Safety monitoring system configuration method and apparatus, electronic device, and storage medium

Abstract

The present invention provides a safety monitoring system configuration method. The method comprises: receiving point cloud data of a target area acquired by a point cloud data acquisition device, and configuring the target area to be displayed as a three-dimensional space in an operation interface; receiving configuration parameters set by a user for an event, and configuring an event monitoring area in the operation interface on the basis of the configuration parameters, wherein the configuration parameters comprise an area type, bounding box parameters, and an event monitoring algorithm; and performing event monitoring within the event monitoring area on the basis of the configured event monitoring area and the point cloud data of the target area. In the present invention, by introducing a combination of a superordinate computer and the point cloud data acquisition device, along with corresponding software algorithm support, three-dimensional spatial display of the target area and event monitoring within the target area can be implemented, thereby improving the accuracy and flexibility of safety monitoring.

Description

Configuration methods, devices, electronic equipment and storage media of safety monitoring systems Technical Field

[0001] This invention relates to the field of visual technology, and in particular to a configuration method, apparatus, electronic device, and storage medium for a security monitoring system. Background Technology

[0002] A 3D safety sensor, also known as a Time-of-Flight (ToF) camera or 3D TOF camera, is primarily used for 3D security in industrial applications. Its working principle involves measuring the time it takes for light emitted from the sensor to reach an object in the detection area and reflect back to the camera to calculate distance information, thereby acquiring 3D spatial data of the scene. When a target object intrudes into the detected area, a safety control signal is output to a host computer or relay, thereby controlling the shutdown of operating equipment.

[0003] Currently, traditional security monitoring systems have fixed functions. For example, a system only monitors one type of event. When monitoring different types of events in different areas, additional hardware is required. Therefore, existing security monitoring systems have fixed configurations, poor flexibility, and cannot adapt to diverse monitoring scenarios. Summary of the Invention

[0004] This invention provides a configuration method for a safety monitoring system, which can improve the accuracy of monitoring the detection area and increase the system's working efficiency. By receiving point cloud data of a target area from a point cloud data acquisition device and configuring the target area as a three-dimensional space displayed on the operation interface, the system receives user configuration parameters for events and configures the event monitoring area on the operation interface according to these parameters. Using the configured event monitoring area and the point cloud data of the target area, event monitoring is performed on the event monitoring area. By introducing a combination of a host computer and a point cloud data acquisition device, along with corresponding software algorithm support, three-dimensional spatial display and event monitoring of the target area can be achieved, thereby improving the accuracy and flexibility of safety monitoring.

[0005] In a first aspect, embodiments of the present invention provide a configuration method for a security monitoring system. The security monitoring system includes a host computer and at least one point cloud data acquisition device. The host computer includes an interactive operating interface, and the point cloud data acquisition device is used to acquire point cloud data of a target area. The method includes the following steps: receiving point cloud data of the target area from the point cloud data acquisition device and configuring the target area as a three-dimensional space displayed in the operating interface; receiving configuration parameters for events from the user and configuring an event monitoring area in the operating interface according to the configuration parameters, wherein the configuration parameters include area type, area bounding box parameters, and event monitoring algorithm; and performing event monitoring on the event monitoring area based on the configured event monitoring area and the point cloud data of the target area.

[0006] Optionally, configuring the event monitoring area in the operation interface according to the configuration parameters includes: matching an event monitoring algorithm corresponding to the region type in the event monitoring algorithm library, wherein different region types correspond to different event monitoring algorithms in the event monitoring algorithm library; drawing a region bounding box in the three-dimensional space according to the region bounding box parameters; associating the event monitoring algorithm with the region bounding box, and configuring the event monitoring area in the operation interface.

[0007] Optionally, the region bounding box parameters include bounding box length, bounding box width, and bounding box height. Drawing the region bounding box in the three-dimensional space based on the region bounding box parameters includes: determining the three-dimensional coordinates of the target's location point in the three-dimensional space; and drawing the region bounding box in the three-dimensional space based on the bounding box length, the bounding box width, the bounding box height, and the three-dimensional coordinates of the location point, with each target corresponding to one region bounding box.

[0008] Optionally, the step of monitoring events in the event monitoring area based on the configured point cloud data of the event monitoring area and the target area includes: receiving a user's selection instruction and selecting at least one target event monitoring area from the configured event monitoring area; and monitoring events in the target event monitoring area based on the point cloud data of the target area and in conjunction with the event monitoring algorithm corresponding to the target event monitoring area.

[0009] Optionally, event monitoring is performed on the event monitoring area based on the configured point cloud data of the event monitoring area and the target area, including: receiving a user's switching instruction and determining a target event monitoring area in the event monitoring area; switching the current event monitoring area to the target event monitoring area; and performing event monitoring on the target event monitoring area based on the point cloud data of the target area and in conjunction with the event monitoring algorithm corresponding to the target event monitoring area.

[0010] Optionally, the region type includes at least one of shielded region, protected region, alarm region, contour analysis region, attention region, and AI analysis region. The step of monitoring events in the event monitoring region based on the configured event monitoring region and the point cloud data of the target region includes: If the configured event monitoring region is a protected region or alarm region, then based on each two-dimensional projection plane of the protected region or alarm region, determining whether an intrusion event exists in the protected region or alarm region; if an intrusion event exists in the protected region or alarm region, then outputting a first security control signal; If the configured event monitoring region is a attention region, then based on the depth dispersion rate of each point cloud in the attention region, determining whether an intrusion event exists in the attention region; if an intrusion event exists in the attention region, then outputting a second security control signal; If the configured event monitoring region is a contour analysis region, then based on the distance change between the reference frame point cloud data and the current frame point cloud data in the contour analysis region, determining whether a contour change event exists in the contour analysis region; if a contour change event exists in the contour analysis region, then outputting a third security control signal. If the configured event monitoring area is an AI analysis area, then based on intensity image data and 3D point cloud data, it is determined whether there is a target person in the AI ​​analysis area. If there is a target person in the AI ​​analysis area, then a fourth security control signal is output.

[0011] Secondly, embodiments of the present invention also provide a configuration device for a security monitoring system. The security monitoring system includes a host computer and at least one point cloud data acquisition device. The host computer includes an interactive operating interface, and the point cloud data acquisition device is used to acquire point cloud data of a target area. The configuration device for the security monitoring system includes: a first configuration module, used to receive the point cloud data of the target area from the point cloud data acquisition device and configure the target area as a three-dimensional space displayed in the operating interface; a second configuration module, used to receive user configuration parameters for events and configure an event monitoring area in the operating interface according to the configuration parameters, the configuration parameters including area type, area bounding box parameters, and event monitoring algorithm; and an event monitoring module, used to perform event monitoring on the event monitoring area based on the configured event monitoring area and the point cloud data of the target area.

[0012] Thirdly, embodiments of the present invention provide a security monitoring system, the security monitoring system including a host computer and at least one point cloud data acquisition device, the host computer including an interactive operation interface, the point cloud data acquisition device being used to acquire point cloud data of a target area, and the host computer being used to execute the steps in the configuration method of the security monitoring system provided in the embodiments of the present invention.

[0013] Fourthly, embodiments of the present invention provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the configuration method of the security monitoring system provided in the embodiments of the present invention.

[0014] Fifthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the configuration method of the security monitoring system provided in the embodiments of the present invention.

