CN116052103A

CN116052103A - Method, device, computer equipment and storage medium for processing monitoring data

Info

Publication number: CN116052103A
Application number: CN202310007656.5A
Authority: CN
Inventors: 周东开
Original assignee: Wuhan Lotus Technology Co Ltd
Current assignee: Ningbo Lutes Robotics Co ltd
Priority date: 2023-01-04
Filing date: 2023-01-04
Publication date: 2023-05-02

Abstract

The application relates to a monitoring data processing method, a monitoring data processing device, computer equipment and a storage medium. The method comprises the following steps: acquiring image data in a vehicle target monitoring area; detecting target events of the image data and outputting detection results; when the detection result triggers a preset first event, desensitizing the image data, and outputting first monitoring data which can be directly checked; when the detection result triggers a preset first event and a preset second event, second monitoring data corresponding to the second event in the image data are stored, and after preset viewing authority is configured, a preset alarm processing flow is executed. By adopting the method, privacy disclosure caused by displayed monitoring data can be avoided, and the monitoring data of the alarm event which can be checked according to the authority can be saved, so that privacy disclosure is further avoided.

Description

Method, device, computer equipment and storage medium for processing monitoring data

Technical Field

The present invention relates to the field of computer technologies, and in particular, to a method and apparatus for processing monitoring data, a computer device, and a storage medium.

Background

With the rapid development of automobile technology, automobile intelligent configuration is becoming higher and higher. The security system of the vehicle takes a guard mode as an example, and the guard mode provides a guarantee for the security of the vehicle, but also causes privacy disputes.

In the sentry mode, the surrounding environment of the vehicle is monitored through the vehicle-mounted camera, clear video images around the vehicle body can be acquired no matter in the daytime or at night, and the existing convolutional neural network technology is utilized to identify adverse behaviors of people or objects on the surrounding side of the vehicle to the vehicle, such as rubbing the vehicle, placing personal articles on the head of the vehicle and the like, and starting alarm reminding to inform the vehicle owner when the adverse behaviors occur. However, in the sentry mode, because sensitive information such as faces of pedestrians and license plates of the vehicles can be recorded in the collected video images, privacy leakage is caused to the pedestrians and vehicles in the past.

Disclosure of Invention

Based on the above, it is necessary to provide a method, an apparatus, a computer device and a storage medium for processing monitoring data, which are capable of adding data desensitization processing to pedestrians/vehicles in a target monitoring area, avoiding privacy disclosure caused by displayed monitoring data, and preserving the monitoring data of alarm events which can be checked according to authority, so as to further avoid privacy disclosure.

The application provides a monitoring data processing method, which comprises the following steps:

acquiring image data in a vehicle target monitoring area;

detecting target events of the image data and outputting detection results;

when the detection result triggers a preset first event, desensitizing the image data, and outputting first monitoring data which can be directly checked;

when the detection result triggers a preset first event and a preset second event, second monitoring data corresponding to the second event in the image data are stored, and after preset viewing authority is configured, a preset alarm processing flow is executed.

In one embodiment, the step of acquiring image data within a target area of a vehicle includes:

acquiring source data in a target monitoring area through a vehicle-mounted camera;

and preprocessing the source data, and caching the image data to be processed.

In one embodiment, the vehicle-mounted camera is a fisheye camera;

preprocessing source data to obtain image data to be processed, wherein the preprocessing comprises the following steps:

performing internal reference calibration by using an image processing open source library OpenCV in advance;

and correcting the distortion of the source data according to the calibration internal parameters to obtain the image data to be processed.

In one embodiment, the step of detecting the target event on the image data and outputting the detection result includes:

detecting a target object of the image data, and triggering a first event when a preset target object exists;

and performing behavior recognition on the target object, outputting a recognition result, matching the recognition result with a preset alarm behavior, and triggering a second event when the matching is successful.

In one embodiment, the step of detecting a target object for the image data, and triggering the first event when a preset target object exists includes:

detecting a target object of the image data by using a preset neural network model;

if the preset target object exists, the type of the target object is identified, and the target object in the image data is marked to serve as a first triggering event.

