CN113947867A - Method, system, electronic device and storage medium for detecting abnormal target behaviors - Google Patents

Method, system, electronic device and storage medium for detecting abnormal target behaviors Download PDF

Info

Publication number
CN113947867A
CN113947867A CN202111115922.3A CN202111115922A CN113947867A CN 113947867 A CN113947867 A CN 113947867A CN 202111115922 A CN202111115922 A CN 202111115922A CN 113947867 A CN113947867 A CN 113947867A
Authority
CN
China
Prior art keywords
target
detected
height value
event
point cloud
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202111115922.3A
Other languages
Chinese (zh)
Other versions
CN113947867B (en
Inventor
陈辰
陈凯
马莉莉
涂钊锋
江一波
姚中扬
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Bestechnic Shanghai Co Ltd
Original Assignee
Ningbo Xitang Information Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Ningbo Xitang Information Technology Co ltd filed Critical Ningbo Xitang Information Technology Co ltd
Priority to CN202111115922.3A priority Critical patent/CN113947867B/en
Publication of CN113947867A publication Critical patent/CN113947867A/en
Application granted granted Critical
Publication of CN113947867B publication Critical patent/CN113947867B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
    • G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/04—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons
    • G08B21/0407—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis
    • G08B21/043—Alarms for ensuring the safety of persons responsive to non-activity, e.g. of elderly persons based on behaviour analysis detecting an emergency event, e.g. a fall
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Health & Medical Sciences (AREA)
  • Psychiatry (AREA)
  • Psychology (AREA)
  • Social Psychology (AREA)
  • General Health & Medical Sciences (AREA)
  • Gerontology & Geriatric Medicine (AREA)
  • Business, Economics & Management (AREA)
  • Emergency Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

The application relates to a method, a device, a system, an electronic device and a storage medium for detecting target abnormal behaviors, wherein the method comprises the following steps: acquiring multi-frame point cloud data acquired by a millimeter wave radar in a monitoring scene, wherein the multi-frame point cloud data comprises multi-frame historical frame point cloud data and current frame point cloud data; extracting a historical height value corresponding to the target to be detected from each frame of historical frame point cloud data, and selecting the highest historical height value from a plurality of historical height values as a reference height value corresponding to the target to be detected; extracting a current height value corresponding to the target to be detected from the current frame point cloud data, comparing the current height value with a reference height value, and determining the change action of the height value of the target to be detected; and determining whether the target to be detected has abnormal behavior according to the current height value and the change action of the height value. By the method and the device, the problem of low accuracy rate of abnormal detection of the indoor target in the related technology is solved, and the technical effect of improving the accuracy rate of the abnormal detection of the indoor target is achieved.

