Fall detection method based on infrared camera and millimeter wave radar and storage medium

Abstract

According to the fall detection method based on the infrared camera and the millimeter wave radar and the storage medium, a non-contact detection mode is adopted, all-weather and all-scene monitoring can be realized without wearing equipment, and the privacy of a user is effectively protected; the millimeter-wave radar and the infrared camera are combined, so that the efficient working capability of the system in a poor illumination condition and a night environment is ensured, and all-weather detection is really realized; when the system detects a falling event, communication can be carried out through a voice function, related personnel are notified in time, and it is ensured that the accident can be rapidly handled; the skeleton posture display function of the mobile terminal also enhances the visual monitoring capability of the system, and a guardian can check the real-time dynamic state in a monitoring area at any time through a mobile phone.

Description

Technical Field

[0001] The present invention relates to the technical field of fall detection for the elderly, and in particular to a fall detection method and a storage medium based on an infrared camera and a millimeter-wave radar. Background Art

[0002] Falls among the elderly are a common and difficult-to-prevent emergency event, especially secondary injuries caused by falls among elderly people living alone without timely discovery and rescue, which has a great negative impact on the elderly. Therefore, how to quickly and effectively detect the fall status of the elderly in the single-living elderly care scenario is a technical problem that needs to be urgently solved in healthy elderly care.

[0003] At present, fall detection methods mainly include fall detection systems based on visual images and fall detection systems based on wearable devices. Among them, although the detection system based on visual images has high accuracy, it cannot protect personal privacy well, and it is costly and requires a lot of calculations; the fall detection system based on wearable devices is a relatively mature detection method, but it requires users to wear wearable devices all day and in all scenarios, and cannot protect user privacy, which will inevitably cause inconvenience.

[0004] Therefore, it is necessary to provide a new method to solve the above technical problems. Summary of the invention

[0005] In order to achieve the above-mentioned purpose and other advantages of the present invention, the first purpose of the present invention is to provide a fall detection method based on an infrared camera and a millimeter-wave radar, comprising the following steps:

[0006] Acquire millimeter wave signals;

[0007] Detect whether there is a human body in the environment based on millimeter wave signals;

[0008] When the presence of a human body is detected in the environment, the fall detection function based on the infrared video stream is triggered;

[0009] Get real-time infrared video stream;

[0010] Dynamically analyzing the posture of the human skeleton through the infrared video stream to achieve fall detection;

[0011] When a fall event is detected, an alarm message is sent.

[0012] Furthermore, the method further comprises the steps of:

[0013] When the presence of a human body is detected in the environment, the changing characteristics of the reflected wave are analyzed to determine whether the target is within the monitoring range.

[0014] Furthermore, the step of dynamically analyzing the posture of a human skeleton through the infrared video stream to realize fall detection includes:

[0015] extracting a foreground object from the infrared video stream;

[0016] Extract joint points for each frame in the video;

[0017] Track the joint positions of multiple consecutive frames to calculate the dynamic information of the human body;

[0018] By monitoring the movement changes of joints, it is possible to determine whether a fall has occurred.

[0019] Furthermore, the step of extracting a foreground object from the infrared video stream comprises:

[0020] Extracting foreground objects from the infrared video stream using a static background model;

[0021] Perform image enhancement on the extracted foreground object;

[0022] Generate a binary image based on the set threshold.

[0023] Furthermore, the static background model construction step is also included:

[0024] A Gaussian mixture model is used to model the static background;

[0025] By continuously capturing multiple frames of images, a static background model is established;

[0026] The static background model is updated to adapt to dynamic changes.

[0027] Furthermore, the step of extracting joint points from each frame in the video includes:

[0028] Use a lightweight mobil enet neural network model combined with a self-attention mechanism to perform posture estimation and extract the positions of the main joints of the human body, including the head, shoulders, elbows, knees, and ankles;

[0029] Standardize the joint point coordinates and unify them into a relative coordinate system between 0 and 1.

[0030] Furthermore, the step of tracking the joint point positions of a plurality of consecutive frames includes:

[0031] The Kalman filter is used to track the joint position in multiple consecutive frames.

[0032] Furthermore, the step of determining whether a fall has occurred by monitoring the movement changes of the joints includes:

[0033] Detect the height change of the head joint in a short period of time. When the change exceeds the threshold and lasts for a long time, the fall detection is triggered.

[0034] The dynamic model and XGboost machine learning algorithm are used to analyze the movement of the current joints in real time and give the probability of falling.

