate similarity, detect specific patterns, perform classification tasks, etc. If the current data meets the judgment criteria of a certain feature (i.e., the match is successful), then the definition and rules of the feature determine that there is a corresponding alarm in the current control area. Once it is determined that there is an alarm, the system immediately generates an alarm report. The alarm report can determine the alarm type based on the matched alarm features, record the exact time when the alarm occurred, and clarify the specific location of the alarm. It may include detailed coordinates or identification of the control area, and attach relevant data that triggered the alarm, such as video clips, audio recordings, and facial images. After the alarm report is generated, the system will report it to the alarm system in a timely manner. So that relevant departments or personnel can quickly obtain and process it. Through the above steps, effective monitoring and timely response to the control area can be achieved, and the ability of security prevention and emergency handling can be improved.

[0033] Optionally, when the corresponding alarm feature is traversed, it also includes: The audio and video data of the control site are written into the optimized sample set. When the number of samples in the optimized sample set reaches the set threshold, the composite model is optimized and trained based on the optimized sample set.

[0034] When the system traverses the corresponding alarm feature and confirms the existence of an alarm, in addition to generating and submitting an alarm report, the audio and video data of the scene that triggered the alarm (or its key parts, such as video clips containing abnormal behavior, audio records containing specific audio events, etc.) can be written into the optimized sample set. These samples should be marked as associated with specific alarm features so that they can be used for subsequent optimization training of the model.

[0035] The system continuously monitors the number of samples in the optimization sample set. For example, a counter is set to increase the corresponding value each time a new sample is written into the optimization sample set. When the number of samples reaches the set threshold (the threshold can be adjusted according to the actual situation to ensure that there are enough samples for training while avoiding too long waiting time), the system triggers the optimization training process of the model.

[0036] After the number of samples in the optimized sample set reaches the set threshold, the system uses these samples to optimize the composite model for training. The audio and video data in the optimized sample set are preprocessed, including data cleaning and feature extraction, to ensure that they are consistent with the data format and features used in model training. The composite model is then trained using the preprocessed optimized sample set. During the training process, the model's parameters, structure, or training strategy can be adjusted to improve the model's recognition performance on specific alarm features. After training is complete, the optimized model is evaluated for performance using a validation set or a test set. If the optimized model performs well in the performance evaluation, the system updates it to the model currently in use and deploys it to the video surveillance system at the site. In this way, the system is able to use the updated model to more accurately identify and handle future alarms.

[0037] Optionally, after outputting the alarm report of the current control area according to the traversed alarm features, the method further includes: According to the traversed police situation characteristics, historical police processing data is queried, processing experience information is extracted from the historical police processing data, and the processing experience information is written into the police situation report.

[0038] After outputting the police situation report of the current control area according to the traversed police situation characteristics, the historical police processing data can be further queried according to these police situation characteristics, and the relevant processing experience information can be extracted, and then this information can be written into the police situation report.

[0039] Understandably, historical police processing data contains a wealth of processing experience and successful cases, which can provide valuable reference and decision-making support for the current police situation. By drawing on historical experience, relevant departments or personnel can formulate effective response measures more quickly, thereby shortening response time and improving processing efficiency. Processing solutions based on historical experience are often more mature and effective, which helps to improve the current police situation processing effect and reduce potential losses and risks.

[0040] Based on this, this application queries the system or database for relevant historical police processing data based on the traversed police situation characteristics (such as police situation type, location, related data, etc.). These data may include past police situation reports, processing records, decision-making processes, result feedback, etc. Processing experience information that matches the current police situation characteristics is extracted from the historical police processing data. This information includes strategies, key steps, precautions, resource allocation plans, etc. for successfully handling similar police situations. The extracted processing experience information is then written into the current police situation report. As part of the report, this information provides valuable reference and guidance for relevant personnel receiving the report. Through the above process, the police situation report not only contains the basic situation of the current police situation, but also incorporates information on historical processing experience, providing more comprehensive and powerful support for the rapid and effective handling of police situations.

[0041] S130. Based on the alarm report, call the associated cameras in the control area, track the alarm based on the associated cameras, and obtain real-time tracking video; obtain the police officer terminals within the set range, select at least one target terminal from the police officer terminals, upload the alarm report and tracking video to the target terminal, so as to provide the corresponding police officer with an alarm prompt; after the set time period, query the location distance between the target terminal and the control area, and when it is detected that the location distance exceeds the set distance threshold, forward the alarm report according to the alarm type of the alarm report.