In one embodiment, the target object comprises a pedestrian, a vehicle; the target features comprise a pedestrian face and a vehicle license plate;

when the detection result triggers a preset first event, performing desensitization processing on the image data, and outputting first monitoring data which can be directly checked, wherein the method comprises the following steps of:

acquiring the type of a target object;

screening a preset feature recognition model according to the type of the target object to recognize target features in the target object and determining a feature area where the target features are located;

generating a target mask corresponding to the feature region by using a preset mask generation technology;

splicing the target mask with the image data according to the size and the position to obtain a plurality of desensitized image frames;

and caching a preset number of desensitized image frames by using an image queue, generating first monitoring data and outputting the first monitoring data for direct viewing.

In one embodiment, performing behavior recognition on the target object, outputting a recognition result, matching the recognition result with a preset alarm behavior, and triggering a second event when the matching is successful, including:

performing behavior recognition on the target object by using a preset behavior recognition model, and outputting a recognition result;

and matching the identification result with a preset alarm behavior, and triggering a second event if the matching is successful.

The application provides a monitoring data processing device, the device includes:

the data acquisition module is used for acquiring image data in a vehicle target area;

the target detection module is used for detecting target events of the image data and outputting detection results;

the data desensitization module is used for performing desensitization processing on the image data when the detection result triggers a preset first event and outputting first monitoring data which can be directly checked;

and the alarm processing module is used for storing second monitoring data corresponding to the second event in the image data when the detection result triggers the preset first event and second event, and executing a preset alarm processing flow after configuring the preset checking authority.

The application provides a computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

acquiring image data in a vehicle target monitoring area;

detecting target events of the image data and outputting detection results;

The present application provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of:

acquiring image data in a vehicle target monitoring area;

detecting target events of the image data and outputting detection results;

The monitoring data processing method, the device, the computer equipment and the storage medium have the following technical effects.

Because the target event detection is performed on the image data, the preset event occurring in the target monitoring area can be detected. The first event and the second event are preset, when the first event is triggered, the fact that a preset target object appears in a target monitoring area of a vehicle is indicated, a certain potential safety hazard is caused to the vehicle, desensitization processing is carried out on the target object in image data at the moment, and first monitoring data which can be directly checked by a vehicle owner are generated, so that the problem of privacy leakage caused to past pedestrians/vehicles in a whistle mode of the vehicle is solved.

When the detection result is adopted to trigger the first event and the second event, the second monitoring data corresponding to the second event in the image data is stored, and after the preset viewing authority is configured, the preset alarm processing flow is executed, so that the vehicle owner cannot directly view the second monitoring data and can only view the second monitoring data by using a supervision platform with the preset viewing authority, the privacy of the monitoring data is further improved, and disputes caused by the second event are reduced.

The first monitoring data and the second monitoring data of the alarm event are output after the desensitization processing to configure the checking authority, so that after the intelligent vehicle parking system is integrated into a guard mode of the vehicle parking, the privacy of pedestrians around the intelligent vehicle can be protected while the function is ensured, the sensitive information is avoided, the social public order is maintained, the guard mode system is commercialized to be more mature, the detection speed is fully ensured while the desensitization effect is ensured by the output first monitoring data, and the coexistence of the accuracy and the real-time performance is realized.

The monitoring data processing method provided by the embodiment can be applied to unexpected scenes of vehicles, such as civil security, and sensitive information can be seen only when the monitoring data processing method has high-level authority, for example, a monitoring hall of a security room can only display first monitoring data after desensitization (faces and license plates are demosaiced); but the local public security authority is authorized to view the image data (original image without desensitization) before desensitization, and corresponding data is acquired according to the authority, so that privacy is protected and evidence is reserved, and the technical scheme has certain universality.