[0035] Furthermore, the method further comprises the steps of:

[0036] Record a video before and after the fall and store it on a local hard drive or cloud server;

[0037] In response to a user's remote monitoring request, a real-time video image of the monitoring area is provided. The video image is processed for privacy and does not display a real person image, but only displays skeleton contour information extracted from a real person in the scene.

[0038] The alarm record of each fall is saved to facilitate users to view historical alarm information and related video recordings.

[0039] A second object of the present invention is to provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0040] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0041] The present invention provides a fall detection method and storage medium based on infrared cameras and millimeter-wave radars. It adopts a non-contact detection method and can achieve all-weather and all-scenario monitoring without wearing equipment, effectively protecting the privacy of users. The combination of millimeter-wave radars and infrared cameras ensures the system's efficient working ability in poor lighting conditions and nighttime environments, truly achieving all-weather detection. When the system detects a fall event, it can communicate through the voice function to notify relevant personnel in a timely manner to ensure that the accident can be handled quickly. The skeleton posture display function of the mobile terminal also enhances the system's visual monitoring capability, and guardians can view the real-time dynamics in the monitoring area at any time through their mobile phones.

[0042] The present invention realizes accurate fall detection and privacy protection through the combination of millimeter wave radar and infrared video images. Millimeter wave radar can effectively sense the presence of the human body, and detect the movement trajectory of the target in real time through the electromagnetic wave reflection information it emits. It is particularly suitable for working at night or in low light environments. The infrared camera dynamically analyzes the skeleton posture of the human body based on the collected infrared video images. The system can make a quick judgment when a fall occurs, and display the skeleton posture in real time through the mobile terminal, providing accurate monitoring information for the guardian.

[0043] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention and implement it according to the contents of the specification, the following is a detailed description of the preferred embodiments of the present invention in conjunction with the accompanying drawings. The specific implementation of the present invention is given in detail by the following embodiments and their accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0045] Figure 1 The process of fall detection method based on infrared camera and millimeter wave radar Figure 1 ;

[0046] Figure 2 The process of fall detection method based on infrared camera and millimeter wave radar Figure 2 ;

[0047] Figure 3 It is a flow chart of fall detection;

[0048] Figure 4 Extracting flow charts for foreground objects;

[0049] Figure 5 Build a flow chart for the static background model;

[0050] Figure 6 Extract flow charts for joint points;

[0051] Figure 7 Provide a flow chart for joint point monitoring;

[0052] Figure 8 It is a schematic diagram of computer equipment;

[0053] Fig. 9 A schematic diagram of a computer-readable storage medium. DETAILED DESCRIPTION

[0054] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. It should be noted that, under the premise of no conflict, the embodiments or technical features described below can be arbitrarily combined to form a new embodiment.

[0055] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0056] The figure numbers in this application are only used to distinguish the various steps in the scheme, and are not used to limit the execution order of the various steps. The specific execution order is subject to the description in the specification.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0058] Example 1

[0059] A fall detection method based on infrared camera and millimeter wave radar is used to detect falls of the elderly. It uses a non-contact video fall detection device to sense the presence of the human body through millimeter wave radar, evaluate the fall status based on infrared video images, communicate through voice when a fall occurs, and display the skeleton posture in real time on the mobile terminal. Figure 1 , Figure 2 As shown, the following steps are included:

[0060] S1. Acquire millimeter wave signals;

[0061] S2. Detect whether there is a human body in the environment based on the millimeter wave signal;

[0062] In this embodiment, the millimeter wave radar detection module determines the presence of a human body by emitting millimeter wave signals and receiving reflected signals. Millimeter waves have the characteristics of strong penetration and high resolution, and can accurately detect targets in complex environments, especially in low-light or completely dark night scenes. When the radar detects the presence of a human body in the environment, it quickly determines whether the target is within the monitoring range by analyzing the changing characteristics of the reflected wave. This radar detection technology can not only sense the static state of the human body, but also capture tiny movements, such as breathing or subtle limb movements, thereby ensuring the sensitivity and accuracy of the detection.

[0063] To optimize the overall performance of the system, the millimeter wave radar module works closely with the video streaming monitoring system.

[0064] S3. When the presence of a human body is detected in the environment, the fall detection function based on the infrared video stream is triggered; this can significantly reduce the amount of continuous monitoring calculations for the video stream in an unmanned environment. Through this mechanism, the system avoids the high computing requirements of video monitoring around the clock, effectively reduces the energy consumption of the equipment, and extends the service life of the system. In addition, the high sensitivity of the millimeter-wave radar enables the system to maintain efficient and stable fall detection capabilities under complex lighting conditions, especially at night or in low-light environments, ensuring full-scene monitoring coverage.