[0015] In this embodiment of the invention, point cloud data of a target area is received from a point cloud data acquisition device, and the target area is configured as a three-dimensional space and displayed in the operation interface. User configuration parameters for events are received, and an event monitoring area is configured in the operation interface according to these parameters. The configuration parameters include area type, area bounding box parameters, and event monitoring algorithm. Event monitoring is performed on the event monitoring area based on the configured event monitoring area and the point cloud data of the target area. This invention, by receiving point cloud data of a target area from a point cloud data acquisition device and configuring the target area as a three-dimensional space in the operation interface, receiving user configuration parameters for events, configuring an event monitoring area in the operation interface according to these parameters, and utilizing the configured event monitoring area and the point cloud data of the target area to perform event monitoring, achieves three-dimensional spatial display and event monitoring of the target area through the introduction of a host computer and a point cloud data acquisition device, along with corresponding software algorithm support. This improves the accuracy and flexibility of security monitoring. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 is a flowchart of a configuration method for a security monitoring system provided in an embodiment of the present invention; Figure 2 is a structural schematic diagram of a configuration device for a security monitoring system provided in an embodiment of the present invention; Figure 3 is a structural schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention provides a security monitoring system, which includes a host computer and at least one point cloud data acquisition device. The host computer includes an interactive interface, the point cloud data acquisition device is used to collect point cloud data of a target area, and the host computer is used to execute the steps in the configuration method of the security monitoring system provided in this invention.

[0020] Specifically, the aforementioned safety monitoring system includes a point cloud acquisition module, a signal input module, a configuration module, a processing module, and an output module.

[0021] The aforementioned point cloud acquisition module is configured to acquire point cloud data of the target area, and the aforementioned point cloud acquisition module is located inside the point cloud data acquisition device.

[0022] The aforementioned signal input module, configuration module, processing module, and output module are located in the host computer.

[0023] The aforementioned signal input module is configured to receive area group event switching signals, external trigger signals, reset signals, EDM signals, wake-up signals, power inputs, etc.

[0024] The above configuration modules include an account management module, a network configuration module, a regional group configuration module, a regional configuration module, an input configuration module, an output configuration module, a working mode configuration module, a sleep / wake-up function configuration module, a data acquisition terminal parameter configuration module, and an event configuration module.

[0025] The account management module is used to add or delete accounts, or to manage account permissions.

[0026] The region group configuration module is used to configure the addition and removal of region groups, the addition and removal of regions within a region group, and region configuration. A region group can be understood as a collection of regions of the same type, with each region group corresponding to one region type. Of course, multiple regions of different types can also be manually combined into a region group. Regions within a region group can be added or removed using the region configuration module.

[0027] The area configuration module is used for area drawing, area type (shielded area / alarm area / protected area / contour analysis area / ROI area / AI analysis area), automatic adaptation of output OSSD (Output Switching Signal Device) / IO type, configuration of specific OSSD / IO output interfaces, and selection of output N / P type.

[0028] The input configuration module is used for configuring OSSD reset mode (manual / automatic), input delay, and EDM monitoring time. It is also used for regional group event input signal acquisition, input delay processing, and input signal verification (parity check, etc.). Furthermore, it is used for interrupt event input signals, specifically wake-up information (entering normal working state) and external trigger signals (only effective when the camera mode is configured for external trigger mode, and each external trigger only acquires one 3D point cloud instance).

[0029] The output configuration module is used for OSSD reset mode configuration.

[0030] The working mode configuration is used to configure the working mode, which can include live mode, soft trigger, and hardware trigger.

[0031] The hibernation / wake-up function configuration module is used to configure the hibernation / wake-up function.

[0032] The acquisition end parameter configuration module is used to configure the frame rate, camera exposure time, time synchronization, and shooting mode of the point cloud acquisition module. Specifically, the configuration of the 3D point cloud acquisition module can include frame rate, camera exposure time, time synchronization, shooting mode, filtering, etc. After configuring the point cloud acquisition module, point cloud data of the corresponding area can be acquired.

[0033] The event configuration module is used for event storage configuration and event type configuration (event / hibernation).

[0034] The processing modules may include an alarm zone protection processing module (AI / traditional), a protected area protection processing module (AI / traditional), a contour recognition zone protection processing module, and an ROI (area of ​​interest) protection processing module.

[0035] The alarm area is configured with resolution and resampling times. The resolution is configured to automatically calculate the furthest alarm distance corresponding to a smaller size threshold or object point count threshold, and the resampling times are N. Alarm area intrusion detection (alarm area detection algorithm): Step 1: Check if there is a point cloud within the alarm area. If there is no point cloud, then there is no intrusion; the IO outputs no intrusion signal. Otherwise, proceed to the next step.

[0036] Step 2: Point cloud filtering. If there is no point cloud, there is no intrusion, and the IO output shows no intrusion signal; otherwise, proceed to the next step.

[0037] Step 3: Point cloud clustering processing. Calculate the object size (or number of points). If the object size or the maximum number of points is less than the threshold, there is no intrusion, and the I / O output shows no intrusion signal; otherwise, proceed to the next step.

[0038] Step 4: If the number of consecutive N sampling intrusions exceeds M1, the IO outputs an intrusion signal.

[0039] Configure the resolution and resampling times for the protected area. The resolution is configured to automatically calculate the maximum alarm distance corresponding to the smaller size threshold or object point count threshold, and the resampling times are N. Protected area intrusion detection (protected area detection algorithm): Step 1: Check if there is a point cloud within the protected area. If there is no point cloud, there is no intrusion, and OSSD outputs a safety signal (no intrusion); otherwise, proceed to the next step.

[0040] Step 2: Point cloud filtering. If there is no point cloud, there is no intrusion, and OSSD outputs a security signal (no intrusion). Otherwise, proceed to the next step.

[0041] Step 3: Point cloud clustering processing, calculate object size (or number of points). If the object size or the maximum number of points is less than the threshold, there is no intrusion, and OSSD outputs a security signal (no intrusion). Otherwise, proceed to the next step.

[0042] Step 4: If the number of consecutive intrusions exceeds M2 after N consecutive samplings, the OSSD will output a danger signal (intrusion has occurred).

[0043] Step 5: When the OSSD pre-output is a danger signal, and a reset signal is received and there is no intrusion in the current frame, the OSSD reset outputs a safety signal (no intrusion).

[0044] AI analysis area configuration: Add / remove AI analysis types: human body, head, hand, arm, leg, robotic arm, AGV, safety helmet, safety clothing, etc.; automatically calculate the farthest analysis distance; configure the access attribute (allow / stop) for target type identification; resampling times N. AI analysis area intrusion detection (AI area detection algorithm): Step 1: Perform AI analysis within the analysis area. If there is no point cloud in the analysis area, there is no intrusion, and the OSSD outputs a security signal (no prohibited target); otherwise, proceed to the next step.

[0045] Step 2: Point cloud filtering. If there is no point cloud, there is no intrusion, and OSSD outputs a security signal (no prohibited target). Otherwise, proceed to the next step.

[0046] Step 3: Target identification and classification. If all targets are passable targets, output "No intrusion" and OSSD outputs a security signal (no prohibited targets). If there are prohibited targets, proceed to the next step.

[0047] Step 4: If M3 intrusions are detected in N consecutive samples, the OSSD will output a danger signal (a prohibited target exists).

[0048] Step 5: When the OSSD pre-output is a danger signal, and a reset signal is received and there is no prohibited target in the current frame, the OSSD reset outputs a safety signal (no prohibited target).

