An emergency call method, device, equipment and medium based on artificial intelligence

Abstract

The present invention relates to the technical field of personnel behavior monitoring, and specifically, to an emergency call method, device, equipment and medium based on artificial intelligence. The present invention marks and grades the more dangerous areas in the hospital in advance, and configures key monitoring groups of different age groups based on the grades, and configures corresponding cameras for monitoring the area. When multiple elderly people enter the vicinity of the corresponding high-risk area, the key monitoring objects are screened out based on the human parameter monitoring equipment worn by the elderly. After the key monitoring objects enter the high-risk area, the camera corresponding to the high-risk area is used to zoom and focus on monitoring the body posture of the elderly when passing through the high-risk area, and then detect whether the elderly have fallen, etc. Since the entire process is automatically analyzed and intelligently determined through high-definition video monitoring, Internet of Things sensors and big data, the personnel monitoring task volume in the nursing home or sanatorium is effectively reduced, and the labor cost expenditure is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of personnel behavior monitoring, and more specifically, to an emergency call method, device, equipment and medium based on artificial intelligence. Background Art

[0002] In today's society, with the aging population becoming increasingly serious, improving the operational efficiency and service quality of nursing homes and sanatoriums, as important places for elderly care and health management, has become a key issue that needs to be addressed. Traditional manual monitoring methods are not only labor-intensive but also difficult to provide comprehensive monitoring for every elderly person.

[0003] The rapid development of artificial intelligence (AI) technology in recent years has provided new solutions to these problems. Therefore, leveraging technologies like high-definition video surveillance, IoT sensors, and big data analytics to automatically identify, analyze, and issue early warnings about elderly people's daily behaviors, particularly in high-risk areas like plaza steps, has become a pressing issue. Summary of the Invention

[0004] The purpose of the present invention is to provide an artificial intelligence-based emergency call method, device, equipment and medium to improve the problem that traditional manual monitoring is difficult to monitor the elderly in some high-risk activity areas in nursing homes or sanatoriums around the clock without blind spots.

[0005] In order to achieve the above objectives, the embodiments of the present application provide the following technical solutions:

[0006] On the one hand, an embodiment of the present application provides an artificial intelligence-based emergency call method, which is applicable to a human parameter monitoring device, wherein the human parameter monitoring device is used to monitor the physiological data and location information of the wearer. The method includes: responding to the location information fed back by multiple human parameter monitoring devices, and marking the human parameter monitoring device located in a dangerous radiation zone as a risk device based on the location information; retrieving the wearer's health data corresponding to the risk device, and screening out high-risk devices from multiple risk devices based on the danger level corresponding to the dangerous radiation zone; constructing a forward range based on the running speed and vector of the high-risk device, and when the forward range intersects with the high-risk zone, determining the focus probability based on the historical motion trajectory of the high-risk device, and when the focus probability is greater than a first threshold, generating a first focus instruction based on the position code of the high-risk area, so that the corresponding surveillance camera focuses on the high-risk area according to a preset focal length, and feeds back a surveillance video; determining the body posture of the wearer of the high-risk device when passing through the high-risk area based on the surveillance video, and making an emergency call to the corresponding service personnel if the body posture is abnormal.

[0007] Optionally, determining the focus probability based on the historical motion trajectory of the high-risk device includes:

[0008] Dividing the historical running trajectory into a first running trajectory and a second running trajectory based on a time limit, wherein the first running trajectory is the running trajectory within the dangerous radiation zone today, and the second running trajectory is the activity trajectory within the past 1-2 weeks;

[0009] Marking a plurality of reference points on the first running trajectory based on a preset time interval, generating a forward range corresponding to each reference point, and calculating a proportion of reference points that intersect the forward range with the high-risk area;

[0010] The number of times the second running trajectory crosses the high-risk area is counted, and the probability of the current high-risk device crossing the high-risk area is calculated based on the proportion and the number of times through a weighted algorithm, which is the focus probability.