[0065] In some embodiments, the steps include:

[0066] When the presence of a human body is detected in the environment, the changing characteristics of the reflected wave are analyzed to determine whether the target is within the monitoring range. When the target is determined to be within the monitoring range, an infrared camera is used to capture the image in the monitoring area.

[0067] This embodiment uses an infrared camera to capture images in the monitoring area. The infrared camera can work effectively at night and in low light environments, so it can provide support for all-weather monitoring. Specifically, the infrared camera captures real-time video streams, and the resolution and frame rate of the video frames can be adjusted according to actual needs. Generally, a frame rate of 30 frames per second can ensure good real-time performance while ensuring image smoothness.

[0068] S4, obtaining real-time infrared video stream;

[0069] S5. Dynamically analyzing the posture of the human skeleton through the infrared video stream to achieve fall detection;

[0070] In some embodiments, Figure 3 As shown, the step of dynamically analyzing the human skeleton posture through the infrared video stream to realize fall detection includes:

[0071] S51, extracting a foreground object from the infrared video stream;

[0072] Furthermore, if Figure 4 As shown, the step of extracting the foreground object from the infrared video stream includes:

[0073] S511, extracting a foreground object from the infrared video stream by using a static background model;

[0074] S512, performing image enhancement on the extracted foreground object;

[0075] Since the resolution of infrared cameras may be affected in low-light conditions, the system ensures that the human body outlines in the image are clear by applying image enhancement techniques such as contrast enhancement and filtering methods.

[0076] S513: Generate a binary image according to the set threshold.

[0077] For example, the pixel value of the foreground area (such as a human body) is 255, and the pixel value of the background area is 0. Binary images can help simplify subsequent pose estimation and joint point detection.

[0078] Furthermore, if Figure 5 As shown, it also includes the steps of building a static background model:

[0079] S510, using a Gaussian mixture model (GMM) to model the static background;

[0080] S520: Establish a static background model by continuously capturing multiple frames of images; compare the new frame image with the background model to extract the foreground object.

[0081] Through the background subtraction method, the moving foreground (such as the human body) is separated from the static background.

[0082] S530: Update the static background model to adapt to dynamic changes.

[0083] The update frequency of background modeling can be adjusted according to the actual application environment to adapt to dynamic changes such as lighting changes.

[0084] Posture estimation is a critical step in the system. The posture of the human body is inferred based on the position of the joints, thus providing an important basis for subsequent fall detection. This embodiment implements posture estimation through the following steps:

[0085] S52, extracting joint points from each frame in the video;

[0086] Furthermore, if Figure 6 As shown, the step of extracting joint points for each frame in the video includes:

[0087] S521, using a lightweight mobile neural network model combined with a self-attention mechanism to perform posture estimation and extract the positions of the main joints of the human body, wherein the joints include the head, shoulders, elbows, knees, ankles, etc.;

[0088] Through the relative positions of joints, the system can construct a human skeleton model, which can reflect the posture and movement trajectory of the human body.

[0089] Since the images captured by the infrared camera under different conditions may have different scales and resolutions, S522 standardizes the joint point coordinates and unifies them into a relative coordinate system between 0 and 1 to ensure the consistency of posture estimation in different devices and environments.

[0090] S53, tracking the joint positions of multiple consecutive frames to calculate the dynamic information of the human body; by tracking the movement trajectory of the joint points, calculating the dynamic information such as the speed and direction of the human body movement.

[0091] Furthermore, the step of tracking the joint point positions of a plurality of consecutive frames includes:

[0092] Kalman filtering is used to track the joint positions of multiple consecutive frames to ensure that good joint tracking effects can be maintained when the target is occluded or image noise occurs.

[0093] S54. Determine whether a fall has occurred by monitoring the movement changes of the joints.

[0094] Since a fall is usually accompanied by a sharp change in human posture, especially a sudden drop in head height, by monitoring these changes, it is possible to effectively determine whether a fall has occurred.

[0095] The extracted joint point motion vector ω j ={Ψ j ,θ j} for analysis, where Ψ j is the amplitude of the joint motion, θ j It is the direction of movement. Through the changes in the joint movement of consecutive frames, the system can determine the movement trend of the human body. Special attention is paid to the height changes of key joints such as the head, ankles, and knees. Generally speaking, a sharp drop in the height of the head and contact with the ground is an important sign of a fall.