[0049] Contour analysis region configuration includes reference contour configuration, contour correlation threshold, and resampling times. Contour analysis region anomaly detection (contour analysis region detection algorithm to prevent falls, etc.): Step 1, reference contour feature extraction.

[0050] Step 2: Extract contour features of the current frame.

[0051] Step 3: Analyze the correlation between the current frame contour and the reference contour (e.g., contour matching), and combine the correlation threshold to determine whether the contour has undergone unacceptable changes.

[0052] Step 4: If the contour change is acceptable, then the contour is normal and the OSSD outputs a safety signal. If the contour change is unacceptable, then the contour is abnormal and the OSSD outputs a danger signal.

[0053] Step 5: When the OSSD pre-output is a danger signal, and a reset signal is received and there is no prohibited target in the current frame, the OSSD reset outputs a safety signal (no prohibited target).

[0054] ROI region configuration includes adding / removing reference ROIs, setting the size of reference ROI regions, configuring the safe coordinate range of reference ROIs (including the coordinate range of x, Y, and Z), and adjusting the repetition precision coefficient of reference ROI region points. ROI region anomaly detection (preventing changes in key points, corresponding to the area of ​​interest detection algorithm): Step 1: For each ROI region, extract the coordinate data of all points in that ROI and calculate the centroid coordinates and the point repetition precision threshold.

[0055] Step 2: In the current frame, for each ROI point, calculate the centroid coordinates and repeatability of all points in that ROI.

[0056] Step 3: Determine whether the centroid coordinates of all ROIs are within the safe coordinate range. If the centroid coordinates of any ROI are within the safe coordinate range of the ROI region or the repeatability of any ROI does not exceed the repeatability threshold of the ROI region, it means that there are no unacceptable changes in all ROI regions, all ROIs are abnormal, and OSSD outputs a safe signal.

[0057] Step 4: If the centroid coordinates of any ROI exceed the safe coordinate range of that ROI point, or the repeatability of any ROI exceeds the repeatability threshold of that ROI area point, it indicates an unacceptable change, and the ROI has an abnormal OSSD output danger signal.

[0058] Step 5: When the OSSD pre-output is a danger signal, a reset signal is received and all ROIs in the current frame are normal, the OSSD reset outputs a safety signal.

[0059] The shielded area is configured with dimensions (center coordinates, length, width, and height). Shielded areas are typically used in conjunction with alarm zones, protected zones, and AI analysis zones. Within the alarm zone, no intrusion detection is required within the shielded area. If an intrusion occurs within the alarm zone but outside the shielded area, the IO outputs an intrusion alarm signal. Within the protected zone, no intrusion detection is required within the shielded area. If an intrusion occurs within the protected zone but outside the shielded area, it indicates a danger, and the OSSD outputs a danger signal. Within the AI ​​analysis zone, no specific target intrusion detection is required within the shielded area. If a specific target is within the AI ​​analysis zone but outside the shielded area, it indicates a danger, and the OSSD outputs a danger signal.

[0060] The above output modules are used for OSSD signals, IO signals, and data output. IO signal output can be configured as NPN / PNP output or adapted to the input voltage. OSSD signal output can be configured to support NPN / PNP output, OSSD and IO conversion configuration, OSSD heartbeat, and adapt to the input voltage. Data output modes can be configured as point cloud output mode, depth map output mode, device information, etc.

[0061] In one possible embodiment, the host computer is a single-channel (single-processor) system that can process application schemes based on configuration and configure the corresponding IO or OSSD outputs according to the configuration and event processing results.

[0062] In one possible embodiment, the host computer is dual-channel (dual-processor), with a dual-processor architecture, each channel performing self-tests. The dual-processor architecture allows for independent operation of each channel, with cross-checking. OSSD heartbeat functionality is included. Dual-processor cross-checking is also performed. The dual processors and OSSD are cross-checked.

[0063] As shown in Figure 1, Figure 1 is a flowchart of a configuration method for a safety monitoring system provided by an embodiment of the present invention. The configuration method for the safety monitoring system includes the following steps: 101. Receive point cloud data of a target area from a point cloud data acquisition device, and configure the target area as a three-dimensional space to be displayed in the operation interface.

[0064] In this embodiment of the invention, the safety monitoring system includes a host computer and at least one point cloud data acquisition device. The point cloud data acquisition device may be a 3D camera, a depth camera, or a time-of-flight camera, such as a stereo safety sensor. The host computer includes an interactive interface, and the point cloud data acquisition device is used to acquire point cloud data of the target area.

[0065] The configuration method of the aforementioned safety monitoring system is mainly used in the host computer of the safety monitoring system. The point cloud data mentioned above can be understood as the three-dimensional coordinate data (x, y, z) of each pixel in the depth map collected by the stereo safety sensor.

[0066] The aforementioned point cloud data can also be referred to as 3D point cloud data. In this embodiment, the acquisition of point cloud data can be based on the iTOF ranging principle. Through CW modulation driven by VCSEL, the VCSEL modulates light with a frequency of f. Because it is necessary to analyze the phase difference between the reflected light and the emitted light to calculate the distance, in order to avoid multiple solutions, the distance of the target within 1 / 2 wavelength can only be measured. That is, the maximum ranging is 1 / 2 wavelength of the modulated light wave. In this example, two frequencies, 15MHz and 120MHz, are used. The theoretical maximum detection distance at the 15MHz frequency is 10m, and the theoretical maximum detection distance at the 120MHz frequency is 1.25m. The theoretical maximum detection distance of the dual-frequency rotation is the least common multiple of the theoretical maximum detection distances corresponding to the two frequencies, which is the least common multiple of 1.5m and 10m. That is, the theoretical maximum detection distance under the condition of rotating between the two frequencies of 15MHz and 120MHz is 10m.

[0067] The depth map distortion problem is addressed through camera intrinsic parameter calibration. The intrinsic parameter calibration module uses a common computer vision model, employing the "pinhole + distortion" model to characterize imaging system parameters such as lens distortion and principal point shift. These parameters include: 1. Principal point c. x c y , representing the pixel coordinates of the intersection point between the lens optical axis and the imaging target surface.

[0068] 2. Focal length f x f y This represents the ratio of lens focal length to pixel size.

[0069] 3. Distortion parameters k1, k2, k 3；Let k1, k2, and k3 represent the radial distortion of the lens, and p1 and p2 represent the tangential distortion. In the intrinsic parameter calibration module, the calibration method is used to solve for the model parameters. The camera is controlled to acquire intensity images of the calibration plate. Given the coordinates of the marker points on the calibration plate, features are extracted from the marker points in the image to obtain their image coordinates. An objective function is established and optimized using the LM method. Finally, the principal point c is obtained. x c y Focal length f x f y The calibration results are recorded in the camera, along with the distortion parameters k1, k2, k3, p1, and p2.

[0070] During the depth calibration (phase calibration) process, the measurement accuracy of TOF is affected by many factors. The depth calibration module mainly considers three aspects: 1. When measuring an object at zero distance, the delay difference between the phase of the signal received by RX (RF receiver) and the zero-phase signal when transmitting a zero-phase signal, i.e., zero drift.

[0071] 2. Zero drift, which characterizes the difference in initial exposure time between different pixels, i.e., fixed phase template noise.

[0072] 3. The deviation between the light waveform emitted by the TX (radio frequency transmitter) and the cosine signal causes harmonic errors that vary with distance.

[0073] The absolute error introduced by the above items is usually on the order of centimeters. To ensure the accuracy of the ranging results, the above errors of each stereo safety sensor must be calibrated one by one at the factory.