[0042] Based on the above police reports, this application realizes an efficient and practical emergency response process by calling associated cameras to track police situations, sending police reports and tracking videos to police officers' terminals, and forwarding police reports according to the locations of police officers.

[0043] Among them, according to the control area described in the police report, the system automatically calls the related cameras in the area. Using video analysis technology, the video captured by the camera is analyzed in real time to track the development of the police situation. Specifically, it may include target tracking, behavior recognition, etc. The tracked real-time video data is saved and prepared to be sent to the police.

[0044] In addition, by determining the police officers’ terminals within the set range, these terminals can be located through GPS, network location, etc. According to the location of the police officers’ terminals, at least one target terminal is automatically selected from the acquired terminals. The police situation report and real-time tracking video are sent to the target terminal through wireless communication methods (such as 4G / 5G, Wi-Fi, etc.). Figure 3As shown in the figure, when a police situation is detected in the control area A, the police terminal a, b, and c are determined within the set range, and the police terminal a is selected as the target terminal. After receiving the police situation report and tracking video, the target terminal prompts the corresponding police officer through sound, vibration, screen display, etc., and prompts the police officer to go to handle the current police situation in real time.

[0045] By setting a time period, it is used to monitor whether the police officers have arrived at or are heading to the controlled area. After the set time period, the system will query the location distance between the target terminal and the controlled area again. If it is detected that the location distance exceeds the set distance threshold (that is, the police officer fails to reach or approach the controlled area within the specified time), the next step is executed. According to the type of alarm report, the alarm report is forwarded to other relevant police officer terminals or command centers to ensure that the alarm is handled in a timely and effective manner. In this way, the system can efficiently call resources, track alarms, prompt police officers, and flexibly adjust and handle according to actual conditions, thereby ensuring that the alarm is handled in a timely and effective manner.

[0046] At this point, by calling the associated camera for real-time tracking, more comprehensive and detailed police information is provided, which helps police officers make accurate judgments. In addition, by dynamically selecting the target terminal based on the police officer's location information, rapid response and efficient processing of police situations can be achieved. In addition, when police officers are unable to arrive at the scene in time, forwarding the police report can ensure that the police situation is properly handled to avoid deterioration.

[0047] Optionally, refer to Figure 4 , selecting at least one target terminal from the police officer terminals, including: S1301, determine the alarm level of the alarm report, and determine the corresponding number of processing personnel according to the alarm level; S1302. Taking the control area as the center, select a corresponding number of target terminals from the police officers' terminals from near to far according to the number of personnel to be processed.

[0048] In the process of selecting the target terminal from the police officers' terminals, tasks are reasonably allocated according to the urgency of the alarm and the geographical location of the police officers, thereby improving the response speed and efficiency. First, the system evaluates the alarm level of the alarm report. The alarm level can be divided into urgent, important, general, etc. according to the nature, severity, potential impact and other factors of the alarm. According to the assessed alarm level, the system needs to determine the corresponding number of processing personnel. Generally speaking, the higher the alarm level, the more processing personnel are required to ensure that the alarm can be handled in a timely and adequate manner.

[0049] Then, the current location information of all police officers' terminals within the set range is obtained. Then, with the control area as the center, the police officers are sorted according to the distance between them and the control area, and the selection is performed from near to far. According to the number of processing personnel determined in step S1301, a corresponding number of target terminals are selected from the sorted police officers' terminals. In this way, the system can efficiently select the police officer terminal that is most suitable for handling the current police situation, providing strong support for the rapid response and effective handling of the police situation.

[0050] Among them, determining the alarm level of the alarm report includes: The preset alarm score is queried according to the traversed alarm characteristics, and the alarm level of the alarm report is determined according to the alarm score.

[0051] The present application determines the alarm level of the alarm report, which is a process of quantitative evaluation based on the alarm characteristics. In this process, the alarm level of the alarm report is determined by querying the preset alarm scores and determining these scores. Prior to this, corresponding alarm scores are defined for different types of alarm characteristics. These scores can be set based on historical data and specified standards to reflect the degree of influence of different characteristics on the urgency and severity of the alarm. When the alarm report is triggered, the system traverses all relevant alarm characteristics and assigns a corresponding score to each characteristic according to the preset score table. These scores are then added up to get the total score. Based on the calculated total score, the system compares it with the preset alarm level threshold to determine the alarm level of the alarm report.

[0052] Optionally, forwarding the police report according to the police type of the police report includes: According to the type of police situation reported, select the backend dispatch center or the nearest police officer terminal to forward the police situation report.