[0096] Furthermore, if Figure 7 As shown, the step of determining whether a fall has occurred by monitoring the movement changes of the joints includes:

[0097] Several thresholds and rules are set to assist in the judgment of falling behavior. S541. Detect the height change of the head joint in a short period of time. When the change exceeds the threshold and lasts for a long time, the fall detection is triggered;

[0098] S542. Analyze the movement of the current joints in real time through the dynamic model and XGboost machine learning algorithm, and give the probability of falling.

[0099] In order to improve the accuracy of fall detection, the system introduces a dynamic model and XGboost machine learning algorithm. By analyzing the dynamic motion characteristics of human joints, it can effectively distinguish normal behaviors (such as sitting and walking) from abnormal behaviors (such as falling). The model is trained with a large number of normal behavior and fall samples to learn the characteristics of different behaviors. The trained model can analyze the motion of the current joints in real time and give the probability of falling.

[0100] S6. When a fall event is detected, an alarm message is sent.

[0101] When a fall event is detected, the system will immediately perform an alarm operation to ensure that relevant personnel or institutions are notified in time for rescue. After a fall event occurs, the system will immediately send an alarm signal to the preset contact through the network. The alarm information includes the time and location of the fall, as well as real-time video screenshots, etc. The system provides a variety of alarm methods, including SMS, phone calls, emails, push notifications, etc. Users can select the appropriate alarm method in the system settings.

[0102] In some embodiments, the steps include:

[0103] Record a video before and after the fall and store it on a local hard drive or cloud server; these video records can be used for post-analysis and verification, and can also help medical staff understand the specific circumstances of the accident.

[0104] In order to facilitate users to remotely monitor and control, the system provides a user-side interface, and users can operate the system through mobile devices (such as mobile phones, tablets) or computers. The functions of the user-side include:

[0105] Real-time video monitoring based on skeleton information: Users can view the video footage of the monitored area in real time through mobile applications or web pages. The footage is processed for privacy and does not display images of real people. It only displays the skeleton outline information extracted from the real person in the scene.

[0106] Accordingly, in response to the user's remote monitoring request, a real-time video picture of the monitoring area is provided.

[0107] The alarm record of each fall is saved to facilitate users to view historical alarm information and related video recordings through the application.

[0108] The installation and deployment of the fall detection system based on infrared cameras is very flexible and can be applied to various scenarios such as homes, nursing homes, hospitals, etc. It is installed at a high place in the monitoring area to ensure that the field of view of the monitoring area is wide enough to capture the dynamics of the entire monitoring site. The number of cameras can be adjusted according to the size of the site. Multiple cameras can be connected through a local area network to jointly complete the monitoring task of a large area. Users can complete the initial configuration through the setting interface provided by the device or the mobile application. The configuration content includes network connection, camera angle adjustment, monitoring area setting, etc. The system supports remote configuration and upgrade. Users can update the software algorithm through the cloud to ensure that the system can be continuously upgraded. Before the system is officially put into operation, debugging and testing are required to ensure that the infrared camera can work normally under different lighting conditions, and the sensitivity and accuracy of the posture estimation and fall detection modules meet the expected results.

[0109] Example 2

[0110] A computer device 200, such as Figure 8 As shown, it includes a memory 210, a processor 220, and a computer program 230 stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a fall detection method based on an infrared camera and a millimeter-wave radar are implemented. For a detailed description of the method, reference may be made to the corresponding description in the above method embodiment, which will not be repeated here.

[0111] Example 3

[0112] A computer readable storage medium such as Fig. 9 As shown, a computer program is stored thereon, and when the computer program is executed by the processor, the steps of a fall detection method based on an infrared camera and a millimeter wave radar are implemented. For a detailed description of the method, reference may be made to the corresponding description in the above method embodiment, and no further description is given here.

[0113] Example 4

[0114] A computer program product, the computer program product comprising a computer program, wherein when the computer program is executed by a processor, the steps of a fall detection method based on an infrared camera and a millimeter wave radar are implemented. For a detailed description of the method, reference may be made to the corresponding description in the above method embodiment, which will not be repeated here.

[0115] The number of devices and processing scales described here are used to simplify the description of the present invention. Applications, modifications and variations of the present invention will be obvious to those skilled in the art.

[0116] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and implementation modes. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.