[0074] To calibrate the aforementioned errors, the true distance to the object under test needs to be known, and TOF images need to be acquired simultaneously to calculate the actual measured distance. An error model is then established between these two methods, and the model parameters are solved. Specifically, a stereo safety sensor is placed on a track platform, and TOF data is acquired by capturing images of the 90% diffuse reflection plane at different distances. Using the distance from the stereo safety sensor to the diffuse reflection plane provided by the track platform, and the intrinsic parameter calibration model, the true distance of each pixel to the diffuse reflection plane is calculated. A depth calibration algorithm is then called to process the acquired TOF data, calculate the actual measured distance, and optimize the model parameters.

[0075] The depth calculation algorithm performs calculations from the raw image to the tap image, from the tap image to the intensity image, amplitude image, phase image, from the phase image to the distance image, from the distance image to the depth image, and from the depth image to the point cloud image. The depth calculation algorithm also has the following two functions: 1. During the phase calculation process, the depth calculation algorithm needs to call the calibration results to compensate for the calculated phase, thereby eliminating systematic errors and obtaining higher measurement accuracy.

[0076] 2. Because the 3D safety sensor operates in dual-frequency mode, the depth calculation algorithm needs to fuse the dual-frequency data to output a higher quality measurement result.

[0077] The depth filtering algorithm operates at different stages of the depth computation algorithm, processing intermediate quantities, filtering out errors and low signal-to-noise ratio results, and ensuring the quality of the output results. The depth filtering algorithm consists of five modules: 1. Temporal filtering module: Averaging the depth map across multiple frames to improve the signal-to-noise ratio. The temporal filtering algorithm can be enabled or disabled, and the fusion weights for the current frame can be set.

[0078] 2. Spatial Filtering Module: Performs spatial Gaussian filtering on the 1Q image to improve the signal-to-noise ratio. The spatial filtering algorithm can be enabled or disabled, and the size of the Gaussian filter window can be set.

[0079] 3. Amplitude Filtering Module: This module filters out pixels with amplitude values ​​below a set threshold. The amplitude filtering algorithm can be enabled or disabled, and the amplitude filtering threshold can be set.

[0080] 4. Fusion Error Filtering Module: Calculates the dual-frequency fusion error. When the phase value is greater than a set value, the corresponding pixel depth is filtered out. The fusion error filtering algorithm can be enabled or disabled, and the value of the interval can be set.

[0081] 5. Flying Point Filtering Module: Converts depth to point cloud, calculates the distance between a point and its surrounding points, and filters out the corresponding pixel depth when the minimum distance is greater than a set value. The function of flying point filtering can be turned on and off, and the distance value can be set.

[0082] The following is an example of point cloud generation: Through CW modulation driven by VCSEL, the VCSEL modulates light at two frequencies of 15MHz and 120MHz in a time-division manner. The light at the two frequencies is exposed separately under the illumination conditions to obtain the energy integral map of four phases at 0°, 180°, 90° and 270° at each frequency. Therefore, a total of 8 energy integral maps need to be collected: Rawdata120_0: 0° phase energy integral map under 120MHz frequency modulated light.

[0083] Rawdata120_90: 90° phase energy integral plot under 120MHz frequency modulated light.

[0084] Rawdata120_180: 180° phase energy integral plot under 120MHz frequency modulated light.

[0085] Rawdata120_270: 270° phase energy integral plot under 120MHz frequency modulated light.

[0086] Rawdata15_0: 0° phase energy integral plot under 15MHz frequency modulated light.

[0087] Rawdata15_90: 90° phase energy integral plot under 15MHz frequency modulated light.

[0088] Rawdata15_180: 180° phase energy integral plot under 15MHz frequency modulated light.

[0089] Rawdata15_270: 270° phase energy integral plot under 15MHz frequency modulated light.

[0090] Dual-frequency rotation: 1. First, 120MHz is used to take 4 pictures. Each picture has 2 phases (2 taps). The exposure time of each picture is the set exposure value (e.g., if the exposure is set to 1000us, 4 pictures will take 4000us. There is also a waiting time between two pictures, with a minimum of about 4ms).

[0091] 2. Next, at 15MHz, take 4 images. Each image has 2 phases (2 taps). The exposure time for each image is the set exposure value (e.g., if the exposure is set to 1000us, 4 images would take 4000us. There is also a waiting time between two images, with a minimum of about 4ms).

[0092] 3. For each modulation frequency, the tap phase order of the 4 images is: Tap order (0°, 180°) → (90°, 270°) → (180°, 0°) → (270°, 90°).

[0093] The value of each pixel in the raw data at a certain frequency and phase represents the integral of the reflected energy of the target in the corresponding direction. For the same pixel, the integral energy Q0, Q1, Q2, and Q3 of the four phases are used to calculate the energy. 90 Q 180 Q 270 The phase offset value at that point can then be calculated.

[0094]

[0095] Q0 represents the energy integral of the pixel at phase 0.

[0096] Q 90 The energy integral representing the 0 phase of this pixel.

[0097] Q 180 The energy integral representing the 0 phase of this pixel.

[0098] Q 270 The energy integral representing the 0 phase of this pixel.

[0099] Based on the phase of this pixel The detection range at that frequency can be calculated, and the corresponding distances d1 and d2 can be calculated for the two frequencies respectively.

[0100]

[0101] The final distance is obtained by dual-frequency calculation.

[0102] (1) Calculate the combined frequency of the two frequencies.

[0103] f_max is the greatest common divisor of frequencies f1 and f2: when f1 = 15 MHz and f2 = 120 MHz, f_max = 15 MHz (2) Calculate M_f1 and M_f2 as follows: M_f1 = f1 / f_max M_f2 = f2 / f_max

[0104] Calculate A_f1 and A_f2 using the following formulas:

[0105] ω is calculated using the following formula:

[0106] The final dual-frequency distance is: d = d1·ω + d2·(1-ω)

[0107] Depth map generation: ① Based on the phase calibration results, the relationship between the actual detected phase and the theoretical phase is obtained; ② Based on the current actual detected phase, the current true phase can be calculated, and the calculated distance, which is the actual detected distance value, can be obtained.

[0108] Depth map converted into point cloud map: (1) According to the calibration of camera intrinsic parameters, the azimuth angle corresponding to each pixel can be obtained.

[0109] (2) The three-dimensional coordinates x, y, and z can be calculated based on the depth d and the azimuth angle.

[0110] The aforementioned three-dimensional space can be understood as a three-dimensional coordinate system used to describe the position and shape of an object. The three-dimensional space consists of three dimensions: length, width, and height. The first three-dimensional space can be the camera monitoring area. This monitoring area can be defined as a cube centered at the camera's optical center O, with vertices ABCDEFGH. The six square pyramids within the cube (OABEH, OCDFG, OABCD, OEFGH, OADEF, OBCGH) serve as sub-regions.

[0111] The system receives point cloud data of the target area from the point cloud data acquisition device, uploads the point cloud data of the target area to the host computer, and displays it in a three-dimensional space through the host computer's operation interface.

[0112] The aforementioned user interface refers to the graphical interface on software or equipment that allows users to interact with the system.

[0113] The aforementioned three-dimensional space is used to describe the position and shape of an object. The three-dimensional space consists of three dimensions: length, width, and height.

[0114] It should be noted that displaying the target area in a three-dimensional space in the user interface can give the target area a three-dimensional effect, allowing users to understand and operate it more intuitively.