[0011] Optionally, determining the body posture of the wearer of the high-risk device when passing through the high-risk area based on the surveillance video includes:

[0012] Construct a key reference plane based on the building characteristics in the high-risk area, and divide the key reference plane into location zones, thereby obtaining the location parameter range corresponding to each location zone;

[0013] The PAF algorithm is used to construct the joint points of multiple characters in the surveillance video, and the shoulder and foot joint points of each character are extracted;

[0014] Based on the inclusion relationship between the foot joint point of each character and the key reference plane, characters that are not within the key reference plane are eliminated, thereby obtaining a plurality of first shoulder joint points and first foot joint points corresponding to the characters within the key reference plane;

[0015] Determining the location zone where the wearer of the high-risk device is located based on the location information fed back by the high-risk device and the location parameter range corresponding to each location zone, and then finding the second shoulder joint point and the second foot joint point corresponding to the wearer of the high-risk device among multiple people located in the key reference plane;

[0016] The body posture of the wearer of the high-risk device when passing through the high-risk area is determined based on the position changes of the second shoulder joint point and the second foot joint point when passing through the high-risk area.

[0017] Optionally, the determining of the body posture of the wearer of the high-risk device when passing through the high-risk area based on the position changes of the second shoulder joint point and the second foot joint point when passing through the high-risk area includes:

[0018] detecting the integrity of the joint points corresponding to the wearer of the high-risk device in real time, and identifying the second shoulder joint point and the second foot joint point if they are intact, and monitoring the height difference between the two in real time, and calculating the time it takes for the height difference to collapse to the first threshold if the height difference is less than a first threshold, recording it as a first time, and identifying the head pixel point coordinate set and the foot pixel point coordinate set of the suspected wrestler using a shape-based image feature extraction algorithm after the height difference is less than the first threshold;

[0019] Determine the key reference plane where the feet of the suspected wrestler are located based on the set of foot pixel coordinates, and record it as the wrestling reference plane;

[0020] The estimated collapse degree of the suspected wrestler is determined based on the wrestling reference surface, the foot pixel coordinate set and the head pixel coordinate set, and then the suspected wrestler is determined to be in a falling state or a sitting state based on the estimated collapse degree and the first time length weighting.

[0021] Optionally, the step of determining an estimated collapse degree of the suspected wrestling character based on the wrestling reference surface, the foot pixel coordinate set, and the head pixel coordinate set includes:

[0022] Determine the first center coordinates based on the set of pixel coordinates of the lower body, and determine the second center coordinates based on the set of pixel coordinates of the head;

[0023] The corresponding collapse degree estimation is evaluated based on the longitudinal arrangement order of the first center coordinate, the second center coordinate and the wrestling reference surface and the height difference between them. The greater the difference between the longitudinal arrangement order and the preset arrangement order, the higher the collapse degree estimation. The greater the difference between the height difference between the first center coordinate, the second center coordinate and the wrestling reference surface and the preset threshold range, the higher the collapse degree estimation. The collapse degree estimation is used to characterize the degree of aggregation between joint points. The greater the collapse degree estimation, the greater the probability of the target person falling.

[0024] In a second aspect, this embodiment provides an artificial intelligence-based emergency call device, characterized in that the device includes:

[0025] a first calculation module, configured to respond to position information fed back by a plurality of human parameter monitoring devices and mark a human parameter monitoring device located in a dangerous radiation zone as a risky device based on the position information;

[0026] A second calculation module is used to retrieve the wearer's health data corresponding to the risky device and screen out high-risk devices from multiple risky devices based on the risk level corresponding to the dangerous radiation area;

[0027] A third calculation module is configured to construct a forward range based on the operating speed and vector of the high-risk device, and when the forward range intersects the high-risk area, determine a focus probability based on the historical motion trajectory of the high-risk device. When the focus probability is greater than a first threshold, a first focus instruction is generated based on the position code of the high-risk area, so that the corresponding surveillance camera focuses on the high-risk area according to a preset focal length and feeds back surveillance video.