[0117] The apparatus, computer device, non-volatile computer storage medium and method provided in the embodiments of this specification correspond to each other, and therefore, the apparatus, computer device and non-volatile computer storage medium also have similar beneficial technical effects as the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding apparatus, computer device and non-volatile computer storage medium will not be repeated here.

[0118] Those skilled in the art also know that, in addition to implementing the controller in a purely computer-readable program code, the controller can be made to implement the same function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, such a controller can be considered as a hardware component, and the devices for implementing various functions included therein can also be considered as structures within the hardware component. Or even, the devices for implementing various functions can be considered as both software units for implementing the method and structures within the hardware component.

[0119] The systems, devices or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. For the convenience of description, the above devices are described separately by functions in various units. Of course, when implementing one or more embodiments of this specification, the functions of each unit can be implemented in the same or more software and / or hardware.

[0120] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification may be in the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the embodiments of this specification may be in the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0121] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0122] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0124] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0125] The specification may be described in the general context of computer-executable instructions executed by a computer, such as program units. Generally, program units include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program units may be located in local and remote computer storage media, including storage devices.

[0126] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0127] The above description is only an embodiment of this specification and is not intended to limit one or more embodiments of this specification. For those skilled in the art, one or more embodiments of this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included in the scope of the claims of one or more embodiments of this specification.

Claims

1. A fall detection method based on infrared camera and millimeter wave radar, characterized in that: The following steps are involved: Acquire millimeter wave signals; Detect whether there is a human body in the environment based on millimeter wave signals; When the presence of a human body is detected in the environment, the fall detection function based on the infrared video stream is triggered; Get real-time infrared video stream; Dynamically analyzing the posture of the human skeleton through the infrared video stream to achieve fall detection; When a fall event is detected, an alarm message is sent.

2. A fall detection method based on infrared camera and millimeter wave radar as claimed in claim 1, characterized in that: Also includes the steps: When the presence of a human body is detected in the environment, the changing characteristics of the reflected wave are analyzed to determine whether the target is within the monitoring range.

3. A fall detection method based on infrared camera and millimeter wave radar as claimed in claim 1, characterized in that: The step of dynamically analyzing the human skeleton posture through the infrared video stream to realize fall detection includes: extracting a foreground object from the infrared video stream; Extract joint points for each frame in the video; Track the joint positions of multiple consecutive frames to calculate the dynamic information of the human body; By monitoring the movement changes of joints, it is possible to determine whether a fall has occurred.

4. A fall detection method based on infrared camera and millimeter wave radar as claimed in claim 3, characterized in that: The step of extracting a foreground object from the infrared video stream comprises: Extracting foreground objects from the infrared video stream using a static background model; Perform image enhancement on the extracted foreground object; Generate a binary image based on the set threshold.

5. A fall detection method based on infrared camera and millimeter wave radar as claimed in claim 4, characterized in that: It also includes the static background model construction steps: A Gaussian mixture model is used to model the static background; By continuously capturing multiple frames of images, a static background model is established; The static background model is updated to adapt to dynamic changes.

6. A fall detection method based on infrared camera and millimeter wave radar as claimed in claim 3, characterized in that: The step of extracting joint points from each frame in the video comprises: Use a lightweight mobile neural network model combined with a self-attention mechanism to perform posture estimation and extract the positions of the main joints of the human body, including the head, shoulders, elbows, knees, and ankles; Standardize the joint point coordinates and unify them into a relative coordinate system between 0 and 1.

7. A fall detection method based on infrared camera and millimeter wave radar as claimed in claim 3, characterized in that: The step of tracking the joint point positions of a plurality of consecutive frames comprises: The Kalman filter is used to track the joint positions in multiple consecutive frames.

8. A fall detection method based on infrared camera and millimeter wave radar as claimed in claim 6, characterized in that: The step of determining whether a fall has occurred by monitoring the movement changes of the joints comprises: Detect the height change of the head joint in a short period of time. When the change exceeds the threshold and lasts for a long time, the fall detection is triggered. The dynamic model and XGboost machine learning algorithm are used to analyze the movement of the current joints in real time and give the probability of falling.

9. A fall detection method based on infrared camera and millimeter wave radar as claimed in claim 6, characterized in that: Also includes the steps: Record a video before and after the fall and store it on a local hard drive or cloud server; In response to a user's remote monitoring request, a real-time video image of the monitoring area is provided. The video image is processed for privacy and does not display a real person image, but only displays skeleton contour information extracted from a real person in the scene. The alarm record of each fall is saved to facilitate users to view historical alarm information and related video recordings.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images