Description

Method, system, electronic device and storage medium for detecting abnormal target behaviors
Technical Field
The present application relates to the field of target detection technologies, and in particular, to a method, an apparatus, a system, an electronic apparatus, and a storage medium for detecting abnormal behavior of a target.
Background
With the continuous development and application of smart cities, smart homes and smart buildings, how to ensure the privacy and safety of users is very important. For example, in the aspect of personnel safety monitoring, the detection of personnel targets is particularly important; in the aspect of human body sign detection, a target to be detected needs to be positioned, and further analysis of signals is performed based on the positioning.
The old people are easy to fall down, have sudden diseases and other accidents due to various reasons, lose the help seeking ability after the accidents happen, and the serious consequences can be caused when the old people do not find the abnormal conditions in time and apply treatment to the abnormal conditions.
The nursing organization (nursing home, hospital, etc.) makes regular rounds through nursing staff, nurses, etc. to investigate abnormal conditions, but nursing staff's quantity and energy are limited after all, how can be under the condition that reduces personnel's input, the very first time discovers abnormal conditions, and accurate efficient provides help and rescues for the old man, is a difficult problem that nursing organization needs to face. On the other hand, the problem of home-based care for the aged is more difficult, the professional staff cannot make a ward visit regularly, and the professional staff can only depend on the care of relatives and friends, in recent years, the solitary old people often fall down or no people can rescue after sudden cardiovascular diseases, and even the solitary old people still cannot find the problem after a long time. Therefore, how to accurately and efficiently realize the anomaly monitoring of the indoor target is a major problem to be solved urgently.
Currently, there are many methods available in the related art for detecting an abnormality of an indoor target, such as: a user can carry out one-button type alarm equipment and actively press a button to call for help in case of accidents; the user can wear the abnormality detection equipment, and the equipment can send alarm information in time when the user falls down or other abnormal conditions occur; and a visible light or infrared camera can be adopted, and the target is subjected to abnormity detection through image processing and the like. However, in such technical solutions, the one-key alarm device requires that the user has a certain behavior ability, and cannot give an alarm to the user who has fallen down or loses the ability to move due to other situations, the wearable abnormality detection device can only detect a heartbeat signal or a falling action, and has the problems of high false alarm rate, high missed detection rate, and the privacy of the user is affected by using a visible light or infrared camera.
At present, no effective solution is provided for the problem of low accuracy of the abnormal detection of the indoor target in the related technology.
Disclosure of Invention
The embodiment of the application provides a method, a device, a system, an electronic device and a storage medium for detecting abnormal behaviors of a target, so as to at least solve the problem of low accuracy rate of detecting the abnormality of an indoor target in the related art.
In a first aspect, an embodiment of the present application provides a method for detecting target abnormal behavior, where the method includes: acquiring multi-frame point cloud data, which are acquired by a millimeter wave radar in a preset monitoring scene and correspond to a target to be detected, wherein the multi-frame point cloud data comprise multi-frame historical frame point cloud data and current frame point cloud data; extracting a historical height value corresponding to the target to be detected from each frame of historical frame point cloud data, and selecting the highest historical height value from a plurality of historical height values as a reference height value corresponding to the target to be detected; extracting a current height value corresponding to the target to be detected from the current frame point cloud data, comparing the current height value with the reference height value, and determining a change action of the height value of the target to be detected; and determining whether the target to be detected has abnormal behaviors or not according to the current height value and the change action of the height value of the target to be detected.
In some embodiments, determining whether the target to be measured has an abnormal behavior according to the current height value and the change action of the height value of the target to be measured includes: determining a target observation event corresponding to the current height value and height value change action according to the current height value and height value change action of the target to be detected; judging whether the risk state corresponding to the target observation event is an abnormal state; and under the condition that the risk state corresponding to the target observation event is determined to be an abnormal state, determining that the target to be detected has abnormal behavior.
In some of these embodiments, the height value change action comprises a height value decrease action and a height value increase action; comparing the current height value with the reference height value, and determining the change action of the height value of the target to be measured comprises the following steps: determining that the height value of the target to be detected is reduced under the condition that the current height value of the target to be detected is smaller than the reference height value and the difference value between the reference height value and the current height value is larger than or equal to a preset first threshold value within a preset first time period; and in the first time period, determining that the height value of the target to be detected rises under the condition that the current height value of the target to be detected is greater than the reference height value and the difference value between the current height value and the reference height value is greater than or equal to a preset second threshold value.
In some of these embodiments, the height value change action comprises a height value decrease action and a height value increase action; the target observation event comprises a target fall event and a target height reduction event; determining a target observation event corresponding to the current height value and height value change action according to the current height value and height value change action of the target to be detected comprises: in a preset second time period, the height value of the target to be detected is reduced, the current height value of the target to be detected is smaller than or equal to a preset third threshold value, and a target falling event of the target to be detected is determined under the condition that the target to be detected with the height value larger than a preset fourth threshold value is not detected in a preset monitoring scene in the second time period; and in the second time period, the target to be detected has a height value reducing action, the current height value of the target to be detected is greater than the third threshold value, and under the condition that no height value increasing action is detected in the second time period, the target to be detected has a target height reducing event.
In some embodiments, determining whether the risk state corresponding to the target observation event is an abnormal state includes: determining that the target to be detected has an abnormal state under the condition that a target falling event occurs to the target to be detected and the duration of the target falling event is greater than a preset fifth threshold; and determining that the target to be detected has an abnormal state under the condition that a target height reduction event occurs to the target to be detected and the duration of the target height reduction event is greater than a preset sixth threshold.
In some of these embodiments, the target observation event comprises a target appearance event and a target disappearance event; determining a target observation event corresponding to the current height value and height value change action further comprises: under the condition that a target to be detected is detected to appear in point cloud data acquired by a millimeter wave radar for the first time, determining that a target occurrence event appears in a monitoring scene corresponding to the millimeter wave radar; and after the target appearance event appears in the monitoring scene is detected, determining that the target disappearance event appears in the monitoring scene under the condition that the point cloud data does not detect the appearance of the target to be detected in a preset third time period.
In some embodiments, millimeter wave radars are arranged in a plurality of monitoring scenes, and the monitoring ranges of the millimeter wave radars are not overlapped; judging whether the risk state corresponding to the target observation event is an abnormal state comprises the following steps: in a preset fourth time period, under the condition that the times of the occurrence of the target events and/or the occurrence of the target disappearance events in a preset first monitoring scene are larger than a preset seventh threshold value, determining that the target to be detected has an abnormal state; in the fourth time period, counting the appearance duration of the target to be detected in the first monitoring scene according to the target appearance event and the target disappearance event, and determining that the target to be detected has an abnormal state under the condition that the appearance duration is greater than a preset eighth threshold; determining that the target to be detected has an abnormal state under the condition that the target occurrence event and/or the target disappearance event do not appear in a preset second monitoring scene within a preset fifth time period; and determining that the target to be detected has an abnormal state under the condition that the target occurrence event occurs in at least two monitoring scenes and the duration of the target occurrence event is greater than a preset ninth threshold.
In some embodiments, the acquiring multi-frame point cloud data of the target to be detected, which is acquired by the millimeter wave radar in a preset monitoring scene, includes: acquiring a point cloud data packet acquired by a millimeter wave radar in a preset monitoring scene, wherein the point cloud data comprises point cloud data of a plurality of detection points acquired by the millimeter wave radar; clustering the point cloud data packets to obtain at least one point cloud cluster; respectively judging whether the number of the detection points in each point cloud cluster is greater than a preset tenth threshold value or not, and respectively judging whether the maximum signal-to-noise ratio of the detection points in each point cloud cluster is greater than a preset eleventh threshold value or not; under the condition that the number of the detection points in the point cloud cluster is larger than the tenth threshold value and the maximum signal-to-noise ratio of the detection points in the point cloud cluster is larger than the eleventh threshold value, taking the point cloud cluster as the point cloud data corresponding to the target to be detected; and taking all point cloud clusters associated with the target to be detected in the point cloud clusters as multi-frame point cloud data corresponding to the target to be detected.
In some of these embodiments, the method further comprises: and in the first time period, determining that the height value of the target to be detected maintains action under the condition that the difference value between the current height value and the reference height value is greater than the negative value of the first threshold and smaller than the second threshold.
In some of these embodiments, the method further comprises: in the second time period, the height value of the target to be detected rises, the current height value of the target to be detected is larger than the fourth threshold value, and under the condition that the height value reduction action is not detected in the second time period, the target to be detected is determined to have a target rising event; in the second time period, the height value of the target to be detected is increased, the current height value of the target to be detected is smaller than or equal to the fourth threshold, and a target height increasing event of the target to be detected is determined under the condition that no height value decreasing action is detected in the second time period; and in the second time period, the target to be detected has any one of three height value change actions, and under the condition that no target falling event, no target height reducing event, no target rising event or no target height increasing event is detected in the second time period, the target to be detected is determined to have a target height maintaining event.