[0028] The early warning module is used to determine the body posture of the wearer of high-risk equipment when passing through high-risk areas based on the surveillance video, and to call the corresponding service personnel urgently if the body posture is abnormal.

[0029] In a third aspect, an embodiment of the present application provides an emergency call device based on artificial intelligence, which includes a memory and a processor.

[0030] The memory is used to store computer programs; the processor is used to implement the steps of the above-mentioned artificial intelligence-based emergency call method when executing the computer program.

[0031] In a fourth aspect, an embodiment of the present application provides a medium having a computer program stored thereon, which implements the steps of the above-mentioned artificial intelligence-based emergency call method when executed by a processor.

[0032] The beneficial effects of the present invention are:

[0033] The present invention marks and grades the more dangerous areas in the hospital in advance, configures key monitoring groups of different age groups based on the grades, and configures corresponding cameras to monitor the areas. When multiple elderly people enter the vicinity of the corresponding high-risk areas, the key monitoring objects are screened out based on the human parameter monitoring equipment worn by the elderly. After the key monitoring objects enter the high-risk areas, the cameras corresponding to the high-risk areas are used to zoom and focus on monitoring the body postures of the elderly passing through the high-risk areas, and then detect whether the elderly have fallen. Since the entire process is automatically analyzed and intelligently determined through high-definition video monitoring, Internet of Things sensors and big data, the workload of personnel monitoring in nursing homes or sanatoriums is effectively reduced, and the labor cost expenditure is reduced.

[0034] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or understood by practicing the embodiments of the present invention. The purposes and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0036] Figure 1 This is a flow chart of an emergency call method based on artificial intelligence according to an embodiment of the present invention;

[0037] Figure 2 This is a structural diagram of an emergency call device based on artificial intelligence described in an embodiment of the present invention. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0039] It should be noted that similar reference numerals or letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0040] Embodiment 1:

[0041] like Figure 1 As shown, this embodiment provides an emergency call method based on artificial intelligence, which is applicable to a human body parameter monitoring device, wherein the human body parameter monitoring device is used to monitor the physiological data and location information of the wearer, and the method includes step S100, step S200, step S300 and step S400.

[0042] Step S100: In response to the location information fed back by multiple human parameter monitoring devices, in order to provide more accurate positioning, an additional positioning base station is configured in the hospital, and based on the location information, the human parameter monitoring devices located in the dangerous radiation zone are marked as risky devices. Before the overall monitoring system is put into place, it is necessary to manually classify and number the buildings in the hospital according to their distribution and classify them into different levels. The classification of the dangerous level of the place is mainly used to screen the key monitoring groups during the later monitoring. Since the age span of users in nursing homes or sanatoriums is large, the early staff will classify the occupants according to their age and specific physical condition when entering the information of the occupants, thereby facilitating the intelligent screening and monitoring of the later system. That is, in the same place, such as an activity square with a small number of stairs, the danger level is different for people of different age groups, and the camera will give priority to monitoring the older or physically poorer people in the area;

[0043] Step S200: retrieve the health data of the wearer corresponding to the risk device, and screen out high-risk devices from multiple risk devices based on the risk level corresponding to the dangerous radiation zone. The dangerous radiation zone is an area artificially divided in the early stage, which is used to represent the area close to the high-risk zone, such as the flat area close to the stairs. Based on the risk level corresponding to the dangerous radiation zone, the system will determine whether there are elderly people who are older or whose physical fitness is not suitable for activities in the current dangerous zone in multiple dangerous radiation zones, and mark the human parameter monitoring devices worn by the corresponding elderly people as high-risk devices;