Drawings

FIG. 1 is a diagram of an application environment for a method of monitoring data processing in one embodiment;

FIG. 2 is a flow chart of a prior art sentinel pattern monitoring data processing method;

FIG. 3 is a flow chart of a sentinel mode monitoring data processing method in the present embodiment;

FIG. 4 is a schematic diagram of a vehicle monitoring area in one embodiment;

FIG. 5 is a flow chart of a method of monitoring data processing in one embodiment;

FIG. 6 is a schematic diagram of source data correction effects in one embodiment;

FIG. 7 is a schematic diagram of a control motherboard in one embodiment;

FIG. 8 is a schematic diagram of a face detection annotation in one embodiment;

FIG. 9 is a schematic diagram of the effect of desensitization treatment in one embodiment;

FIG. 10 is a diagram of triggering an alarm event notification in one embodiment;

FIG. 11 is a schematic diagram showing details of a flow of a method of processing monitoring data in one embodiment;

FIG. 12 is a block diagram of a monitoring data processing apparatus in one embodiment;

fig. 13 is an internal structural view of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.

The monitoring data processing method provided by the application can be applied to a vehicle environment shown in fig. 1. The monitoring terminal 10 is connected with the vehicle-mounted camera 20 and the vehicle owner terminal 30, the vehicle-mounted camera 20 is arranged on the periphery of the vehicle and is used for collecting source data of surrounding environments of the vehicle, the monitoring terminal 10 obtains the source data collected by the vehicle-mounted camera 20, and after preset image data processing, the monitoring terminal is notified to the vehicle owner terminal 30 when an alarm event of a whistle mode is triggered. As shown in fig. 2, the present data processing flow of the sentinel mode is that the camera acquires the image data, and after the image data is processed, the image data is matched with the alarm condition of the sentinel mode, if the image data is consistent with the alarm condition, the alarm is triggered, otherwise, the image data is continuously acquired. Fig. 3 is a diagram showing a method for processing monitoring data in this embodiment, after image data processing, two steps are required to be executed, one is to perform desensitization processing on the data to obtain first monitoring data, and the other is to match the image data with alarm conditions of a whistle mode, if the matching is successful, an alarm is triggered, a video is recorded, and second monitoring data is obtained. The first monitoring data is cached in the vehicle local area, and the vehicle owner can directly check the first monitoring data in the vehicle local area. And the second monitoring data is configured with preset checking authority, and after an alarm event occurs, the vehicle owner can inform the supervision platform with the checking authority of the second monitoring data so as to execute corresponding punishment measures. In fig. 4, four target monitoring areas are shown, but the number of the target monitoring areas is not limited in theory, and the whole surrounding environment of the vehicle can be included.

In one embodiment, as shown in fig. 5, a method for processing monitoring data is provided, and the method is applied to the monitoring terminal in fig. 1 for illustration, and includes the following steps:

step S101, obtaining image data in a vehicle target monitoring area;

step S102, detecting a target event on image data and outputting a detection result;

step S103, when the detection result triggers a preset first event, desensitizing processing is carried out on the image data, and first monitoring data which can be directly checked is output;

step S104, when the detection result triggers a preset first event and a preset second event, second monitoring data corresponding to the second event in the image data are stored, and after the preset viewing authority is configured, a preset alarm processing flow is executed.

In step S101, the step of acquiring image data in the vehicle target monitoring area includes: acquiring source data in a target monitoring area through a vehicle-mounted camera; and preprocessing the source data to obtain the image data to be processed. The collection behavior of the vehicle-mounted camera is executed under a sentinel mode based on vehicle parking. When a moving target (pedestrian or vehicle) appears in a target monitoring area of the host vehicle, a certain potential safety hazard or loss may be caused to the host vehicle, and the monitoring data processing method in the embodiment needs to be executed. When a first event is triggered, the fact that a preset target object appears in a vehicle target monitoring area is indicated, and a certain potential safety hazard exists in the vehicle. When the second event is triggered, the first target event is triggered, and the preset target object has preset alarm behaviors, so that the damage loss of the vehicle is caused. Among them, the pretreatment modes include but are not limited to: the filtering and screening processes of the source data are not limited to this embodiment.

The vehicle-mounted camera in this embodiment may be a look-around camera. In one embodiment, the in-vehicle camera employs a fisheye camera in a pan around camera. For example, 4 fisheye cameras with an angle of view of 150 degrees and ultra-wide angle can cover a 360-degree angle of view of a vehicle, definition can be ensured by adopting 1080P high resolution and 30FPS frame rate, and images of the periphery of the vehicle body can be acquired in real time. The vehicle-mounted camera in the embodiment supports multiple communication interfaces such as GMSL (Gigabit Multimedia Serial Links, gigabit multimedia serial link), ethernet and the like, and is convenient for vehicle-mounted deployment application.