In a second aspect, an embodiment of the present application provides an apparatus for detecting target abnormal behavior, where the apparatus includes: the acquisition module is used for acquiring multi-frame point cloud data, which are acquired by the millimeter wave radar in a preset monitoring scene and correspond to a target to be detected, wherein the multi-frame point cloud data comprise multi-frame historical frame point cloud data and current frame point cloud data; the extraction module is used for extracting a historical height value corresponding to the target to be detected from each frame of historical frame point cloud data, and selecting the highest historical height value from a plurality of historical height values as a reference height value corresponding to the target to be detected; the comparison module is used for extracting a current height value corresponding to the target to be detected from the current frame point cloud data, comparing the current height value with the reference height value and determining the change action of the height value of the target to be detected; and the output module is used for determining whether the target to be detected has abnormal behaviors or not according to the current height value and the change action of the height value of the target to be detected.
In a third aspect, an embodiment of the present application further provides a system for detecting an abnormal behavior of a target, where the system includes: the system comprises a plurality of event observation modules, an event analysis module and an exception handling module; the event observation module comprises millimeter wave radars and a processor, the event observation modules are arranged in different monitoring scenes, the monitoring ranges of the millimeter wave radars are not overlapped, and the event observation module is used for observing abnormal behaviors occurring in the corresponding monitoring scenes; the event analysis module is in communication connection with the event observation module and the exception handling module, and is used for analyzing the exception behavior obtained by the event observation module and sending a handling instruction corresponding to the exception behavior to the exception handling module; the exception handling module is to perform a corresponding handling action according to the handling instruction.
In a fourth aspect, an embodiment of the present application further provides an electronic apparatus, including a memory and a processor, where the memory stores a computer program, and the processor is configured to execute the computer program to perform the method for detecting target abnormal behavior according to the first aspect.
In a fifth aspect, an embodiment of the present application further provides a storage medium, where a computer program is stored in the storage medium, where the computer program, when executed by a processor, implements the method for detecting target abnormal behavior according to the first aspect.
Compared with the related art, the method, the device, the system, the electronic device and the storage medium for detecting the abnormal behavior of the target provided by the embodiment of the application acquire multi-frame point cloud data, which is acquired by a millimeter wave radar in a preset monitoring scene and corresponds to the target to be detected, wherein the multi-frame point cloud data comprises multi-frame historical frame point cloud data and current frame point cloud data; extracting a historical height value corresponding to the target to be detected from each frame of historical frame point cloud data, and selecting the highest historical height value from a plurality of historical height values as a reference height value corresponding to the target to be detected; extracting a current height value corresponding to the target to be detected from the current frame point cloud data, comparing the current height value with a reference height value, and determining a change action of the height value of the target to be detected; and determining whether the target to be detected has abnormal behavior according to the current height value of the target to be detected and the change action of the height value. The problem of low accuracy rate of abnormal detection of the indoor target in the related technology is solved, and the technical effect of improving the accuracy rate of abnormal detection of the indoor target is achieved.
The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below to provide a more thorough understanding of the application.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a flow chart of a method of detection of target anomalous behavior in accordance with an embodiment of the present application;
FIG. 2 is a logic diagram for determining a target abnormal state according to an embodiment of the present application;
fig. 3 is a block diagram of a structure of a target abnormal behavior detection apparatus according to an embodiment of the present application;
FIG. 4 is a block diagram of a system for detecting target anomalous behavior in accordance with an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments provided in the present application without any inventive step are within the scope of protection of the present application. Moreover, it should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another.
Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the specification. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of ordinary skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments without conflict.
Unless defined otherwise, technical or scientific terms referred to herein shall have the ordinary meaning as understood by those of ordinary skill in the art to which this application belongs. Reference to "a," "an," "the," and similar words throughout this application are not to be construed as limiting in number, and may refer to the singular or the plural. The present application is directed to the use of the terms "including," "comprising," "having," and any variations thereof, which are intended to cover non-exclusive inclusions; for example, a process, method, system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to the listed steps or elements, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. Reference to "connected," "coupled," and the like in this application is not intended to be limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. Reference herein to "a plurality" means greater than or equal to two. "and/or" describes an association relationship of associated objects, meaning that three relationships may exist, for example, "A and/or B" may mean: a exists alone, A and B exist simultaneously, and B exists alone. Reference herein to the terms "first," "second," "third," and the like, are merely to distinguish similar objects and do not denote a particular ordering for the objects.
The present embodiment provides a method for detecting a target abnormal behavior, and fig. 1 is a flowchart of a method for detecting a target abnormal behavior according to an embodiment of the present application, and as shown in fig. 1, the method includes:
step S101, acquiring multi-frame point cloud data, which corresponds to a target to be detected and is acquired by the millimeter wave radar in a preset monitoring scene, wherein the multi-frame point cloud data comprises multi-frame historical frame point cloud data and current frame point cloud data.
In this embodiment, the millimeter waves include electromagnetic waves with a millimeter wavelength, the millimeter wave radar may be a radar system that transmits electromagnetic waves with a wavelength greater than 0.1mm and less than 0.2mm and a frequency between 20GHz and 300GHz, and currently, the common millimeter wave radar mainly uses a transmission frequency band such as 24GHz, 60GHz, or 77 GHz.
In this embodiment, the multi-frame historical frame point cloud data may be obtained a period of time before the current frame point cloud data, for example, may be obtained before the first 10s and the first 20s of the current frame point cloud data, where the current frame point cloud data may also be the last frame point cloud data in a set of historical frame point cloud data.
In the above embodiments, the monitoring scenario may be a bathroom, bedroom, or other location of the target residence to be monitored.
And S102, extracting historical height values corresponding to the target to be detected from each frame of historical frame point cloud data, and selecting the highest historical height value from the historical height values as a reference height value corresponding to the target to be detected.
In this embodiment, the reference height value can be obtained by the following formula:
Href＝max(H0,max{H[t]})；
wherein HrefTo reference the height value, H0Is a minimum reference height value, H0Can be preset, e.g. H0Can be set to 1.4m, H [ t ]]The height value corresponding to the target to be measured in the historical frame point cloud data of the t-th frame is obtained. For example, the height values corresponding to a group of related historical frame point cloud data (i.e. the historical frame point cloud data all have the target to be measured) are respectively H [1 ]]＝1.7，H[2]＝1.67，H[3]＝1.2，H[4]＝1.73，H[5]1.8, then according to Href＝max(H0,max{H[t]}) the reference height value of the target to be measured is 1.8 m.
Step S103, extracting a current height value corresponding to the target to be detected from the current frame point cloud data, comparing the current height value with a reference height value, and determining the change action of the height value of the target to be detected.
In the present embodiment, the time T is observed by the maximum action1Inner (T)1Can take 5s or other numerical values) to compare the current height value with the reference height value, and according to the magnitude relation between the current height value and the reference height value, whether the height value of the target to be measured is lowered or raised can be determinedOr a height value maintenance action.
And step S104, determining whether the target to be detected has abnormal behaviors or not according to the current height value and the change action of the height value of the target to be detected.
In this embodiment, whether the target to be detected has an abnormal behavior such as falling or not may be determined according to the height value changing action and the current height value of the target to be detected, for example, when the height value of the target to be detected has a decreasing action, the current height value of the target to be detected is less than or equal to a certain threshold (e.g., 0.4m), and the duration of the height value decreasing action is greater than or equal to a certain threshold (e.g., 20s), it may be determined that the target to be detected has a falling behavior.
In the above embodiment, a millimeter wave radar (T × R is required to be not less than 3) having T transmitting antennas and R receiving antennas may be used to transmit electromagnetic waves, perform signal processing on echo signals, and output radar detection points; wherein, the signal processing to the echo signal includes: performing A/D sampling on echo signals, FFT (Fast Fourier transform, FFT for short) based on a distance dimension, and FFT based on a Doppler dimension; in the multi-antenna radar system, it is also necessary to perform an FFT based on an angle (horizontal direction angle, vertical pitch angle, and the like) dimension or an angle measurement process based on a Capon algorithm or a MUSIC algorithm on the echo signal.
In this embodiment, each radar detection point may be determined by a CFAR (Constant False Alarm Rate, CFAR for short) algorithm, point cloud data is formed, and data information such as a radial distance, a radial velocity, a signal-to-noise ratio, and the like of each detection point is output; in the multi-antenna radar system, data information such as a direction angle and a pitch angle of each detection point is additionally output.
In this embodiment, the radial distance, the direction angle, and the horizontal angle in the radar coordinate system are converted into a six-dimensional vector in a space rectangular coordinate system, and the height of each detection point in the point cloud data from the ground and the two-dimensional projection coordinate on the ground can be calculated by combining the installation position and the angle measured when the millimeter wave radar is installed.
In this embodiment, Based on information such as the Spatial distance and the velocity of each detection point in the point cloud data, an algorithm such as a DBSCAN (Density-Based Clustering method with Noise, referred to as DBSCAN) may be used to perform cluster analysis on the detection points in the point cloud data, and obtain a point cloud cluster in which a target to be detected exists.
In this embodiment, after obtaining the point cloud cluster having the target to be measured (i.e. the point cloud data corresponding to the target to be measured), the height value of the target to be measured can be obtained by the following formula:
Figure BDA0003275278010000081
wherein H is the height value of the target to be measured, HiHeight of the ith detection point from the ground, SNRiFor the signal-to-noise ratio of the ith detection point, the signal-to-noise ratio needs to be greater than eta3This preset threshold.
The horizontal coordinate of the target to be measured can also be obtained by the following formula:
Figure BDA0003275278010000082
Figure BDA0003275278010000083
wherein X and Y are horizontal coordinates of the target to be measured, and XiAnd yiNamely the two-dimensional projection coordinate of the ith detection point on the ground.
In this embodiment, after obtaining the horizontal coordinates of the target to be detected in the t-1 th frame and the t-1 th frame point cloud data, it can be determined whether the distance between the target to be detected in the t-1 th frame point cloud data and the target to be detected in the t-1 th frame point cloud data is smaller than a preset threshold ηD：
(X[t]-X[t-1])2+(Y[t]-Y[t-1])2≤ηD；