[0044] Step S300: Based on the running speed and vector of the high-risk equipment, a forward range is constructed, and when the forward range intersects with the high-risk area, the focus probability is determined based on the historical motion trajectory of the high-risk equipment, and when the focus probability is greater than a first threshold, a first focus instruction is generated based on the position code of the high-risk area, so that the corresponding surveillance camera focuses on the high-risk area according to a preset focal length and feeds back the surveillance video. In some special occasions, such as on a raised platform for sunbathing or at the edge of a lower step on a large flat area; the forward range is constructed to focus the camera in advance, thereby increasing the detection accuracy of the later posture detection algorithm. In principle, the camera is used to monitor the entire area and will only focus when a dangerous person is about to enter the preset dangerous area. The focal length is preset and cannot be moved or locked according to the person, etc.; the forward range is the area within 45 degrees to the left and right of the current vector direction, and the size of the fan-shaped forward range is proportional to the current running speed;

[0045] Step S400: Determine the body posture of the wearer of the high-risk device when passing through the high-risk area based on the surveillance video, and call the corresponding service personnel urgently if the body posture is abnormal.

[0046] The artificial intelligence-based emergency call described in this embodiment marks and grades the more dangerous areas in the hospital in advance, configures key monitoring groups of different age groups based on the grades, and configures corresponding cameras to monitor the area. When multiple elderly people enter the vicinity of the corresponding high-risk area, key monitoring objects are screened out based on the human parameter monitoring equipment worn by the elderly. After the key monitoring objects enter the high-risk area, the camera corresponding to the high-risk area is used to zoom and focus on monitoring the body posture of the elderly when passing through the high-risk area, and then detect whether the elderly have fallen. Since the entire process is automatically analyzed and intelligently determined through high-definition video monitoring, Internet of Things sensors and big data, the workload of personnel monitoring in nursing homes or sanatoriums is effectively reduced, and labor cost expenditure is reduced.

[0047] For some elderly people who regularly wander around the stairs in the square, it is necessary to determine the probability of them going down the stairs based on their historical movement trajectory and dispatch surveillance cameras based on this probability. For some people who often wander around and rarely go down the stairs, it is unnecessary to dispatch cameras frequently, which may cause damage to the cameras. That is, the implementation method of determining the focus probability based on the historical movement trajectory of high-risk equipment in step S300 is as follows:

[0048] Step S310: Divide the historical running trajectory into a first running trajectory and a second running trajectory based on a time limit, wherein the first running trajectory is the running trajectory within the dangerous radiation zone today, and the second running trajectory is the activity trajectory within the past 1-2 weeks;

[0049] Step S320: Mark multiple reference points on the first running trajectory based on preset time intervals, generate a forward range corresponding to each reference point, and calculate the proportion of reference points whose forward range intersects with the high-risk area. That is, if the vehicle is hovering near a step but always maintains a certain distance from the step, the proportion of reference points whose forward range intersects with the high-risk area is low, and the probability of descending the step is low.

[0050] Step S330: Count the number of times the second running trajectory crosses the high-risk area, and calculate the probability of the current high-risk device crossing the high-risk area through a weighted algorithm based on the proportion and the number of times, which is the focus probability.

[0051] The method for determining the body posture of the wearer of the high-risk device when passing through the high-risk area based on the surveillance video in step S400 is as follows:

[0052] Step S410: Construct a key reference plane based on the building features in the high-risk area, and partition the key reference plane into different locations to obtain a position parameter range corresponding to each location partition. For a plaza staircase, the key reference plane is the plane corresponding to each step.

[0053] Step S420: construct the joint points of multiple characters in the surveillance video using the PAF algorithm, and extract the shoulder joint points and foot joint points of each character;

[0054] Step S430: based on the inclusion relationship between the foot joint point of each character and the key reference plane, characters that are not within the key reference plane are eliminated, thereby obtaining a plurality of first shoulder joint points and first foot joint points corresponding to the characters within the key reference plane;

[0055] Step S440: Determine the location zone where the wearer of the high-risk device is located based on the location information fed back by the high-risk device and the location parameter range corresponding to each location zone, and then find the second shoulder joint point and the second foot joint point corresponding to the wearer of the high-risk device among multiple people located within the key reference plane. That is, when multiple elderly people approach the edge of the stairs at the same time, there are both those who do not need to be monitored and those who need to be monitored in the crowd. It is necessary to further mark the key monitored subjects in the surveillance video. Therefore, it is necessary to divide the key reference plane into location zones, and mark the target person in the video based on the location data range threshold corresponding to each zone and the location information fed back by the human body parameter monitoring device;

[0056] Step S450: Determine the body posture of the wearer of the high-risk device when passing through the high-risk area based on the position changes of the second shoulder joint point and the second foot joint point when passing through the high-risk area.