[0115] 102. Receive the user's configuration parameters for the event, and configure the event monitoring area in the operation interface according to the configuration parameters.

[0116] In this embodiment of the invention, the configuration parameters include region type, region bounding box parameters, and event monitoring algorithm. The region type can be a shielded region / alarm zone / protected zone / contour analysis zone / ROI region / AI analysis zone, etc. The region bounding box parameters can be geometric attributes such as the coordinates, size, and shape of the detection region. The event monitoring algorithm is used to identify specific events in the video, such as whether an object enters, leaves, or stays at a specific location within the monitoring area.

[0117] Based on the region type, the system can match the corresponding monitoring algorithm from the event monitoring algorithm library. Then, based on the region bounding box parameters, it draws the region bounding box in 3D space and associates the event monitoring algorithm with the bounding box. The event monitoring region can be configured in the user interface. Event monitoring algorithms can include shielded area monitoring algorithms, alarm area monitoring algorithms, protected area monitoring algorithms, contour analysis area monitoring algorithms, ROI area monitoring algorithms, and AI analysis area monitoring algorithms, etc. Specifically, the shielded area monitoring algorithm corresponds to the shielded area, the alarm area monitoring algorithm corresponds to the alarm area, the protected area monitoring algorithm corresponds to the protected area, the contour analysis area monitoring algorithm corresponds to the contour analysis area, the ROI area monitoring algorithm corresponds to the ROI area, and the AI ​​analysis area monitoring algorithm corresponds to the ROI area.

[0118] The aforementioned event monitoring algorithm library can be a collection of various algorithms used to detect and identify specific events.

[0119] It should be noted that the event monitoring algorithms are designed for different events. Depending on the area type, the algorithm that best matches the time can be selected from the time monitoring algorithm library. For example, if the area type is an alarm zone or a protection zone, an algorithm specifically designed to monitor equipment malfunctions or unsafe employee behavior will be selected; if the area type is an AI analysis zone, an algorithm designed to monitor crowd density or abnormal gathering behavior will be selected.

[0120] 103. Based on the configured event monitoring area and the point cloud data of the target area, perform event monitoring in the event monitoring area.

[0121] In this embodiment of the invention, the configured event monitoring area is used to monitor events within that area.

[0122] Event monitoring can be performed on the configured event monitoring area and the point cloud data of the target area. This event monitoring includes using event monitoring algorithms to analyze point cloud data, identify potential event indicators, and generate corresponding alarms or reports. By monitoring the target event monitoring area, event activity within the target area can be monitored in real time, effectively improving the accuracy of target event monitoring.

[0123] In this embodiment of the invention, point cloud data of a target area is received from a point cloud data acquisition device, and the target area is configured as a three-dimensional space and displayed in the operation interface. User configuration parameters for events are received, and an event monitoring area is configured in the operation interface according to these parameters. The configuration parameters include area type, area bounding box parameters, and event monitoring algorithm. Event monitoring is performed on the event monitoring area based on the configured event monitoring area and the point cloud data of the target area. This invention, by receiving point cloud data of a target area from a point cloud data acquisition device and configuring the target area as a three-dimensional space in the operation interface, receiving user configuration parameters for events, configuring an event monitoring area in the operation interface according to these parameters, and utilizing the configured event monitoring area and the point cloud data of the target area to perform event monitoring, achieves three-dimensional spatial display and event monitoring of the target area through the introduction of a host computer and a point cloud data acquisition device, along with corresponding software algorithm support. This improves the accuracy and flexibility of security monitoring.

[0124] It is understood that in the specific implementation of this application, data such as point cloud data, configuration data, and device data are involved. When the embodiments in this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0125] Optionally, in the step of configuring the event monitoring area in the operation interface according to the configuration parameters, the event monitoring algorithm corresponding to the region type can be matched in the event monitoring algorithm library according to the region type. Different region types correspond to different event monitoring algorithms in the event monitoring algorithm library. The region box is drawn in three-dimensional space according to the region box parameters. The event monitoring algorithm is associated with the region box, and the event monitoring area is configured in the operation interface.

[0126] In this embodiment of the invention, the above-mentioned area type may be a shielded area / alarm area / protected area / contour analysis area / ROI area / AI analysis area, etc.

[0127] The aforementioned event monitoring algorithm library can be understood as a collection of algorithms used to detect and identify specific events, for analyzing and recognizing different events. Within this library, different region types correspond to different event monitoring algorithms.

[0128] For example, if the area type is an alarm zone, the alarm zone intrusion detection algorithm is used to detect intrusion in the alarm zone: step 1, whether there is a point cloud in the alarm zone. If there is no point cloud, then there is no intrusion IO output and no intrusion signal is output; otherwise, proceed to the next step.

[0129] Step 2: Point cloud filtering. If there is no point cloud, there is no intrusion, and I0 outputs no intrusion signal; otherwise, proceed to the next step.

[0130] Step 3: Point cloud clustering processing. Calculate the object size (or number of points). If the object size or the maximum number of points is less than the threshold, there is no intrusion, and the I / O output shows no intrusion signal; otherwise, proceed to the next step.

[0131] Step 4: If the number of consecutive N sampling intrusions exceeds M1, the IO outputs an intrusion signal.

[0132] It's important to note that event monitoring algorithms are designed for different events or behavioral patterns, such as algorithms for equipment malfunctions, object intrusion, and personnel trespassing. Based on the area type, the most suitable algorithm is selected from the event monitoring algorithm library. Specifically, event monitoring algorithms can include shielded area monitoring algorithms, alarm area monitoring algorithms, protected area monitoring algorithms, contour analysis area monitoring algorithms, ROI area monitoring algorithms, and AI analysis area monitoring algorithms, etc. Among these, shielded area monitoring algorithms correspond to shielded areas, alarm area monitoring algorithms to alarm areas, protected area monitoring algorithms to protected areas, contour analysis area monitoring algorithms to contour analysis areas, ROI area monitoring algorithms to ROI areas, and AI analysis area monitoring algorithms to ROI areas.

[0133] The aforementioned region bounding box parameters can be geometric attributes such as the coordinates, size, and shape of the detection region.

[0134] Specifically, a bounding box can be drawn in 3D space using parameters such as position, size, and shape. Then, the event monitoring algorithm is associated with this bounding box. Through this association, the event monitoring area can be configured in the user interface. This allows users to visually view the monitored area and the events occurring within it.

[0135] Optionally, the region bounding box parameters include the bounding box length, bounding box width, and bounding box height. In the step of drawing the region bounding box in three-dimensional space based on the region bounding box parameters, the three-dimensional coordinates of the target's positioning point can be determined in three-dimensional space; and the region bounding box can be drawn in three-dimensional space based on the bounding box length, bounding box width, bounding box height, and the three-dimensional coordinate values ​​of the positioning point.

[0136] In this embodiment of the invention, each target corresponds to a region box. The aforementioned targets may be devices such as robots or robotic arms.

[0137] First, determine the three-dimensional coordinates of the target location point in three-dimensional space. Then, using the frame length, frame width, frame height, and the three-dimensional coordinates of the location point, draw the region box in three-dimensional space. Each target corresponds to a region box, and these region boxes can be accurately located and drawn in three-dimensional space.

[0138] Optionally, in the step of monitoring events in the event monitoring area based on the point cloud data of the configured event monitoring area and the target area, a user selection instruction can be received to select at least one target event monitoring area from the configured event monitoring area; and event monitoring can be performed on the target event monitoring area based on the point cloud data of the target area and the event monitoring algorithm corresponding to the target event monitoring area.