If less than or equal to the threshold etaDThen the targets to be measured of the two frames can be associated togetherUntil a group of associated point cloud data, namely the point cloud data corresponding to the multi-frame target to be measured, eta is obtainedDCan be set according to actual needs, and the application is not limited herein.
In the above embodiments, the radar has advantages over other optical sensors in that it is not affected by ambient light and has a certain barrier penetration capability; compare in the bigger radar system of wavelength (frequency is less than 20GHz), the millimeter wave radar has the characteristics that the wavelength is short, the bandwidth is big, thereby possess higher resolution ratio, the precision is higher when using the unusual action to indoor target to detect, sensitivity is higher simultaneously, utilize the millimeter wave radar to use the wavelength signal of millimeter level to detect, there is good detection performance to the micro-action of target more than the centimeter level, reduce false positive rate and the rate of missing detection to the unusual detection of indoor target, and can not infringe the privacy of target in the monitoring range.
Through the steps from S101 to S104, multi-frame point cloud data corresponding to the target to be detected and collected by the millimeter wave radar in a preset monitoring scene is obtained, wherein the multi-frame point cloud data comprises multi-frame historical frame point cloud data and current frame point cloud data; extracting a historical height value corresponding to the target to be detected from each frame of historical frame point cloud data, and selecting the highest historical height value from a plurality of historical height values as a reference height value corresponding to the target to be detected; extracting a current height value corresponding to the target to be detected from the current frame point cloud data, comparing the current height value with a reference height value, and determining a change action of the height value of the target to be detected; whether abnormal behaviors exist in the target to be detected is determined according to the current height value and the change action of the height value of the target to be detected, millimeter-wave radar is used for detecting by using millimeter-scale wavelength signals, the detection performance is good for micro actions of the target above the centimeter scale, the false alarm rate and the omission factor of abnormal detection of the indoor target are reduced, and the privacy of the target in the monitoring range cannot be violated. By the method and the device, the problem of low accuracy rate of abnormal detection of the indoor target in the related technology is solved, and the technical effect of improving the accuracy rate of the abnormal detection of the indoor target is achieved.
In some embodiments, determining whether the target to be measured has abnormal behavior according to the current height value and the change action of the height value of the target to be measured is implemented by the following steps:
step 1, determining a target observation event corresponding to the current height value and height value change action according to the current height value and height value change action of the target to be detected.
And 2, judging whether the risk state corresponding to the target observation event is an abnormal state.
And 3, determining that the target to be detected has abnormal behavior under the condition that the risk state corresponding to the target observation event is determined to be an abnormal state.
TABLE 1
Figure BDA0003275278010000091
Figure BDA0003275278010000101
As shown in table 1, in this embodiment, the height value change action of the target to be measured may include a height value lowering action, a height value raising action, and a height value maintaining action, and in addition to the height value change action, the target to be measured also includes a target appearing action and a target disappearing action, where the current height value is compared with the reference height value, and the height value change action of the target to be measured is determined by the following steps:
step 1, determining that the height value of the target to be detected is reduced under the condition that the current height value of the target to be detected is smaller than a reference height value and the difference value between the reference height value and the current height value is larger than or equal to a preset first threshold value within a preset first time period.
And 2, determining that the height value of the target to be detected rises under the condition that the current height value of the target to be detected is greater than the reference height value and the difference value between the current height value and the reference height value is greater than or equal to a preset second threshold value in the first time period.
And 3, determining that the height value of the target to be detected maintains the action under the condition that the difference value between the current height value and the reference height value is larger than the negative value of the first threshold value and smaller than the second threshold value in the first time period.
As shown in Table 1, TAI.e. a first time period, T, preset in the present application1For a predetermined maximum motion observation time, H0]A reference height value corresponding to the target to be measured or a height value, eta of the target to be measured at the moment when t is 01Is a preset first threshold value, eta2Is a preset second threshold value, T1The value can be set to 5S, and other values can be set. When the current height value H [ T ] of the target to be measuredA]Less than a reference height value H [0 ]]And H [0 ]]-H[TA]≥η1And then, the height value of the target to be measured can be determined to be reduced.
When the current height value H [ T ] of the target to be measuredA]Greater than a reference height value H [0 ]]And H [ T ]A]-H[0]≥η2In time, the height value of the target to be detected can be determined to be increased; when the current height value H [ T ] of the target to be measuredA]When the value is basically unchanged, i.e. for any value of T is more than 0 and less than or equal to T1Satisfy- η1＜H[t]-H[0]＜η2And determining that the height value maintaining action of the target to be measured occurs.
In the above embodiment, the first threshold η1Can be set to 0.1m, the second threshold η2May be set to 0.2m, for example, when referring to the height value H [0 ]]Is 1.4m, T is more than 0 and less than or equal to T1Satisfy-0.1 < H [ t ]]And (4) determining that the height value of the target to be detected is maintained when the value is-1.4 is less than 0.2.
In the above embodiment, the height value lowering action a1, the height value raising action a2, and the height value maintaining action A3 may be associated together for subsequent target observed event determination.
TABLE 2
Figure BDA0003275278010000111
As shown in table 2, in this embodiment, the target observation events may include a target fall event E1, a target height reduction event E2, a target uprighting event E3, a target height increase event E4, a target height maintaining event E5, a target occurrence event E6, and a target disappearance event E7, and the determination of the target observation event corresponding to the current height value and the change action of the height value is implemented by the following steps according to the current height value and the change action of the height value of the target to be detected:
step 1, in a preset second time period, the height value of the target to be detected is reduced, the current height value of the target to be detected is smaller than or equal to a preset third threshold, and in the second time period, under the condition that the target to be detected with the height value larger than a preset fourth threshold is not detected in a preset monitoring scene, the target to be detected is determined to fall down.
And 2, determining that the target height of the target to be detected is reduced under the condition that the height value of the target to be detected is reduced within the second time period, the current height value of the target to be detected is greater than the third threshold value, and the height value of the target to be detected is not increased within the second time period.
And 3, determining that the target to be detected has a target getting-up event under the conditions that the height value of the target to be detected rises within the second time period, the current height value of the target to be detected is greater than the fourth threshold value, and the height value reduction action is not detected within the second time period.
And 4, in a second time period, the height value of the target to be detected rises, the current height value of the target to be detected is smaller than or equal to a fourth threshold, and under the condition that the height value reduction action is not detected in the second time period, the target height rising event of the target to be detected is determined.
And 5, in a second time period, determining that a target height maintaining event occurs on the target to be detected under the condition that any one of three height value change actions occurs on the target to be detected and no target falling event, target height reducing event, target rising event and target height increasing event are detected in the second time period.
And 6, under the condition that the target to be detected is detected to appear in the point cloud data acquired by the millimeter wave radar for the first time, determining that the target appearing event appears in the monitoring scene corresponding to the millimeter wave radar.
And 7, after the target occurrence event is detected in the monitoring scene, and in a preset third time period, under the condition that the target to be detected does not appear in the point cloud data, determining that the target disappearance event appears in the monitoring scene.
In this example, T is shown in Table 22For a preset second time period, i.e. the maximum event observation time (T)2May be set to 10s, or may be set to other values), H [ T ]A]I.e. the current height value, delta, of the object to be measured1×HrefIs a preset third threshold value, HrefIs a reference height value, delta, of the object to be measured2×HrefIs a preset fourth threshold value, wherein1Can be set to 0.2, delta2And may be set to 0.4 or other values, and the application is not limited herein.
In the above embodiment, the third period of time may be NDA period of time T1，T1For a preset maximum motion observation time (which can be set to 5s), NDThe setting can be made according to the actual requirement, for example, 10, 20, etc.
In this embodiment, the corresponding target observation event can be extracted according to the judgment logic shown in table 2 through a plurality of actions of each group of associated targets, and for different radar installation positions and application scene requirements, the method can be applied only by adjusting and programming the relevant parameters and the judgment logic in table 1 and table 2, and has better applicability and higher flexibility for the anomaly detection of the indoor targets in different scenes.
Fig. 2 is a logic diagram for determining a target abnormal state according to an embodiment of the present application, and as shown in fig. 2, in some embodiments, determining whether a risk state corresponding to a target observation event is an abnormal state is implemented by:
step 1, determining that the target to be detected has an abnormal state under the condition that a target falling event occurs to the target to be detected and the duration time of the target falling event is greater than a preset fifth threshold value;
and 2, determining that the target to be detected has an abnormal state under the condition that a target height reduction event occurs to the target to be detected and the duration of the target height reduction event is greater than a preset sixth threshold.
In this embodiment, target observation events reported by one or more event observation modules (including millimeter wave radars and processors) in different installation locations may be collected, that is, the event observation modules are disposed in different monitoring scenarios, and the first handling instruction may be generated according to a target fall event E1, a target height reduction event E2, a target rise event E3, a target height increase event E4, a target height maintenance event E5, a target occurrence event E6, and a target disappearance event E7, which are reported by any one of the event observation modules and are associated together.
In this embodiment, as shown in fig. 2, a state machine may be established for a group of associated events, and when the height value of the target to be measured is normal, it is determined that the target to be measured is in a normal state; when the height value of the target to be detected is lower, determining that the target to be detected is in a low risk state; when the height value of the target to be detected is very low, determining that the target to be detected is in a high risk state; if the duration of the high-risk state of the target to be detected is greater than a preset fifth threshold value Tw1And/or the duration of the low-risk state of the target to be detected is greater than a preset sixth threshold value Tw2And if so, determining that the target to be detected has an abnormal state, and generating a first handling instruction at the moment.