[0057] The method for determining the body posture of the wearer of the high-risk device when passing through the high-risk area based on the position changes of the second shoulder joint point and the second foot joint point when passing through the high-risk area in step S450 is as follows:

[0058] Step S451: Detect the integrity of the joint points corresponding to the wearer of the high-risk device in real time, and if they are intact, identify the second shoulder joint point and the second foot joint point, and monitor the height difference between the two in real time. If the height difference is less than a first threshold, calculate the time it takes for the height difference to collapse to the first threshold, recorded as the first time, and identify the head pixel point coordinate set and the foot pixel point coordinate set of the suspected wrestler using a shape-based image feature extraction algorithm after the height difference is less than the first threshold.

[0059] Step S452: determining the key reference plane where the feet of the suspected wrestler are located based on the foot pixel coordinate set, and recording it as the wrestling reference plane;

[0060] Step S453: Determine an estimated degree of collapse of the suspected wrestler based on the wrestling reference surface, the foot pixel coordinate set, and the head pixel coordinate set, and then determine whether the suspected wrestler is in a falling state or a sitting state based on the estimated degree of collapse and the first duration weighting.

[0061] To avoid false falls in human posture recognition, how to distinguish whether the target person is sitting down on the edge of a staircase or has fallen down? These two phenomena are quite similar in human posture recognition based on the PAF algorithm, that is, the joints of the person will converge / collapse towards the center. Therefore, this embodiment further determines whether the target person is sitting down or falling based on the collapse speed and the positional relationship between the key joints after collapse and the corresponding reference plane.

[0062] Secondly, it should be noted that the PAF algorithm is better at recognizing the standing posture of the human body, but it is more difficult to recognize the sitting or wrestling posture. Therefore, the recognition after collapse in this embodiment is based on the shape-based image features to identify key positions, which are used to find the pixel clusters that best reflect the head and foot areas in the video pattern, and determine the body posture based on the sorting relationship between the center positions of the pixel clusters and the distance relationship between each other.

[0063] The specific implementation method of determining the estimated collapse degree of the suspected wrestler based on the wrestling reference surface, the foot pixel coordinate set, and the head pixel coordinate set in step S453 is as follows:

[0064] Step S4531: determining a first center coordinate based on the body pixel point coordinate set, and determining a second center coordinate based on the head pixel point coordinate set;

[0065] Step S4531: Evaluate the corresponding collapse degree based on the vertical arrangement order of the first center coordinate, the second center coordinate, and the wrestling reference surface and the height difference between them. The greater the difference between the vertical arrangement order and the preset arrangement order, the higher the collapse degree estimate. The greater the difference between the height difference between the first center coordinate, the second center coordinate, and the wrestling reference surface and the preset threshold range, the higher the collapse degree estimate. The collapse degree estimate is used to characterize the degree of aggregation between joints. The greater the collapse degree estimate, the greater the probability of the target person falling.

[0066] In a sitting position, the head, feet, and reference plane where the feet are located have a strict height order and a distance threshold range between the two points. The more chaotic the height order and the greater the difference between the height difference between any two points and the corresponding distance threshold, the greater the probability of falling.

[0067] Example 2:

[0068] This embodiment provides an artificial intelligence-based emergency call device, the device comprising:

[0069] a first calculation module, configured to respond to position information fed back by a plurality of human parameter monitoring devices and mark a human parameter monitoring device located in a dangerous radiation zone as a risky device based on the position information;

[0070] A second calculation module is used to retrieve the wearer's health data corresponding to the risky device and screen out high-risk devices from multiple risky devices based on the risk level corresponding to the dangerous radiation area;

[0071] A third calculation module is configured to construct a forward range based on the operating speed and vector of the high-risk device, and when the forward range intersects the high-risk area, determine a focus probability based on the historical motion trajectory of the high-risk device. When the focus probability is greater than a first threshold, a first focus instruction is generated based on the position code of the high-risk area, so that the corresponding surveillance camera focuses on the high-risk area according to a preset focal length and feeds back surveillance video.