[0053] Different types of responses are made to different types of alarms based on the alarm types in the identified alarm reports. For example, for certain specific types of alarms or alarms with a high degree of urgency, the system directly forwards the alarm report to the backend dispatch center. The dispatch center will coordinate relevant resources to respond and handle the alarm based on the type and severity of the alarm. For alarms that require on-site processing and are of moderate urgency, the system can choose to forward the alarm report to the nearest police officer terminal to ensure that the police officer can quickly arrive at the scene and conduct preliminary processing and investigation. After determining the forwarding target, the system immediately sends the alarm report electronically to the selected target to ensure that the target can immediately notice and respond to the alarm.

[0054] In the above, audio and video data of the control site are collected through the corresponding set control area, and the target behavior data, target audio data and target face data are extracted based on the audio and video data of the control site; the set police feature library is traversed based on the target behavior data, target audio data and target face data, and when the corresponding police feature is traversed, the police report of the current controlled area is output according to the traversed police feature, and the police report is reported to the police system; based on the police report, the associated camera of the controlled area is called, and the police tracking is performed based on the associated camera to obtain real-time tracking video; the police officer terminals within the set range are obtained, at least one target terminal is selected from the police officer terminals, and the police report and tracking video are uploaded to the target terminal to provide a police prompt to the corresponding police officer; after the set time period, the location distance between the target terminal and the control area is queried, and when it is detected that the location distance exceeds the set distance threshold, the police report is forwarded according to the police type of the police report. By adopting the above-mentioned technical means, it is possible to accurately identify on-site police situations by combining the audio and video data of the controlled area, and to realize police situation reports by interacting with the associated cameras and associated police officer terminals at the controlled area, so as to promptly notify police officers to handle police situations and improve police handling efficiency. While reducing the pressure of patrol and control on police officers, it also effectively assists police officers in handling police situations.

[0055] Embodiment 2: Based on the above embodiments, Figure 5 This is a schematic diagram of the structure of a digital human business intelligent processing system for smart police services provided in Example 2 of this application. Figure 5 The digital human business intelligent processing system for smart policing provided in this embodiment specifically includes: The acquisition module 21 is used to collect the audio and video data of the control site corresponding to the set control area, and extract the target behavior data, target audio data and target face data based on the audio and video data of the control site; The alarm recognition module 22 is used to traverse the set alarm feature library based on the target behavior data, the target audio data and the target face data, and when the corresponding alarm feature is traversed, output the alarm report of the current control area according to the traversed alarm feature, and report the alarm report to the alarm system; The alarm business processing module 23 is used to call the associated cameras in the control area based on the alarm report, track the alarm based on the associated cameras, and obtain real-time tracking video; obtain the police officer terminals within the set range, select at least one target terminal from the police officer terminals, and upload the alarm report and tracking video to the target terminal to provide alarm prompts to the corresponding police officers; after a set period of time, query the location distance between the target terminal and the control area, and when it is detected that the location distance exceeds the set distance threshold, forward the alarm report according to the alarm type of the alarm report.

[0056] Specifically, based on the on-site audio and video data, target behavior data, target audio data and target face data are extracted, including: The audio and video data of the control site are input into the pre-built composite model, and the target behavior data, target audio data and target face data are output based on the composite model. The composite model pre-builds training samples based on the historical control site audio and video, and the model is trained based on the training samples.

[0057] Specifically, when the corresponding alarm feature is traversed, it also includes: The audio and video data of the control site are written into the optimized sample set. When the number of samples in the optimized sample set reaches the set threshold, the composite model is optimized and trained based on the optimized sample set.

[0058] Specifically, after outputting the alarm report of the current control area according to the traversed alarm features, it also includes: According to the traversed police situation characteristics, historical police processing data is queried, processing experience information is extracted from the historical police processing data, and the processing experience information is written into the police situation report.

[0059] Specifically, selecting at least one target terminal from the police officer terminals includes: Determine the alarm level of the alarm report and determine the corresponding number of processing personnel according to the alarm level; Taking the control area as the center, select the corresponding number of target terminals from the police officers' terminals from near to far according to the number of personnel to be processed.

[0060] Specifically, determine the alarm level of the alarm report, including: The preset alarm score is queried according to the traversed alarm characteristics, and the alarm level of the alarm report is determined according to the alarm score.

[0061] Specifically, the police report is forwarded according to the type of the police report, including: According to the type of police situation reported, select the backend dispatch center or the nearest police officer terminal to forward the police situation report.