[0139] In this embodiment of the invention, the user's selection instruction can be a feature selection or operation made by the user in the system, such as clicking a button or selecting an option.

[0140] In one possible implementation, users can select events by clicking buttons on the interface, choosing drop-down menu options, or using other graphical representations. The user's selection instruction can be clicking a specific area marker, or entering a specific area number or name. Upon receiving the user's selection instruction, the system will select at least one target event monitoring area from the configured event monitoring areas. The target event monitoring area can be a specific location that the user deems to pose a high security risk or require special attention. For each selected target event monitoring area, point cloud data of the target area can be used, combined with the corresponding event monitoring algorithm, to perform event monitoring in the target event monitoring area.

[0141] Specifically, in this embodiment of the invention, a region group can be understood as a collection of event monitoring regions of the same region type. Each region group corresponds to an event monitoring region of one region type. Of course, multiple event monitoring regions of different region types can also be manually combined into a region group. Regions within a region group can be added or removed through the region configuration module.

[0142] When receiving a region group event switching signal, the corresponding region group (event monitoring region) is switched according to the state of the IO. In this embodiment of the invention, there are a total of four general-purpose IO input signals. Three IOs have eight combinations of different input levels to each IO, corresponding to the events of the eight region groups. Another IO is used for parity checking, counting the number of high-level events on the three pins of IO1-3. If the number of high-level events is odd, the IO needs to be set to high, and if the number of high-level events is even, it needs to be set to low.

[0143] For example, by default, these pins are not connected to signals. In the selected event group (111-->7), the number of 1s in IO1-3 is counted. If the count is odd, the IO needs to be set to 1; if even, it is set to 0. IO4 is for parity checking. In event group (111-->7), all IO1-3 are 1, so the count of 1s is odd, and IO4 is set to 1. In the selected event group (011-->3), there are 2 1s and 1 0 in IO1-3, meaning one IO is 0, so the count of 1s is even, and IO4 is set to 0.

[0144] A single thread polls and checks four I / O ports. IO4 checks I / O1-3. If the check is successful, the value formed by I / O1-3 is used as the current selected event group. For example, if (111-->7), the event group is the region group marked 7. If (011-->3), the event group is the event group marked 3.

[0145] Switching strategy: The current detection area group is determined by either the value input from the host computer or the external I / O input. The choice of method is configured through a parameter. Think of it this way: each event monitoring area has a unique number. The system finds the corresponding number based on the value input from the host computer or the external I / O input and switches accordingly. Typically, switching is done through the four positions mentioned above: IO1-IO4.

[0146] Optionally, in the step of monitoring events in the event monitoring area based on the configured point cloud data of the event monitoring area and the target area, the user's switching command can be received to determine the target event monitoring area in the event monitoring area; the current event monitoring area can be switched to the target event monitoring area; and event monitoring can be performed in the target event monitoring area based on the point cloud data of the target area and the event monitoring algorithm corresponding to the target event monitoring area.

[0147] In this embodiment of the invention, after receiving the user's switching instruction, the system will determine the target event monitoring area in the event monitoring area. This means that the system will adjust its monitoring focus according to the instruction to ensure that the key event area of ​​the user's switching is monitored.

[0148] The current event monitoring area mentioned above can be understood as the area currently used to monitor a specific event. The target time monitoring area mentioned above can be understood as the monitoring area that is expected to be reached.

[0149] The aforementioned event monitoring algorithm is used to identify specific events in a video, such as whether an object enters, leaves, or stays in a specific location within the monitoring area.

[0150] Furthermore, based on the point cloud data of the target area and combined with the event monitoring algorithm corresponding to the target event monitoring area, event monitoring operations can be performed in the target event monitoring area.

[0151] The aforementioned event monitoring includes using event monitoring algorithms to analyze point cloud data, identify potential event indicators, and generate corresponding alarms or reports. By monitoring the target event monitoring area, event activities within the target area can be monitored in real time, effectively improving the accuracy of target event monitoring.

[0152] In one possible embodiment, in the step of monitoring events in the event monitoring area based on the configured point cloud data of the event monitoring area and the target area, the monitoring activity of different area types at different time periods can be determined according to the monitoring results of historical event monitoring areas. Based on the monitoring activity at different time periods, target event monitoring areas for different time periods are determined within the event monitoring area. When the corresponding time period is reached, the current event monitoring area is switched to the target event monitoring area for the corresponding time period. Based on the point cloud data of the target area and combined with the event monitoring algorithm corresponding to the target event monitoring area, event monitoring is performed on the target event monitoring area.

[0153] The aforementioned monitoring activity level can be calculated using the following formula:

[0154] Among them, the above r j T represents the monitoring activity of a certain region type in the j-th time period, H represents the number of times events corresponding to the region type were detected in the j-th time period, and T represents the monitoring activity of a certain region type in the j-th time period. i，j t represents the duration of the event corresponding to the i-th event of the region type detected in the j-th time period. i，j t represents the start time of the detection of the i-th event corresponding to the region type in the j-th time period. i-1，j G represents the start time of the detection of the (i-1)th event corresponding to the region type in the j-th time period. uot This indicates the number of times the system switches to the event monitoring area corresponding to this area type via an externally triggered switching command (user-input switching command) during the j-th time period. It can be seen that monitoring activity increases with the duration of the event, the number of detected events, and the number of externally triggered switching commands, and also increases with the decrease in the interval between two consecutive detected events.

[0155] The higher the monitoring activity, the better the monitoring effect of the event monitoring area corresponding to that area type. Therefore, the event monitoring area corresponding to the maximum monitoring activity in different time periods can be determined as the target event monitoring area for different time periods; when the corresponding time period is reached, the current event monitoring area is switched to the target event monitoring area for the corresponding time period.

[0156] Optionally, the region type includes at least one of the following: shielded region, protected region, alarm region, contour analysis region, area of ​​interest, and AI analysis region. In the step of monitoring events in the event monitoring region based on the point cloud data of the configured event monitoring region and the target region, if the configured event monitoring region is a protected region or alarm region, then based on each two-dimensional projection plane of the protected region or alarm region, it is determined whether an intrusion event exists in the protected region or alarm region. If an intrusion event exists in the protected region or alarm region, then a first security control signal is output. If the configured event monitoring region is an area of ​​interest, then based on the depth dispersion rate of each point cloud within the area of ​​interest, it is determined whether an intrusion event exists within the area of ​​interest. If an intrusion event is detected in the area of ​​interest, a second security control signal is output. If the configured event monitoring area is a contour analysis area, the presence of a contour change event is determined based on the distance changes between the reference frame point cloud data and the current frame point cloud data. If a contour change event is detected, a third security control signal is output. If the configured event monitoring area is an AI analysis area, the presence of a target person is determined based on intensity image data and 3D point cloud data. If a target person is detected, a fourth security control signal is output.

[0157] In this embodiment of the invention, taking a robot or robotic arm as an example, the working area of ​​the robot or robotic arm is a shielded area, and the area outside the shielded area is a protected area or alarm area. When point cloud data appears in the protected area or alarm area, it indicates that an object has invaded the protected area or alarm area from the shielded area during the operation of the robot or robotic arm. A safety control signal can be output to the host computer or calculator to control the corresponding working equipment to stop working.

[0158] The aforementioned two-dimensional projection plane is the position of a three-dimensional point on a two-dimensional plane as seen from the projection viewpoint.