Wherein, Tw1Less than or equal to Tw2Their values may be configured according to a particular monitoring scenario, e.g. when the event observation module is installed in a toilet, Tw1Can be set to 30s, Tw2Can be set to 30 min; when the event observation module is installed in the bedroom, Tw1Can be set to 5min, Tw2May be set to 12 h.
In this embodiment, as shown in table 2, when the target fall event E1 occurs in the target to be detected, and the duration time of the target fall event E1 is greater than the preset fifth threshold Tw1Under the condition of (1), the abnormal state of the target to be detected can be determined;when the target height reduction event E2 occurs in the target to be detected, and the duration time of the target height reduction event E2 is greater than a preset sixth threshold value Tw2For example, when the target to be detected is in a high risk state, the target to be detected may be re-entered into a normal state through the target rising event E3, or the target to be detected may be entered into a low risk state through the target height increasing event E4, and then the target to be detected may be entered into a normal state through the target rising event E3.
In some embodiments, millimeter wave radars may be set in a plurality of monitoring scenes, and the monitoring ranges of each millimeter wave radar are not overlapped; and judging whether the risk state corresponding to the target observation event is an abnormal state or not, and further implementing the following steps:
step 1, in a preset fourth time period, under the condition that the frequency of occurrence of target events and/or target disappearance events in a preset first monitoring scene is greater than a preset seventh threshold, determining that the target to be detected has an abnormal state.
And 2, in a fourth time period, counting the appearance time of the target to be detected in the first monitoring scene according to the target appearance event and the target disappearance event, and determining that the target to be detected has an abnormal state under the condition that the appearance time is greater than a preset eighth threshold.
And 3, determining that the target to be detected has an abnormal state under the condition that the target appearance event and/or the target disappearance event does not appear in the preset second monitoring scene within a preset fifth time period.
And 4, determining that the target to be detected has an abnormal state under the condition that the target occurrence events occur in at least two monitoring scenes and the duration of the target occurrence events is greater than a preset ninth threshold value.
In this embodiment, the second handling instruction may be generated according to the target occurrence event E6 and the target disappearance event E7 reported by a specific group of event observation modules, for example, the event observation modules in all washrooms (i.e., the first monitoring scenario) in a certain residential building are grouped, the times of reporting the target occurrence event E6 and/or the target disappearance event E7 by the group of event observation modules within a certain time (i.e., a preset fourth time period, which may be set to 1 day) are counted, the time length for using the washrooms in an accumulated manner is calculated, and if the occupied times are greater than a preset seventh threshold value or the time length of occurrence exceeds a preset eighth threshold value, it is determined that the target to be measured has an abnormal state, and the second handling instruction is generated.
In this embodiment, a third handling instruction may be generated according to the target occurrence event E6 and the target disappearance event E7 reported by a specific group of event observation modules, for example, event observation modules in all toilets (i.e., in a second monitoring scenario) in a certain residential building are grouped, and in a fifth time period (which may be set to 12h), when none of the group of observation modules reports the target occurrence event E6 and/or the target disappearance event E7, it is determined that the target to be measured has an abnormal state, and the third handling instruction is generated.
In this embodiment, the fourth processing instruction may also be generated according to the target occurrence event E6 and the target disappearance event E7 reported by all the event observation modules, for example, millimeter wave radars are installed in multiple rooms/areas of a residence of a certain individual person, and during installation, it is ensured that the monitoring ranges of each millimeter wave radar do not overlap with each other, the target occurrence event E6 and the target disappearance event E7 reported by each event observation module in sequence are recorded, and if it is found that a target exists in more than one area (i.e., the target exists in the reported E6 and does not report E7) and the duration of more than one target occurrence event exceeds a preset ninth threshold (the ninth threshold needs to be more than N)D*T1，T1For a preset maximum motion observation time (which can be set to 5s), NDMay be set according to actual needs, for example, set to 10, 20, etc.), it is determined that the target to be measured is in an abnormal state, and a fourth processing instruction is generated.
In the above embodiment, when it is determined that the target to be measured has the abnormal behavior, the indication information may be sent to the preset communication address.
In this embodiment, corresponding handling actions may be executed according to the first handling instruction, the second handling instruction, the third handling instruction, and the fourth handling instruction, and common handling actions include ringing alarm, lighting alarm, calling a preset emergency contact to alarm, and the like.
In the above embodiment, the first treatment instruction indicates that the target to be measured has abnormal behaviors such as falling, failing to fall, or failing to trip, and the corresponding treatment action applied in the endowment institution is usually to notify the nursing staff to rescue by ringing or lighting; in a home scenario, the corresponding handling action may be sending an instruction message or dialing a phone call to a preset address.
The second treatment instruction indicates that the target to be measured uses an abnormal behavior such as a toilet too frequently or too long, so that a health hazard exists, and the corresponding treatment action may be to send the record to a health management APP of the monitored person or a database of a medical institution.
The third handling instruction indicates that the object to be detected does not use the toilet for too long time, and if the condition that the object to be detected goes out is eliminated, the object to be detected has abnormal behavior of losing mobility outside the monitoring range coverage area; the corresponding handling action can be automatically sending an indication message or dialing a phone to a preset communication address in a home scene.
The fourth processing instruction indicates that the house of the target to be detected has abnormal behaviors such as outsider intrusion, and the corresponding processing action can be automatically sending an indication message or dialing a telephone to a preset communication address (such as an emergency contact or an alarm telephone address).
In some embodiments, acquiring multi-frame point cloud data of a target to be detected, which is acquired by a millimeter wave radar in a preset monitoring scene, is realized by the following steps:
step 1, a point cloud data packet acquired by a millimeter wave radar in a preset monitoring scene is acquired, wherein the point cloud data comprises point cloud data of a plurality of detection points acquired by the millimeter wave radar.
And 2, clustering the point cloud data packets to obtain at least one point cloud cluster.
And 3, respectively judging whether the number of the detection points in each point cloud cluster is greater than a preset tenth threshold value or not, and respectively judging whether the maximum signal-to-noise ratio of the detection points in each point cloud cluster is greater than a preset eleventh threshold value or not.
And 4, under the condition that the number of the detection points in the point cloud cluster is greater than a tenth threshold value and the maximum signal-to-noise ratio of the detection points in the point cloud cluster is greater than an eleventh threshold value, taking the point cloud cluster as point cloud data corresponding to the target to be detected.
And 5, taking all point cloud clusters associated with the target to be detected in the point cloud clusters as multi-frame point cloud data corresponding to the target to be detected.
In this embodiment, the point cloud data includes point cloud data of a plurality of detection points acquired by the millimeter wave radar, and the point cloud data includes a radial distance between the detection point and the millimeter wave radar, a radial speed of the detection point, and a signal-to-noise ratio of the detection point.
In the above embodiment, when the millimeter wave radar is a multi-antenna radar system, the point cloud data further includes information such as a direction angle and a pitch angle of the detection point; in addition, the four-dimensional vector (radial distance, radial speed, direction angle and horizontal angle) output by the millimeter wave radar in the spherical coordinate system is converted into a six-dimensional vector (namely X, Y, Z position coordinates with the radar as the origin and corresponding speed components of three coordinate axes) in the spatial rectangular coordinate system, so that the data processing accuracy of the point cloud data is improved.
In the above embodiment, the detection points may be sorted from high to low according to the signal-to-noise ratio to obtain sorted detection points, the spatial rectangular coordinate position and the radial velocity of each detection point are determined according to the radial distance and the relative angle (pitch angle, direction angle, etc.) between each detection point and the millimeter wave radar, and the sorted detection points are radially clustered based on the spatial rectangular coordinate position and the radial velocity of each detection point to obtain a plurality of point cloud clusters.
After a plurality of point cloud clusters are obtained, the number N of detection points contained in each point cloud cluster can be judgedpcWhether it is greater than a preset tenth threshold ηpcAnd judging the maximum signal-to-noise ratio max { SNR } of the detection points in each point cloud clusteriWhether it is greater than a preset eleventh threshold ηsnrWhen the number of the detection points contained in the point cloud cluster is less than a tenth threshold value, marking the detection points contained in the point cloud cluster as noise; and setting a signal-to-noise ratio detection threshold value, namely an eleventh threshold value etasnrDiscarding the maximum signal-to-noise ratio max { SNR } for the detection points in the point cloud clusteriIs lower than the signal-to-noise ratio detection threshold ηsnrThe effective point cloud cluster is obtained, namely the point cloud cluster corresponding to a target to be detected can be used as point cloud data corresponding to the target to be detected, invalid interference data is reduced, only the effective point cloud cluster is left, partial noise is removed, the calculation amount is reduced, the calculation speed is improved, and the determination accuracy of the target point cloud cluster is improved.
The present embodiment provides a target abnormal behavior detection apparatus, and fig. 3 is a block diagram of a structure of the target abnormal behavior detection apparatus according to an embodiment of the present application, and as shown in fig. 3, the apparatus includes: the acquisition module 30 is configured to acquire multi-frame point cloud data, which is acquired by the millimeter wave radar in a preset monitoring scene and corresponds to a target to be detected, where the multi-frame point cloud data includes multi-frame historical frame point cloud data and current frame point cloud data; the extracting module 31 is configured to extract a historical height value corresponding to the target to be detected from each frame of historical frame point cloud data, and select a highest historical height value from the multiple historical height values as a reference height value corresponding to the target to be detected; the comparison module 32 is used for extracting a current height value corresponding to the target to be detected from the current frame point cloud data, comparing the current height value with a reference height value and determining a height value change action of the target to be detected; and the output module 33 is configured to determine whether the target to be detected has an abnormal behavior according to the current height value of the target to be detected and the change action of the height value.
In some embodiments, the output module 33 is further configured to determine, according to the current height value and the height value change action of the target to be measured, a target observation event corresponding to the current height value and the height value change action; judging whether the risk state corresponding to the target observation event is an abnormal state; and under the condition that the risk state corresponding to the target observation event is determined to be an abnormal state, determining that the target to be detected has abnormal behavior.