[0062] In the above, audio and video data of the control site are collected through the corresponding set control area, and the target behavior data, target audio data and target face data are extracted based on the audio and video data of the control site; the set police feature library is traversed based on the target behavior data, target audio data and target face data, and when the corresponding police feature is traversed, the police report of the current controlled area is output according to the traversed police feature, and the police report is reported to the police system; based on the police report, the associated camera of the controlled area is called, and the police tracking is performed based on the associated camera to obtain real-time tracking video; the police officer terminals within the set range are obtained, at least one target terminal is selected from the police officer terminals, and the police report and tracking video are uploaded to the target terminal to provide a police prompt to the corresponding police officer; after the set time period, the location distance between the target terminal and the control area is queried, and when it is detected that the location distance exceeds the set distance threshold, the police report is forwarded according to the police type of the police report. By adopting the above-mentioned technical means, it is possible to accurately identify on-site police situations by combining the audio and video data of the controlled area, and to realize police situation reports by interacting with the associated cameras and associated police officer terminals at the controlled area, so as to promptly notify police officers to handle police situations and improve police handling efficiency. While reducing the pressure of patrol and control on police officers, it also effectively assists police officers in handling police situations.

[0063] The digital human business intelligent processing system for smart policing provided in the second embodiment of the present application can be used to execute the digital human business intelligent processing method for smart policing provided in the above-mentioned first embodiment, and has corresponding functions and beneficial effects.

[0064] Embodiment three: Embodiment 3 of the present application provides an electronic device, referring to Figure 6 The electronic device includes: a processor 31, a memory 32, a communication module 33, an input device 34 and an output device 35. The number of processors in the electronic device can be one or more, and the number of memories in the electronic device can be one or more. The processor, memory, communication module, input device and output device of the electronic device can be connected via a bus or other means.

[0065] As a computer-readable storage medium, the memory can be used to store software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the digital human business intelligent processing method for smart policing described in any embodiment of the present application (for example, the various modules in the digital human business intelligent processing system for smart policing). The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0066] The communication module is used for data transmission.

[0067] The processor executes various functional applications and data processing of the device by running software programs, instructions and modules stored in the memory, thereby realizing the above-mentioned digital human business intelligent processing method for smart policing.

[0068] The input device can be used to receive input digital or character information and generate key signal input related to user settings and function control of the device. The output device can include display devices such as display screens.

[0069] The electronic device provided above can be used to execute the digital human business intelligent processing method for smart policing provided in the above embodiment 1, and has corresponding functions and beneficial effects.

[0070] Embodiment 4: The embodiment of the present application also provides a storage medium containing computer executable instructions, which are used to execute a digital human business intelligent processing method for smart policing when executed by a computer processor. The digital human business intelligent processing method for smart policing includes: collecting audio and video data of the control scene corresponding to the set control area, extracting target behavior data, target audio data and target face data based on the audio and video data of the control scene; traversing the set police situation feature library based on the target behavior data, target audio data and target face data, and when the corresponding police situation feature is traversed, generating a police situation feature library according to the traversed police situation feature. The system outputs the alarm report of the current controlled area based on the situation characteristics, and reports the alarm report to the alarm system; based on the alarm report, the associated cameras in the controlled area are called, and the alarm tracking is performed based on the associated cameras to obtain real-time tracking video; the police officer terminals within the set range are obtained, at least one target terminal is selected from the police officer terminals, and the alarm report and tracking video are uploaded to the target terminal to provide alarm prompts to the corresponding police officers; after the set time period, the location distance between the target terminal and the controlled area is queried, and when it is detected that the location distance exceeds the set distance threshold, the alarm report is forwarded according to the alarm type of the alarm report Storage medium - any of various types of memory devices or storage devices. The term "storage medium" is intended to include: installation media, such as CD-ROM, floppy disk or tape device; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (such as hard disk or optical storage); registers or other similar types of memory elements, etc. Storage media may also include other types of memory or combinations thereof. In addition, the storage medium may be located in the first computer system in which the program is executed, or may be located in a different second computer system, which is connected to the first computer system via a network (such as the Internet). The second computer system can provide program instructions to the first computer for execution. The term "storage medium" may include two or more storage media residing in different locations (for example, in different computer systems connected by a network). The storage medium may store program instructions (for example, embodied as a computer program) that can be executed by one or more processors.

[0071] Of course, the storage medium containing computer executable instructions provided in the embodiment of the present application, whose computer executable instructions are not limited to the digital human business intelligent processing method for smart policing as described above, can also execute related operations in the digital human business intelligent processing method for smart policing provided in any embodiment of the present application.