[0159] If the configured event monitoring area is a protected area or an alarm zone, the system determines whether an intrusion event exists within the protected area or alarm zone based on its various two-dimensional projection planes. When an intrusion event is detected within the protected area or alarm zone, a first security control signal is output. This first security control signal can indicate the existence of a security threat and request action.

[0160] The aforementioned areas of interest can be predefined based on the application, or they can be specific structures, etc.

[0161] The aforementioned depth dispersion rate is used to represent the fluctuation range of depth changes in each point cloud within the region of interest. The fluctuation range represents the degree of change in depth values ​​within the region of interest. A larger depth dispersion value indicates more drastic changes in depth values ​​within the region of interest, i.e., larger fluctuations; a smaller depth dispersion value indicates more gradual changes in depth values ​​within the region of interest, i.e., smaller fluctuations.

[0162] If the configured event monitoring area is a region of interest, the system determines whether an intrusion event exists within the region of interest based on the depth dispersion of each point cloud within that region. If an intrusion event is detected, a second security control signal is output. This second security control signal can be used to activate emergency security measures.

[0163] The aforementioned contour analysis area can be understood as the region of contour change between the current frame and the reference frame. The current frame can be understood as a frame in an image sequence, or a frame captured by the sensor in real time. The reference frame is a predefined baseline frame.

[0164] If the configured event monitoring area is the contour analysis area, the system determines whether a contour change event exists within the area based on the distance changes between the reference frame point cloud data and the current frame point cloud data. If a contour change event is detected, a third safety control signal is output. This third safety control signal can be a specific signal issued to ensure operational safety, prevent potential risks, or respond to specific situations. Outputting a third safety control signal indicates that a situation requiring special attention or action has been detected, but it may not yet be severe enough to warrant a complete shutdown or emergency braking.

[0165] The aforementioned AI analysis area can be understood as the region obtained by performing object detection and segmentation on intensity image data using an object detection model. This object detection model can be a deep learning or machine learning-based model that can detect objects in image data and identify the category and location of objects. The aforementioned intensity image data can be understood as the intensity of pixels in a single-channel image; for example, in a grayscale image, it is the image's gray level, and in the RGB color space, it can be understood as the pixel grayscale values ​​of the R, G, and B channels. The aforementioned 3D point cloud data is a collection of points distributed in a large number of three-dimensional spaces, used to represent the surface or shape information of objects.

[0166] If the configured event monitoring area is the AI ​​analysis area, then based on the intensity image data and 3D point cloud data, it is determined whether a target person exists in the AI ​​analysis area. If a target person is found in the AI ​​analysis area, a fourth security control signal is output. The fourth security control signal can be used to provide a security alert for the system or environment.

[0167] As shown in Figure 2, this embodiment of the invention provides a configuration device for a security monitoring system. The security monitoring system includes a host computer and at least one point cloud data acquisition device. The host computer includes an interactive operating interface, and the point cloud data acquisition device is used to collect point cloud data of a target area. The configuration device for the security monitoring system includes: a first configuration module 201, used to receive the point cloud data of the target area from the point cloud data acquisition device and configure the target area as a three-dimensional space displayed in the operating interface; a second configuration module 202, used to receive configuration parameters for events from the user and configure an event monitoring area in the operating interface according to the configuration parameters, the configuration parameters including area type, area bounding box parameters, and event monitoring algorithm; and an event monitoring module 203, used to perform event monitoring on the event monitoring area based on the configured event monitoring area and the point cloud data of the target area.

[0168] Optionally, the second configuration module 202 is further configured to match an event monitoring algorithm corresponding to the region type in the event monitoring algorithm library according to the region type, wherein different region types correspond to different event monitoring algorithms in the event monitoring algorithm library; draw a region box in the three-dimensional space according to the region box parameters; associate the event monitoring algorithm with the region box; and configure the event monitoring region in the operation interface.

[0169] Optionally, the second configuration module 202 is further configured to determine the three-dimensional coordinates of the target's positioning point in the three-dimensional space; and draw a region box in the three-dimensional space according to the frame length, frame width, frame height and the three-dimensional coordinate values ​​of the positioning point, with each target corresponding to one region box.

[0170] Optionally, the event monitoring module 203 is further configured to receive a user's selection instruction, select at least one target event monitoring area from the configured event monitoring areas, and perform event monitoring on the target event monitoring area based on the point cloud data of the target area and in conjunction with the event monitoring algorithm corresponding to the target event monitoring area.

[0171] Optionally, the event monitoring module 203 is further configured to receive a user's switching instruction, determine a target event monitoring area in the event monitoring area; switch the current event monitoring area to the target event monitoring area; and perform event monitoring on the target event monitoring area based on the point cloud data of the target area and in conjunction with the event monitoring algorithm corresponding to the target event monitoring area.

[0172] Optionally, the area type includes at least one of shielded area, protected area, alarm area, contour analysis area, area of ​​interest, and AI analysis area. The event monitoring module 203 is further configured to, if the configured event monitoring area is a protected area or alarm area, determine whether an intrusion event exists in the protected area or alarm area based on each two-dimensional projection plane of the protected area or alarm area; if an intrusion event exists in the protected area or alarm area, output a first security control signal; if the configured event monitoring area is an area of ​​interest, determine whether an intrusion event exists in the area of ​​interest based on the depth dispersion rate of each point cloud within the area of ​​interest; if the area of ​​interest... If an intrusion event is detected in the monitoring area, a second security control signal is output. If the configured event monitoring area is a contour analysis area, the presence of a contour change event in the contour analysis area is determined based on the distance changes between the reference frame point cloud data and the current frame point cloud data. If a contour change event is detected in the contour analysis area, a third security control signal is output. If the configured event monitoring area is an AI analysis area, the presence of a target person in the AI ​​analysis area is determined based on intensity image data and 3D point cloud data. If a target person is detected in the AI ​​analysis area, a fourth security control signal is output.

[0173] This invention also provides a security monitoring system, which includes a host computer and at least one point cloud data acquisition device. The host computer includes an interactive interface, the point cloud data acquisition device is used to collect point cloud data of a target area, and the host computer is used to execute the steps in the configuration method of the security monitoring system provided in this invention.

[0174] As shown in Figure 3, this embodiment of the invention also provides an electronic device, including a processor, which can execute any of the above-mentioned configuration methods for the security monitoring system.

[0175] Specifically, the system includes a processor 301, a memory 302, and a computer program stored in the memory 302 and capable of running on the processor 301, which executes a configuration method for the security monitoring system. The processor 301 runs the computer program storing the configuration method for the security monitoring system in the memory 302, performing the following steps: receiving point cloud data of the target area from the point cloud data acquisition device and configuring the target area as a three-dimensional space displayed in the operation interface; receiving user configuration parameters for events and configuring an event monitoring area in the operation interface according to the configuration parameters, wherein the configuration parameters include area type, area bounding box parameters, and event monitoring algorithm; and performing event monitoring on the event monitoring area based on the configured event monitoring area and the point cloud data of the target area.

[0176] Optionally, the process executed by processor 301 to configure the event monitoring region in the operation interface according to the configuration parameters includes: matching an event monitoring algorithm corresponding to the region type in an event monitoring algorithm library, wherein different region types correspond to different event monitoring algorithms in the event monitoring algorithm library; drawing a region bounding box in the three-dimensional space according to the region bounding box parameters; associating the event monitoring algorithm with the region bounding box, and configuring the event monitoring region in the operation interface.