However, the source data collected by the existing fisheye camera has distortion, so the preprocessing step in this embodiment includes correcting the distortion of the source data.

Further, the step of preprocessing the source data to obtain the image data to be processed includes:

and correcting the distortion of the source data according to the calibration internal parameters, and caching the image data to be processed.

As shown in fig. 6, fig. 6 (a) is a fish-eye viewing angle, and fig. 6 (b) is a corrected viewing angle.

The monitor terminal in this embodiment may use the high-performance control motherboard NVIDIA orin to execute each flow of the monitor data processing method. As shown in FIG. 7, the AI performance computing power of NVIDIA orin of the high-performance control main board reaches 200TOPS, the CPU can adopt Arm-A78 with 12 cores, the frequency can reach 2.2GHZ, and the method can be used for ADAS auxiliary driving and L4-level automatic driving.

In one embodiment, the image data acquired in step S101 is the latest image data buffered after preprocessing, for example, the image data buffered within 2 minutes.

In step S102, the step of detecting a target event on the image data and outputting a detection result includes:

Further, when the image data triggers a second event, indicating that the first event has been triggered, the second event is based on a further complement of the first event.

In one embodiment, the step of performing target object detection on the image data, and triggering the first event when the preset target object exists includes:

Wherein the target object includes, but is not limited to, moving pedestrians, vehicles, animals, loads, represented as objects loaded on other vehicles.

The neural network model preset in the embodiment can adopt a YOLO model, and the YOLO model is used as a classical target detection model, so that a target object can be detected rapidly and accurately, and the neural network model is not limited to an application scene.

As shown in fig. 8, in the actual application scenario, pedestrians and vehicles are marked by using YOLO model YOLOV5 series algorithm.

The target object in the present embodiment includes pedestrians and vehicles; the target features include pedestrian faces and vehicle license plates. In step S103, when the detection result triggers a preset first event, a desensitization process is performed on the image data, and a step of outputting directly viewable first monitoring data includes:

acquiring the type of a target object;

screening a preset feature recognition technology according to the type of the target object to recognize target features in the target object and determining a feature area where the target features are located;

In one embodiment, the size and position of the target feature are obtained to determine the feature region in which the target feature is located. The target mask generated by the mask generation technique in this embodiment may be, but is not limited to, a mosaic, as shown in fig. 9, that is, in this embodiment, a pedestrian face and a vehicle license plate are used as sensitive privacy information. In this embodiment, the first monitoring data after desensitization is stored locally in the vehicle, so that the owner can view and save the first monitoring data locally, and privacy protection of pedestrians/vehicles is ensured.

The feature recognition technique in this embodiment is also based on a neural network technique. In one embodiment, when the target object that appears is a pedestrian, then the target feature is a pedestrian face. The feature recognition technology in this embodiment may employ a libfacedetection model to perform face detection. The libfacedetection model is based on a CNN image face detection open source library, the model parameters can reach 232 ten thousand, and the volume is 3.34M. The CNN model has been converted to static variables in the C source file and the source code is independent of any other library, the source code can be compiled under the Windows, linux system using a c++ compiler and SIMD instructions are used to speed up detection.

Therefore, NEON instruction set acceleration can be used on the control main board orin to rapidly detect the face in the image, so that the detection efficiency is improved, and when the confidence of the detection result is greater than 0.85, the detection result is used for indicating that the corresponding target feature is detected. The size and the position of the target feature are obtained to form a required feature area, and the feature area is coated with a preset target mask.

In one embodiment, when the target object is a vehicle, then the target feature is a vehicle license plate. The feature recognition technique in this embodiment may employ a YOLOV5 model to recognize the license plate of the vehicle. In this embodiment, the same YOLOV5 model can be adopted as the neural network model in the target detection, so as to improve the utilization rate of hardware resources. Further, a detection category-license plate is newly added in the YOLOV5 model, license plate categories are marked on the data set, the license plate can be detected through model training, and a target mask is coated on a characteristic area of the license plate.

The libfacedetection model and YOLOV5 model in this embodiment are generated by the following steps.