In some of these embodiments, the height value change action comprises a height value decrease action and a height value increase action; the comparison module 32 is further configured to determine that a height value reduction action occurs in the target to be measured when the current height value of the target to be measured is smaller than the reference height value and the difference between the reference height value and the current height value is greater than or equal to a preset first threshold value within a preset first time period; and determining that the height value of the target to be detected rises under the condition that the current height value of the target to be detected is greater than the reference height value and the difference value between the current height value and the reference height value is greater than or equal to a preset second threshold value in the first time period.
In some of these embodiments, the height value change action comprises a height value decrease action and a height value increase action; the target observation event comprises a target falling event and a target height reduction event; the output module 33 is further configured to determine that a target falling event occurs in the target to be detected when the height value of the target to be detected decreases within a preset second time period, the current height value of the target to be detected is less than or equal to a preset third threshold, and the target to be detected with the height value greater than a preset fourth threshold is not detected within a preset monitoring scene within the second time period; and in the second time period, the target to be detected has a height value reducing action, the current height value of the target to be detected is greater than a third threshold value, and under the condition that no height value increasing action is detected in the second time period, the target to be detected has a target height reducing event.
In some embodiments, the output module 33 is further configured to determine that an abnormal state exists in the target to be detected when the target to be detected has a target falling event, and the duration of the target falling event is greater than a preset fifth threshold; and determining that the target to be detected has an abnormal state under the condition that the target height reduction event occurs to the target to be detected and the duration of the target height reduction event is greater than a preset sixth threshold.
In some of these embodiments, the target observation events include target appearance events and target disappearance events; the output module 33 is further configured to determine that a target occurrence event occurs in a monitoring scene corresponding to the millimeter wave radar when the occurrence of a target to be detected is detected for the first time in the point cloud data acquired by the millimeter wave radar; and after the target occurrence event in the monitoring scene is detected, determining that the target disappearance event occurs in the monitoring scene under the condition that the target to be detected does not appear in the point cloud data in a preset third time period.
In some embodiments, millimeter wave radars are arranged in a plurality of monitoring scenes, and the monitoring ranges of the millimeter wave radars are not overlapped; the output module 33 is further configured to determine that the target to be detected has an abnormal state in a case that the number of times of occurrence of the target occurrence event and/or the target disappearance event in the preset first monitoring scene is greater than a preset seventh threshold in a preset fourth time period; in a fourth time period, counting the occurrence time of the target to be detected in the first monitoring scene according to the target occurrence event and the target disappearance event, and determining that the target to be detected has an abnormal state under the condition that the occurrence time is greater than a preset eighth threshold; determining that the target to be detected has an abnormal state under the condition that the target occurrence event and/or the target disappearance event do not appear in the preset second monitoring scene within a preset fifth time period; and determining that the target to be detected has an abnormal state under the condition that the target occurrence events occur in at least two monitoring scenes and the duration of the target occurrence events is greater than a preset ninth threshold.
In some embodiments, the obtaining module 30 is further configured to obtain a point cloud data packet acquired by the millimeter wave radar in a preset monitoring scene, where the point cloud data includes point cloud data of a plurality of detection points acquired by the millimeter wave radar; clustering the point cloud data packets to obtain at least one point cloud cluster; respectively judging whether the number of the detection points in each point cloud cluster is greater than a preset tenth threshold value or not, and respectively judging whether the maximum signal-to-noise ratio of the detection points in each point cloud cluster is greater than a preset eleventh threshold value or not; under the condition that the number of the detection points in the point cloud cluster is larger than a tenth threshold value and the maximum signal-to-noise ratio of the detection points in the point cloud cluster is larger than an eleventh threshold value, taking the point cloud cluster as point cloud data corresponding to a target to be detected; and taking all point cloud clusters associated with the target to be detected in the point cloud clusters as multi-frame point cloud data corresponding to the target to be detected.
In some embodiments, the apparatus further includes an indication module, configured to send indication information to a preset communication address when it is determined that the target to be detected has an abnormal behavior.
In some embodiments, the comparing module 32 is further configured to determine that the height value maintaining action occurs in the target to be measured if the difference between the current height value and the reference height value is greater than a negative value of the first threshold and less than the second threshold within the first time period.
In some embodiments, the output module 33 is further configured to determine that a target getting-up event occurs in the target to be detected in the case that the height value of the target to be detected is increased within the second time period, the current height value of the target to be detected is greater than the fourth threshold, and the height value is not decreased within the second time period; in a second time period, the height value of the target to be detected rises, the current height value of the target to be detected is smaller than or equal to a fourth threshold value, and under the condition that the height value reduction action is not detected in the second time period, the target to be detected is determined to have a target height rise event; and in the second time period, the target to be detected has any one of three height value change actions, and under the condition that no target falling event, no target height reducing event, no target rising event or no target height increasing event is detected in the second time period, the target to be detected is determined to have a target height maintaining event.
It should be noted that, for specific examples in this embodiment, reference may be made to examples described in the foregoing embodiments and optional implementations, and details of this embodiment are not described herein again.
This embodiment provides a system for detecting a target abnormal behavior, fig. 4 is a block diagram of a structure of a system for detecting a target abnormal behavior according to an embodiment of the present application, and as shown in fig. 4, the system includes: a plurality of event observation modules 40, event analysis modules 41, and exception handling modules 42; the event observation module 40 comprises a millimeter wave radar 400 and a processor 401, the event observation modules 40 are arranged in different monitoring scenes, the monitoring ranges of the millimeter wave radars 400 are not overlapped, and the event observation module 40 is used for observing abnormal behaviors occurring in the corresponding monitoring scenes; the event analysis module 41 is in communication connection with both the event observation module 40 and the exception handling module 42, and is configured to analyze the exception behavior obtained by the event observation module 40 and send a handling instruction corresponding to the exception behavior to the exception handling module 42; exception handling module 42 is to perform corresponding handling actions according to the handling instructions.
It should be noted that, for specific examples in this embodiment, reference may be made to examples described in the foregoing embodiments and optional implementations, and details of this embodiment are not described herein again.
The present embodiment further provides an electronic device, fig. 5 is a schematic diagram of a hardware structure of the electronic device according to an embodiment of the present application, and as shown in fig. 5, the electronic device includes a memory 504 and a processor 502, a computer program is stored in the memory 504, and the processor 502 is configured to execute the computer program to perform the steps in any of the method embodiments.
Specifically, the processor 502 may include a Central Processing Unit (CPU), or A Specific Integrated Circuit (ASIC), or may be configured to implement one or more Integrated circuits of the embodiments of the present Application.
Memory 504 may include, among other things, mass storage for data or instructions. By way of example, and not limitation, memory 504 may include a Hard Disk Drive (Hard Disk Drive, abbreviated to HDD), a floppy Disk Drive, a Solid State Drive (SSD), flash memory, an optical Disk, a magneto-optical Disk, tape, or a Universal Serial Bus (USB) Drive or a combination of two or more of these. Memory 504 may include removable or non-removable (or fixed) media, where appropriate. The memory 504 may be internal or external to the detection device of the target abnormal behavior, where appropriate. In a particular embodiment, the memory 504 is a Non-Volatile (Non-Volatile) memory. In particular embodiments, Memory 504 includes Read-Only Memory (ROM) and Random Access Memory (RAM). The ROM may be mask-programmed ROM, Programmable ROM (PROM), Erasable PROM (EPROM), Electrically Erasable PROM (EEPROM), Electrically rewritable ROM (EAROM), or FLASH Memory (FLASH), or a combination of two or more of these, where appropriate. The RAM may be a Static Random-Access Memory (SRAM) or a Dynamic Random-Access Memory (DRAM), where the DRAM may be a Fast Page Mode Dynamic Random-Access Memory (FPMDRAM), an Extended data output Dynamic Random-Access Memory (EDODRAM), a Synchronous Dynamic Random-Access Memory (SDRAM), and the like.
Memory 504 may be used to store or cache various data files for processing and/or communication purposes, as well as possibly computer program instructions for execution by processor 502.
The processor 502 may implement any one of the above-described embodiments of the method for detecting target abnormal behavior by reading and executing computer program instructions stored in the memory 504.
Optionally, the electronic apparatus may further include a transmission device 506 and an input/output device 508, wherein the transmission device 506 is connected to the processor 502, and the input/output device 508 is connected to the processor 502.
Optionally, in this embodiment, the processor 502 may be configured to execute the following steps by a computer program:
and S1, acquiring multi-frame point cloud data, which are acquired by the millimeter wave radar in a preset monitoring scene and correspond to the target to be detected, wherein the multi-frame point cloud data comprise multi-frame historical frame point cloud data and current frame point cloud data.
And S2, extracting historical height values corresponding to the target to be detected from each frame of historical frame point cloud data, and selecting the highest historical height value from the historical height values as a reference height value corresponding to the target to be detected.
And S3, extracting the current height value corresponding to the target to be detected from the current frame point cloud data, comparing the current height value with the reference height value, and determining the change action of the height value of the target to be detected.
And S4, determining whether the target to be detected has abnormal behavior according to the current height value and the change action of the height value of the target to be detected.
It should be noted that, for specific examples in this embodiment, reference may be made to examples described in the foregoing embodiments and optional implementations, and details of this embodiment are not described herein again.
In addition, in combination with the method for detecting the target abnormal behavior in the foregoing embodiments, the embodiments of the present application may provide a storage medium to implement. The storage medium having stored thereon a computer program; the computer program, when executed by a processor, implements a method of detecting target abnormal behavior as in any of the above embodiments.
It should be understood by those skilled in the art that various features of the above embodiments can be combined arbitrarily, and for the sake of brevity, all possible combinations of the features in the above embodiments are not described, but should be considered as within the scope of the present disclosure as long as there is no contradiction between the combinations of the features.
The above examples are merely illustrative of several embodiments of the present application, and the description is more specific and detailed, but not to be construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims (14)