[0072] The early warning module is used to determine the body posture of the wearer of high-risk equipment when passing through high-risk areas based on the surveillance video, and to call the corresponding service personnel urgently if the body posture is abnormal.

[0073] It should be noted that, regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated on here.

[0074] Example 3:

[0075] Corresponding to the above method embodiment, the embodiment of the present disclosure also provides an emergency call device based on artificial intelligence. The emergency call device based on artificial intelligence described below and the emergency call method based on artificial intelligence described above can refer to each other.

[0076] Figure 2 FIG is a block diagram of an emergency call device 800 based on artificial intelligence according to an exemplary embodiment. Figure 2 As shown, the electronic device 800 may include: a processor 801 , a memory 802 , and may further include one or more of a multimedia component 803 , an I / O interface 804 , and a communication component 805 .

[0077] The processor 801 is used to control the overall operation of the electronic device 800 to complete all or part of the steps in the above-mentioned artificial intelligence-based emergency call method. The memory 802 is used to store various types of data to support the operation of the electronic device 800. Such data may include, for example, instructions for any application or method operating on the electronic device 800, as well as application-related data, such as contact information, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 802 or sent through the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules. The above-mentioned other interface modules may be a keyboard, a mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the electronic device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G or 4G, or a combination of one or more thereof, so the corresponding communication component 805 may include: a Wi-Fi module, a Bluetooth module, an NFC module.

[0078] In an exemplary embodiment, the electronic device 800 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned artificial intelligence-based emergency call method.

[0079] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided. When executed by a processor, the program instructions implement the steps of the aforementioned artificial intelligence-based emergency call method. For example, the computer-readable storage medium may be the aforementioned memory 802 including the program instructions. The program instructions may be executed by the processor 801 of the electronic device 800 to implement the aforementioned artificial intelligence-based emergency call method.

[0080] Embodiment 4:

[0081] Corresponding to the above method embodiment, the embodiment of the present disclosure further provides a readable storage medium. The readable storage medium described below and the artificial intelligence-based emergency call method described above can refer to each other.

[0082] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the artificial intelligence-based emergency call method of the above-mentioned method embodiment.

[0083] The readable storage medium may specifically be any readable storage medium that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0084] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. An artificial intelligence-based emergency call method, applicable to a human parameter monitoring device, wherein the human parameter monitoring device is used to monitor the wearer's physiological data and location information, characterized in that: The method comprises: In response to position information fed back by a plurality of human parameter monitoring devices, and based on the position information, marking a human parameter monitoring device located in a dangerous radiation zone as a risky device; Retrieving health data of the wearer corresponding to the risky device, and screening out high-risk devices from multiple risky devices based on the risk level corresponding to the dangerous radiation area; Based on the running speed and vector of the high-risk device, a forward range is constructed. When the forward range intersects with the high-risk area, a focus probability is determined based on the historical motion trajectory of the high-risk device. When the focus probability is greater than a first threshold, a first focus instruction is generated based on the position code of the high-risk area, so that the corresponding surveillance camera focuses on the high-risk area according to a preset focal length and feeds back surveillance video. Determine the body posture of the wearer of high-risk equipment when passing through high-risk areas based on surveillance video, and call the corresponding service personnel urgently if the body posture is abnormal; Secondly, the determination of focus probability based on the historical motion trajectory of the high-risk device includes: Dividing the historical running trajectory into a first running trajectory and a second running trajectory based on a time limit, wherein the first running trajectory is the running trajectory within the dangerous radiation zone today, and the second running trajectory is the activity trajectory within the past 1-2 weeks; Marking a plurality of reference points on the first running trajectory based on a preset time interval, generating a forward range corresponding to each reference point, and calculating a proportion of reference points that intersect the forward range with the high-risk area; Counting the number of times the second running trajectory crosses the high-risk area, and calculating the probability of the current high-risk device crossing the high-risk area through a weighted algorithm based on the proportion and the number of times, which is the focus probability; Secondly, the method of determining the body posture of the wearer of the high-risk device when passing through the high-risk area based on the surveillance video includes: Construct a key reference plane based on the building characteristics in the high-risk area, and divide the key reference plane into location zones, thereby obtaining the location parameter range corresponding to each location zone; The PAF algorithm is used to construct the joint points of multiple characters in the surveillance video, and the shoulder and foot joint points of each character are extracted; Based on the inclusion relationship between the foot joint point of each character and the key reference plane, characters that are not within the key reference plane are eliminated, thereby obtaining a plurality of first shoulder joint points and first foot joint points corresponding to the characters within the key reference plane; Determining the location zone where the wearer of the high-risk device is located based on the location information fed back by the high-risk device and the location parameter range corresponding to each location zone, and then finding the second shoulder joint point and the second foot joint point corresponding to the wearer of the high-risk device among multiple people located in the key reference plane; The body posture of the wearer of the high-risk device when passing through the high-risk area is determined based on the position changes of the second shoulder joint point and the second foot joint point when passing through the high-risk area.