And (3) data acquisition: and acquiring source images under the same or similar scene by using the looking-around camera so as to construct an actual scene data set. And (3) model building: constructing a libfacedetection neural network/YOLOV 5 neural network, and modifying the Libfacedetection neural network/YOLOV 5 neural network into a customized target detection network; such as a target detection network acting on a target object, target feature. Model training: the public data set and the actual scene data set are combined and input into a target detection network for training to obtain an initial model. Parameter adjustment: and adjusting parameters of the initial model according to the target effect to optimize the model. Model cutting: and the optimized initial model is cut by reducing parameters and the calculated amount, so that the model operation efficiency and the reasoning speed are improved. Model transplanting: and converting the cut model into an acceleration model adapting to the orin. Algorithm supplementation: performing problem supplement on missed detection false detection; model test: the model was further optimized by testing.

In one embodiment, the step of performing behavior recognition on the target object, outputting a recognition result, matching the recognition result with a preset alarm behavior, and triggering a second event when the matching is successful includes:

The behavior recognition model in this step may be defined by the neural network technique described above.

The preset alarm behavior can be that a vehicle collides with the vehicle, a person touches the vehicle, and a person attacks the vehicle. Thus, when the behavior recognition model recognizes the behavior of the target object, possible recognition results include: the vehicle/pedestrian is not contacted with the vehicle, the vehicle collides with the vehicle, the person touches the vehicle, and the person attacks the vehicle by taking other objects.

In step S104, when the detection result triggers the preset first event and second event, second monitoring data corresponding to the second event in the image data is saved, after the preset viewing authority is configured, in the step of executing the preset alarm processing flow, it is detected that a pedestrian or a vehicle exists in the target detection area, and the behavior action of the pedestrian or the vehicle is consistent with the preset alarm action, then the image data triggering the second event is saved, and is saved as the second monitoring data in the local or management platform, and the preset viewing authority is configured. That is, when an alarm event occurs, the stored second monitoring data needs to be checked by a person or platform with authority, for example, a public security system, so that the privacy security is further improved, and disputes caused by privacy are reduced. In this embodiment, when an alarm event occurs, as shown in fig. 10, the vehicle-mounted terminal contacts the vehicle-owner terminal to remind the vehicle owner of the alarm behavior of the vehicle.

Further describing, referring to the steps of the flow shown in fig. 11, the first monitoring data and the second monitoring data acquired in this embodiment may be all checked by a central control screen local to the vehicle. The second monitoring data are set to be authorized to be checked, and the second monitoring data can be checked only when the level of the preset checking authority of the vehicle supplier is met, so that the monitoring privacy is improved. The second monitoring data may be configured to be automatically stored at a local or supervisory platform and directly notified to the vehicle owner terminal to alert the vehicle owner of a second event occurring with the vehicle. The first monitoring data can be set to be in a cache state, a vehicle owner can only check the latest cached first monitoring data through the central control screen, and only when the vehicle owner performs local storage, the first monitoring data is stored, so that the first monitoring data is prevented from being stored all the time, the hard disk space is wasted, and recording to necessary data is ensured.

In the monitoring data processing method, the target event detection is performed on the image data, so that the preset event occurring in the target monitoring area can be detected. The first event and the second event are preset, when the first event is triggered, the fact that a preset target object appears in a target monitoring area of a vehicle is indicated, a certain potential safety hazard is caused to the vehicle, desensitization processing is carried out on the target object in image data at the moment, and first monitoring data which can be directly checked by a vehicle owner are generated, so that the problem of privacy leakage caused to past pedestrians/vehicles in a whistle mode of the vehicle is solved. When the detection result is adopted to trigger the first event and the second event, the second monitoring data corresponding to the second event in the image data is stored, and after the preset viewing authority is configured, the preset alarm processing flow is executed, so that the vehicle owner cannot directly view the second monitoring data and can only view the second monitoring data by using a supervision platform with the preset viewing authority, the privacy of the monitoring data is further improved, and disputes caused by the second event are reduced.

It should be understood that, although the steps in the flowchart are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the figures may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor does the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of other steps or other steps.