1. A method for detecting abnormal behavior of a target, the method comprising:
acquiring multi-frame point cloud data, which are acquired by a millimeter wave radar in a preset monitoring scene and correspond to a target to be detected, wherein the multi-frame point cloud data comprise multi-frame historical frame point cloud data and current frame point cloud data;
extracting a historical height value corresponding to the target to be detected from each frame of historical frame point cloud data, and selecting the highest historical height value from a plurality of historical height values as a reference height value corresponding to the target to be detected;
extracting a current height value corresponding to the target to be detected from the current frame point cloud data, comparing the current height value with the reference height value, and determining a change action of the height value of the target to be detected;
and determining whether the target to be detected has abnormal behaviors or not according to the current height value and the change action of the height value of the target to be detected.
2. The method for detecting the abnormal behavior of the target according to claim 1, wherein determining whether the abnormal behavior of the target to be detected exists according to the current height value and the change action of the height value of the target to be detected comprises:
determining a target observation event corresponding to the current height value and height value change action according to the current height value and height value change action of the target to be detected;
judging whether the risk state corresponding to the target observation event is an abnormal state;
and under the condition that the risk state corresponding to the target observation event is determined to be an abnormal state, determining that the target to be detected has abnormal behavior.
3. The method for detecting the target abnormal behavior according to claim 2, wherein the height value change action includes a height value lowering action and a height value raising action; comparing the current height value with the reference height value, and determining the change action of the height value of the target to be measured comprises the following steps:
determining that the height value of the target to be detected is reduced under the condition that the current height value of the target to be detected is smaller than the reference height value and the difference value between the reference height value and the current height value is larger than or equal to a preset first threshold value within a preset first time period;
and in the first time period, determining that the height value of the target to be detected rises under the condition that the current height value of the target to be detected is greater than the reference height value and the difference value between the current height value and the reference height value is greater than or equal to a preset second threshold value.
4. The method for detecting the target abnormal behavior according to claim 2, wherein the height value change action includes a height value lowering action and a height value raising action; the target observation event comprises a target fall event and a target height reduction event; determining a target observation event corresponding to the current height value and height value change action according to the current height value and height value change action of the target to be detected comprises:
in a preset second time period, the height value of the target to be detected is reduced, the current height value of the target to be detected is smaller than or equal to a preset third threshold value, and a target falling event of the target to be detected is determined under the condition that the target to be detected with the height value larger than a preset fourth threshold value is not detected in a preset monitoring scene in the second time period;
and in the second time period, the target to be detected has a height value reducing action, the current height value of the target to be detected is greater than the third threshold value, and under the condition that no height value increasing action is detected in the second time period, the target to be detected has a target height reducing event.
5. The method for detecting the abnormal behavior of the target according to claim 4, wherein determining whether the risk status corresponding to the target observation event is an abnormal status comprises:
determining that the target to be detected has an abnormal state under the condition that a target falling event occurs to the target to be detected and the duration of the target falling event is greater than a preset fifth threshold;
and determining that the target to be detected has an abnormal state under the condition that a target height reduction event occurs to the target to be detected and the duration of the target height reduction event is greater than a preset sixth threshold.
6. The method for detecting the abnormal behavior of the target according to claim 2, wherein the target observation event comprises a target appearance event and a target disappearance event; determining a target observation event corresponding to the current height value and height value change action further comprises:
under the condition that a target to be detected is detected to appear in point cloud data acquired by a millimeter wave radar for the first time, determining that a target occurrence event appears in a monitoring scene corresponding to the millimeter wave radar;
and after the target appearance event appears in the monitoring scene is detected, determining that the target disappearance event appears in the monitoring scene under the condition that the point cloud data does not detect the appearance of the target to be detected in a preset third time period.
7. The method for detecting the abnormal behavior of the target according to claim 6, wherein millimeter wave radars are arranged in a plurality of monitoring scenes, and the monitoring ranges of the millimeter wave radars are not overlapped; judging whether the risk state corresponding to the target observation event is an abnormal state comprises the following steps:
in a preset fourth time period, under the condition that the times of the occurrence of the target events and/or the occurrence of the target disappearance events in a preset first monitoring scene are larger than a preset seventh threshold value, determining that the target to be detected has an abnormal state;
in the fourth time period, counting the appearance duration of the target to be detected in the first monitoring scene according to the target appearance event and the target disappearance event, and determining that the target to be detected has an abnormal state under the condition that the appearance duration is greater than a preset eighth threshold;
determining that the target to be detected has an abnormal state under the condition that the target occurrence event and/or the target disappearance event do not appear in a preset second monitoring scene within a preset fifth time period;
and determining that the target to be detected has an abnormal state under the condition that the target occurrence event occurs in at least two monitoring scenes and the duration of the target occurrence event is greater than a preset ninth threshold.
8. The method for detecting the abnormal behavior of the target according to claim 1, wherein the step of acquiring multi-frame point cloud data of the target to be detected, which is acquired by the millimeter wave radar in a preset monitoring scene, comprises the steps of:
acquiring a point cloud data packet acquired by a millimeter wave radar in a preset monitoring scene, wherein the point cloud data comprises point cloud data of a plurality of detection points acquired by the millimeter wave radar;
clustering the point cloud data packets to obtain at least one point cloud cluster;
respectively judging whether the number of the detection points in each point cloud cluster is greater than a preset tenth threshold value or not, and respectively judging whether the maximum signal-to-noise ratio of the detection points in each point cloud cluster is greater than a preset eleventh threshold value or not;
under the condition that the number of the detection points in the point cloud cluster is larger than the tenth threshold value and the maximum signal-to-noise ratio of the detection points in the point cloud cluster is larger than the eleventh threshold value, taking the point cloud cluster as the point cloud data corresponding to the target to be detected;
and taking all point cloud clusters associated with the target to be detected in the point cloud clusters as multi-frame point cloud data corresponding to the target to be detected.
9. The method of detecting anomalous behavior in a target according to claim 3, further comprising:
and in the first time period, determining that the height value of the target to be detected maintains action under the condition that the difference value between the current height value and the reference height value is greater than the negative value of the first threshold and smaller than the second threshold.
10. The method of detecting anomalous behavior in a target according to claim 4, further comprising:
in the second time period, the height value of the target to be detected rises, the current height value of the target to be detected is larger than the fourth threshold value, and under the condition that the height value reduction action is not detected in the second time period, the target to be detected is determined to have a target rising event;
in the second time period, the height value of the target to be detected is increased, the current height value of the target to be detected is smaller than or equal to the fourth threshold, and a target height increasing event of the target to be detected is determined under the condition that no height value decreasing action is detected in the second time period;
and in the second time period, the target to be detected has any one of three height value change actions, and under the condition that no target falling event, no target height reducing event, no target rising event or no target height increasing event is detected in the second time period, the target to be detected is determined to have a target height maintaining event.
11. An apparatus for detecting abnormal behavior of a target, the apparatus comprising:
the acquisition module is used for acquiring multi-frame point cloud data, which are acquired by the millimeter wave radar in a preset monitoring scene and correspond to a target to be detected, wherein the multi-frame point cloud data comprise multi-frame historical frame point cloud data and current frame point cloud data;
the extraction module is used for extracting a historical height value corresponding to the target to be detected from each frame of historical frame point cloud data, and selecting the highest historical height value from a plurality of historical height values as a reference height value corresponding to the target to be detected;
the comparison module is used for extracting a current height value corresponding to the target to be detected from the current frame point cloud data, comparing the current height value with the reference height value and determining the change action of the height value of the target to be detected;
and the output module is used for determining whether the target to be detected has abnormal behaviors or not according to the current height value and the change action of the height value of the target to be detected.
12. A system for detecting anomalous behavior in a target, said system comprising: the system comprises a plurality of event observation modules, an event analysis module and an exception handling module; wherein the content of the first and second substances,
the event observation module comprises millimeter wave radars and a processor, the event observation modules are arranged in different monitoring scenes, the monitoring ranges of the millimeter wave radars are not overlapped, and the event observation module is used for observing abnormal behaviors occurring in the corresponding monitoring scenes;
the event analysis module is in communication connection with the event observation module and the exception handling module, and is used for analyzing the exception behavior obtained by the event observation module and sending a handling instruction corresponding to the exception behavior to the exception handling module;
the exception handling module is to perform a corresponding handling action according to the handling instruction.
13. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, and the processor is configured to execute the computer program to perform the method of detecting target abnormal behavior of any one of claims 1 to 10.
14. A storage medium having a computer program stored therein, wherein the computer program, when executed by a processor, implements the method of detecting target abnormal behavior of any one of claims 1 to 10.
CN202111115922.3A 2021-09-23 2021-09-23 Method, system, electronic device and storage medium for detecting abnormal target behavior Active CN113947867B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111115922.3A CN113947867B (en) 2021-09-23 2021-09-23 Method, system, electronic device and storage medium for detecting abnormal target behavior