[0177] Optionally, the region bounding box parameters include bounding box length, bounding box width, and bounding box height. The step of drawing the region bounding box in the three-dimensional space based on the region bounding box parameters executed by the processor 301 includes: determining the three-dimensional coordinates of the target's positioning point in the three-dimensional space; and drawing the region bounding box in the three-dimensional space based on the bounding box length, the bounding box width, the bounding box height, and the three-dimensional coordinates of the positioning point, with each target corresponding to one region bounding box.

[0178] Optionally, the process executed by processor 301 to perform event monitoring on the event monitoring area based on the configured point cloud data of the event monitoring area and the target area includes: receiving a user's selection instruction and selecting at least one target event monitoring area from the configured event monitoring area; and performing event monitoring on the target event monitoring area based on the point cloud data of the target area and in conjunction with the event monitoring algorithm corresponding to the target event monitoring area.

[0179] Optionally, the processor 301 performs event monitoring on the event monitoring area based on the configured point cloud data of the event monitoring area and the target area, including: receiving a user's switching instruction and determining a target event monitoring area in the event monitoring area; switching the current event monitoring area to the target event monitoring area; and performing event monitoring on the target event monitoring area based on the point cloud data of the target area and in conjunction with the event monitoring algorithm corresponding to the target event monitoring area.

[0180] Optionally, the region type includes at least one of shielded region, protected region, alarm region, contour analysis region, attention region, and AI analysis region. The processor 301 executes the event monitoring of the event monitoring region based on the configured event monitoring region and the point cloud data of the target region, including: if the configured event monitoring region is a protected region or alarm region, then based on each two-dimensional projection plane of the protected region or alarm region, determine whether there is an intrusion event in the protected region or alarm region; if there is an intrusion event in the protected region or alarm region, output a first security control signal; if the configured event monitoring region is an attention region, then based on the depth dispersion rate of each point cloud in the attention region, determine whether there is an intrusion event in the attention region; if there is an intrusion event in the attention region, output a second security control signal; if the configured event monitoring region is a contour analysis region, then based on the distance change between the reference frame point cloud data and the current frame point cloud data of the contour analysis region, determine whether there is a contour change event in the contour analysis region; if there is a contour change event in the contour analysis region, output a third security control signal. If the configured event monitoring area is an AI analysis area, then based on intensity image data and 3D point cloud data, it is determined whether there is a target person in the AI ​​analysis area. If there is a target person in the AI ​​analysis area, then a fourth security control signal is output.

[0181] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the configuration method of the security monitoring system provided in this invention and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0182] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0183] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A configuration method for a safety monitoring system, characterized in that, The security monitoring system includes a host computer and at least one point cloud data acquisition device. The host computer includes an interactive interface, and the point cloud data acquisition device is used to acquire point cloud data of a target area. The method includes the following steps: Receive point cloud data of the target area from the point cloud data acquisition device, and configure the target area as a three-dimensional space to be displayed in the operation interface; The system receives configuration parameters for an event from the user and configures an event monitoring area in the operation interface according to the configuration parameters. The configuration parameters include the area type, area bounding box parameters, and event monitoring algorithm. Based on the configured point cloud data of the event monitoring area and the target area, event monitoring is performed on the event monitoring area.

2. The configuration method of the safety monitoring system as described in claim 1, characterized in that, The step of configuring the event monitoring area in the operation interface according to the configuration parameters includes: According to the region type, an event monitoring algorithm corresponding to the region type is matched in the event monitoring algorithm library. In the event monitoring algorithm library, different region types correspond to different event monitoring algorithms. Based on the region bounding parameters, draw the region bounding box in the three-dimensional space; The event monitoring algorithm is associated with the region bounding box, and the event monitoring region is configured in the operation interface.

3. The configuration method of the safety monitoring system as described in claim 2, characterized in that, The region bounding box parameters include bounding box length, bounding box width, and bounding box height. Drawing the region bounding box in the three-dimensional space based on these parameters includes: In the aforementioned three-dimensional space, the three-dimensional coordinates of the target's location point are determined; Based on the frame length, frame width, frame height, and the three-dimensional coordinates of the positioning point, a region box is drawn in the three-dimensional space, with each target corresponding to one region box.

4. The configuration method of the safety monitoring system as described in claim 2, characterized in that, The step of monitoring events in the event monitoring area based on the configured point cloud data of the event monitoring area and the target area includes: Receive the user's selection instruction and select at least one target event monitoring area from the configured event monitoring areas; Based on the point cloud data of the target area, and combined with the event monitoring algorithm corresponding to the target event monitoring area, event monitoring is performed on the target event monitoring area.

5. The configuration method of the safety monitoring system as described in claim 2, characterized in that, The step of monitoring events in the event monitoring area based on the configured point cloud data of the event monitoring area and the target area includes: Receive the user's switching command and determine the target event monitoring area within the event monitoring area; Switch the current event monitoring area to the target event monitoring area; Based on the point cloud data of the target area, and combined with the event monitoring algorithm corresponding to the target event monitoring area, event monitoring is performed on the target event monitoring area.

6. The configuration method of the safety monitoring system as described in claim 1, characterized in that, The region type includes at least one of the following: shielded region, protected region, alarm region, contour analysis region, attention region, and AI analysis region. The event monitoring of the event monitoring region based on the configured event monitoring region and the point cloud data of the target region includes: If the configured event monitoring area is a protected area or an alarm area, then based on each two-dimensional projection plane of the protected area or the alarm area, it is determined whether there is an intrusion event in the protected area or the alarm area. If there is an intrusion event in the protected area or the alarm area, then the first security control signal is output. If the configured event monitoring area is a region of interest, then based on the depth dispersion rate of each point cloud in the region of interest, it is determined whether there is an intrusion event in the region of interest. If there is an intrusion event in the region of interest, then a second security control signal is output. If the configured event monitoring area is a contour analysis area, then based on the distance change between the reference frame point cloud data and the current frame point cloud data in the contour analysis area, it is determined whether there is a contour change event in the contour analysis area. If there is a contour change event in the contour analysis area, then a third safety control signal is output. If the configured event monitoring area is an AI analysis area, then based on intensity image data and 3D point cloud data, it is determined whether there is a target person in the AI ​​analysis area. If there is a target person in the AI ​​analysis area, then a fourth security control signal is output.

7. A configuration device for a safety monitoring system, the safety monitoring system comprising a host computer and at least one point cloud data acquisition device, the host computer including an interactive interface, the point cloud data acquisition device for acquiring point cloud data of a target area, the device comprising: The first configuration module is used to receive point cloud data of the target area from the point cloud data acquisition device, and configure the target area to be displayed in the operation interface as a three-dimensional space. The second configuration module is used to receive configuration parameters for events from the user and configure the event monitoring area in the operation interface according to the configuration parameters. The configuration parameters include area type, area frame parameters and event monitoring algorithm. The event monitoring module is used to monitor events in the event monitoring area based on the configured point cloud data of the event monitoring area and the target area.

8. A safety monitoring system, characterized in that, The security monitoring system includes a host computer and at least one point cloud data acquisition device. The host computer includes an interactive user interface. The point cloud data acquisition device is used to acquire point cloud data of a target area. The host computer is used to execute the steps in the configuration method of the security monitoring system as described in any one of claims 1 to 6.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps in the configuration method of the security monitoring system as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the configuration method for the security monitoring system as described in any one of claims 1 to 6.

Images