[0072] The digital human business intelligent processing system, storage medium and electronic device for smart policing provided in the above embodiments can execute the digital human business intelligent processing method for smart policing provided in any embodiment of the present application. For technical details not described in detail in the above embodiments, please refer to the digital human business intelligent processing method for smart policing provided in any embodiment of the present application.

[0073] The above are only preferred embodiments of the present application and the technical principles used. The present application is not limited to the specific embodiments described herein, and various obvious changes, readjustments and substitutions that can be made by those skilled in the art will not deviate from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the claims.

Claims

1. A digital human business intelligent processing system for smart policing, characterized in that: include: A collection module is used to collect audio and video data of the control site corresponding to the set control area, and extract target behavior data, target audio data and target face data based on the audio and video data of the control site; A warning recognition module is used to traverse the set warning feature library based on the target behavior data, target audio data and target face data, and when the corresponding warning feature is traversed, output a warning report of the current control area according to the traversed warning feature, and report the warning report to the warning system; The alarm service processing module is used to call the associated camera of the control area based on the alarm report, track the alarm based on the associated camera, and obtain real-time tracking video; obtain the police officer terminals within the set range, select at least one target terminal from the police officer terminals, report the alarm report and the tracking video to the target terminal, so as to provide the corresponding police officer with an alarm reminder; After a set period of time, the location distance between the target terminal and the controlled area is queried, and when it is detected that the location distance exceeds a set distance threshold, the alarm report is forwarded according to the alarm type of the alarm report.

2. The digital human business intelligent processing system for smart policing according to claim 1 is characterized in that: The selecting at least one target terminal from the police officer terminals comprises: Determine the alarm level of the alarm report, and determine the corresponding number of processing personnel according to the alarm level; Taking the control area as the center, a corresponding number of target terminals are selected from the police officer terminals from near to far according to the number of the processing personnel.

3. The digital human business intelligent processing system for smart policing according to claim 2 is characterized in that: Determining the alarm level of the alarm report includes: A preset alarm score is queried according to the traversed alarm characteristics, and the alarm level of the alarm report is determined according to the alarm score.

4. The digital human business intelligent processing system for smart policing according to claim 1 is characterized in that: The forwarding of the alarm report according to the alarm type of the alarm report includes: According to the alarm type of the alarm report, a background dispatch center or the nearest police officer terminal is selected to forward the alarm report.

5. The digital human business intelligent processing system for smart policing according to claim 1 is characterized in that: The extracting of target behavior data, target audio data and target face data based on the control site audio and video data includes: The control site audio and video data are input into a pre-constructed composite model, and target behavior data, target audio data and target face data are output based on the composite model. The composite model pre-constructs training samples based on historical control site audio and video, and model training is performed based on the training samples.

6. The digital human business intelligent processing system for smart policing according to claim 5 is characterized in that: When the corresponding alarm features are traversed, it also includes: The audio and video data of the control site are written into an optimized sample set, and when the number of samples in the optimized sample set reaches a set threshold, optimization training of the composite model is performed based on the optimized sample set.

7. The digital human business intelligent processing system for smart policing according to claim 1 is characterized in that: After outputting the alarm report of the current control area according to the traversed alarm characteristics, it also includes: The historical police processing data is queried according to the traversed police situation features, the processing experience information is extracted from the historical police processing data, and the processing experience information is written into the police situation report.

8. A method for intelligent processing of digital human services for smart policing, applied to the intelligent processing system for digital human services for smart policing as claimed in any one of claims 1 to 7, characterized in that: include: Collecting the audio and video data of the control scene corresponding to the set control area, and extracting the target behavior data, target audio data and target face data based on the audio and video data of the control scene; Based on the target behavior data, the target audio data and the target face data, the set alarm feature library is traversed, and when the corresponding alarm feature is traversed, the alarm report of the current control area is output according to the traversed alarm feature, and the alarm report is reported to the alarm system; Based on the alarm report, call the associated camera of the control area, track the alarm based on the associated camera, and obtain real-time tracking video; obtain the police officer terminal within the set range, select at least one target terminal from the police officer terminals, report the alarm report and the tracking video to the target terminal, so as to provide the corresponding police officer with an alarm prompt; After a set period of time, the location distance between the target terminal and the controlled area is queried, and when it is detected that the location distance exceeds a set distance threshold, the alarm report is forwarded according to the alarm type of the alarm report.

9. An electronic device, characterized in that: include: memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the digital human business intelligent processing method for smart policing as described in claim 8.

10. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions, when executed by a computer processor, are used to execute the digital human business intelligent processing method for smart policing as described in claim 8.

Images