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111115922.3A CN113947867B (en) 2021-09-23 2021-09-23 Method, system, electronic device and storage medium for detecting abnormal target behavior

Publications (2)

Publication Number Publication Date
CN113947867A true CN113947867A (en) 2022-01-18
CN113947867B CN113947867B (en) 2023-06-27

Family

ID=79328490

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111115922.3A Active CN113947867B (en) 2021-09-23 2021-09-23 Method, system, electronic device and storage medium for detecting abnormal target behavior

Country Status (1)

Country Link
CN (1) CN113947867B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114900419A (en) * 2022-04-22 2022-08-12 广州番禺电缆集团有限公司 Digital cable and method for intelligently reporting parameters
CN115376277A (en) * 2022-08-05 2022-11-22 美的集团(上海)有限公司 Health safety monitoring and early warning method, electronic equipment and storage medium
CN115968080A (en) * 2023-03-17 2023-04-14 森思泰克河北科技有限公司 Night lamp control method and device based on radar, radar and storage medium

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108670261A (en) * 2018-04-12 2018-10-19 深圳先进技术研究院 Motion state detection method, wearable device and device
CN109855575A (en) * 2019-01-23 2019-06-07 深圳慧安康科技有限公司 Intelligent apparatus, indoor human body 3-D positioning method and wisdom family implementation method
CN111491426A (en) * 2020-05-19 2020-08-04 河南职业技术学院 Intelligent light control system and control method
CN111887861A (en) * 2020-09-30 2020-11-06 西南交通大学 Millimeter wave radar-based integrated monitoring method for indoor personnel safety
CN112036269A (en) * 2020-08-17 2020-12-04 文思海辉元辉科技(无锡)有限公司 Fall detection method and device, computer equipment and storage medium
CN112396804A (en) * 2020-11-11 2021-02-23 湖南时变通讯科技有限公司 Point cloud-based data processing method, device, equipment and medium
TWM611867U (en) * 2021-01-15 2021-05-11 準旺科技股份有限公司 Accurate height detection equipment
CN112782664A (en) * 2021-02-22 2021-05-11 西南交通大学 Toilet fall detection method based on millimeter wave radar
CN112859033A (en) * 2021-02-23 2021-05-28 加特兰微电子科技(上海)有限公司 Target detection method, device and related equipment
CN112949552A (en) * 2021-03-22 2021-06-11 浙江大华技术股份有限公司 Fall detection processing method and device
CN113311428A (en) * 2021-05-25 2021-08-27 山西大学 Intelligent human body falling monitoring system based on millimeter wave radar and identification method
CN217213135U (en) * 2022-02-15 2022-08-16 长沙莫之比智能科技有限公司 Fall detection device based on millimeter wave radar

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108670261A (en) * 2018-04-12 2018-10-19 深圳先进技术研究院 Motion state detection method, wearable device and device
CN109855575A (en) * 2019-01-23 2019-06-07 深圳慧安康科技有限公司 Intelligent apparatus, indoor human body 3-D positioning method and wisdom family implementation method
CN111491426A (en) * 2020-05-19 2020-08-04 河南职业技术学院 Intelligent light control system and control method
CN112036269A (en) * 2020-08-17 2020-12-04 文思海辉元辉科技(无锡)有限公司 Fall detection method and device, computer equipment and storage medium
CN111887861A (en) * 2020-09-30 2020-11-06 西南交通大学 Millimeter wave radar-based integrated monitoring method for indoor personnel safety
CN112396804A (en) * 2020-11-11 2021-02-23 湖南时变通讯科技有限公司 Point cloud-based data processing method, device, equipment and medium
TWM611867U (en) * 2021-01-15 2021-05-11 準旺科技股份有限公司 Accurate height detection equipment
CN112782664A (en) * 2021-02-22 2021-05-11 西南交通大学 Toilet fall detection method based on millimeter wave radar
CN112859033A (en) * 2021-02-23 2021-05-28 加特兰微电子科技(上海)有限公司 Target detection method, device and related equipment
CN112949552A (en) * 2021-03-22 2021-06-11 浙江大华技术股份有限公司 Fall detection processing method and device
CN113311428A (en) * 2021-05-25 2021-08-27 山西大学 Intelligent human body falling monitoring system based on millimeter wave radar and identification method
CN217213135U (en) * 2022-02-15 2022-08-16 长沙莫之比智能科技有限公司 Fall detection device based on millimeter wave radar

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114900419A (en) * 2022-04-22 2022-08-12 广州番禺电缆集团有限公司 Digital cable and method for intelligently reporting parameters
CN114900419B (en) * 2022-04-22 2023-06-16 广州番禺电缆集团有限公司 Digital cable and method for intelligently reporting parameters
CN115376277A (en) * 2022-08-05 2022-11-22 美的集团(上海)有限公司 Health safety monitoring and early warning method, electronic equipment and storage medium
CN115968080A (en) * 2023-03-17 2023-04-14 森思泰克河北科技有限公司 Night lamp control method and device based on radar, radar and storage medium

Also Published As

Publication number Publication date
CN113947867B (en) 2023-06-27

Similar Documents

Publication Publication Date Title
CN113947867B (en) Method, system, electronic device and storage medium for detecting abnormal target behavior
Kosba et al. RASID: A robust WLAN device-free passive motion detection system
CN111887861B (en) Millimeter wave radar-based integrated monitoring method for indoor personnel safety
US20200166610A1 (en) Detection method, detection device, terminal and detection system
US20140362213A1 (en) Residence fall and inactivity monitoring system
JP5350721B2 (en) Resident monitoring system and resident monitoring method
CN108171930A (en) A kind of old man&#39;s exception tracing system applied to home for destitute
US10475326B2 (en) System and method for tracking interaction between monitored population and unmonitored population
WO2015077829A1 (en) System for monitoring subject movement
CN112073901A (en) Safety early warning method, system, device, equipment and storage medium
CN112184626A (en) Gesture recognition method, device, equipment and computer readable medium
WO2021130690A1 (en) Fall detection systems and methods
CN108416974B (en) Automatic alarm device and method based on wireless channel state information
CN111414828A (en) Abnormal aggregation identification method and device
CN114296073A (en) Abnormity warning method and system based on millimeter wave radar and electronic device
CN112396804A (en) Point cloud-based data processing method, device, equipment and medium
CN108245170B (en) Monitoring device for wearable device
Lee et al. Wireless sensing without sensors—An experimental approach
Valera et al. Improving the sensitivity of unobtrusive inactivity detection in sensor-enabled homes for the elderly
US20210225465A1 (en) Tracking individual user health using intrusion detection sensors
CN112731364B (en) Millimeter wave radar intelligent toilet management method, system, platform, medium and equipment
CN109326082B (en) Security monitoring method, device, system, server and air conditioner
JP7127301B2 (en) ABNORMAL SENSOR DETECTION METHOD, ABNORMAL SENSOR DETECTION SYSTEM AND INFORMATION PROCESSING DEVICE
CN113869183A (en) Fall detection method and device
CN115708081A (en) Non-contact daily living state monitoring method, system, computing device and medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
TA01 Transfer of patent application right

Effective date of registration: 20220623

Address after: Room 201, block B, Changtai Plaza, Lane 2889, Jinke Road, Pudong New Area, Shanghai 200120

Applicant after: Hengxuan Technology (Shanghai) Co.,Ltd.

Address before: 315500 room 701, No. 88, Dongfeng Road, Yuelin street, Fenghua City, Ningbo City, Zhejiang Province

Applicant before: NINGBO XITANG INFORMATION TECHNOLOGY Co.,Ltd.

TA01 Transfer of patent application right
GR01 Patent grant
GR01 Patent grant