In one embodiment, as shown in fig. 12, there is provided a monitoring data processing apparatus including: a data acquisition module 201, an object detection module 202, a data desensitization module 203, and an alarm processing module 204, wherein:

the data acquisition module 201 is configured to acquire image data in a target area of a vehicle;

the target detection module 202 is used for detecting a target event of the image data and outputting a detection result;

the data desensitizing module 203 is configured to perform desensitizing processing on the image data when the detection result triggers a preset first event, and output first monitoring data that can be directly checked;

the alarm processing module 204 is configured to store second monitoring data corresponding to the second event in the image data when the detection result triggers the preset first event and second event, and execute a preset alarm processing procedure after configuring a preset viewing authority.

The specific limitation of the monitoring data processing apparatus may be referred to above as limitation of the monitoring data processing method, and will not be described here. The various modules in the monitoring data processing apparatus described above may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

In one embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 13. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing monitoring data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of monitoring data processing.

It will be appreciated by those skilled in the art that the structure shown in fig. 13 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the computer device to which the present application applies, and that a particular computer device may include more or fewer components than shown, or may combine some of the components, or have a different arrangement of components.

In one embodiment, a computer device is provided comprising a memory and a processor, the memory having stored therein a computer program, the processor when executing the computer program performing the steps of:

acquiring image data in a vehicle target monitoring area; detecting target events of the image data and outputting detection results; when the detection result triggers a preset first event, desensitizing the image data, and outputting first monitoring data which can be directly checked; when the detection result triggers a preset first event and a preset second event, second monitoring data corresponding to the second event in the image data are stored, and after preset viewing authority is configured, a preset alarm processing flow is executed.

In one embodiment, a computer readable storage medium is provided having a computer program stored thereon, which when executed by a processor, performs the steps of:

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above examples merely represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.

Claims

1. A method of monitoring data processing, the method comprising:

acquiring image data in a vehicle target monitoring area;

detecting the target event of the image data and outputting a detection result;

when the detection result triggers the preset first event and second event, second monitoring data corresponding to the second event in the image data are stored, and after preset viewing authority is configured, a preset alarm processing flow is executed.

2. The method of claim 1, wherein the step of acquiring image data in a target area of a vehicle comprises:

acquiring source data in the target monitoring area through a vehicle-mounted camera;

and preprocessing the source data, and caching the image data to be processed.

3. The method for processing monitoring data according to claim 1, wherein the vehicle-mounted camera is a fisheye camera;

the step of preprocessing the source data to obtain the image data to be processed comprises the following steps:

4. The method according to claim 1, wherein the step of performing target event detection on the image data and outputting a detection result comprises:

detecting a target object of the image data, and triggering the first event when a preset target object exists;

and performing behavior recognition on the target object, outputting a recognition result, matching the recognition result with a preset alarm behavior, and triggering the second event when the matching is successful.

5. The method according to claim 4, wherein the step of performing object detection on the image data, and triggering the first event when a preset object exists, includes:

if a preset target object exists, identifying the type of the target object, and marking the target object in the image data to serve as the first event triggered.

6. The method of claim 5, wherein the target object comprises a pedestrian or a vehicle; the target features comprise a pedestrian face and a vehicle license plate;

when the detection result triggers a preset first event, performing desensitization processing on the image data, and outputting first monitoring data which can be directly checked, wherein the step comprises the following steps:

acquiring the type of the target object;

and caching a preset number of desensitized image frames by using an image queue, generating the first monitoring data and outputting the first monitoring data for direct viewing.

7. The method for processing monitoring data according to claim 1, wherein the step of performing behavior recognition on the target object, outputting a recognition result, matching the recognition result with a preset alarm behavior, and triggering the second event when the matching is successful includes:

and matching the identification result with a preset alarm behavior, and triggering the second event if the matching is successful.

8. A monitoring data processing apparatus, the apparatus comprising:

the target detection module is used for detecting the target event of the image data and outputting a detection result;

and the alarm processing module is used for storing second monitoring data corresponding to the second event in the image data when the detection result triggers the preset first event and second event, and executing a preset alarm processing flow after configuring a preset viewing authority.

9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1 to 7 when the computer program is executed.

10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method of any of claims 1 to 7.