2. The artificial intelligence-based emergency call method according to claim 1, characterized in that: The determining of the body posture of the wearer of the high-risk device when passing through the high-risk area based on the position changes of the second shoulder joint point and the second foot joint point when passing through the high-risk area includes: detecting the integrity of the joint points corresponding to the wearer of the high-risk device in real time, and identifying the second shoulder joint point and the second foot joint point if they are intact, and monitoring the height difference between the two in real time, and calculating the time it takes for the height difference to collapse to the first threshold if the height difference is less than a first threshold, recording it as a first time, and identifying the head pixel point coordinate set and the foot pixel point coordinate set of the suspected wrestler using a shape-based image feature extraction algorithm after the height difference is less than the first threshold; Determine the key reference plane where the feet of the suspected wrestler are located based on the set of foot pixel coordinates, and record it as the wrestling reference plane; The estimated collapse degree of the suspected wrestler is determined based on the wrestling reference surface, the foot pixel coordinate set and the head pixel coordinate set, and then the suspected wrestler is determined to be in a falling state or a sitting state based on the estimated collapse degree and the first time length weighting.

3. The artificial intelligence-based emergency call method according to claim 2, characterized in that: The step of determining an estimated collapse degree of a suspected wrestling figure based on the wrestling reference surface, the foot pixel coordinate set, and the head pixel coordinate set includes: Determine the first center coordinates based on the set of pixel coordinates of the lower body, and determine the second center coordinates based on the set of pixel coordinates of the head; The corresponding collapse degree estimation is evaluated based on the longitudinal arrangement order of the first center coordinate, the second center coordinate and the wrestling reference surface and the height difference between them. The greater the difference between the longitudinal arrangement order and the preset arrangement order, the higher the collapse degree estimation. The greater the difference between the height difference between the first center coordinate, the second center coordinate and the wrestling reference surface and the preset threshold range, the higher the collapse degree estimation. The collapse degree estimation is used to characterize the degree of aggregation between joint points. The greater the collapse degree estimation, the greater the probability of the target person falling.

4. An emergency call device suitable for the artificial intelligence-based emergency call method according to claim 1, characterized in that: include: a first calculation module, configured to respond to position information fed back by a plurality of human parameter monitoring devices and mark a human parameter monitoring device located in a dangerous radiation zone as a risky device based on the position information; A second calculation module is used to retrieve the wearer's health data corresponding to the risky device and screen out high-risk devices from multiple risky devices based on the risk level corresponding to the dangerous radiation area; A third calculation module is configured to construct a forward range based on the operating speed and vector of the high-risk device, and when the forward range intersects the high-risk area, determine a focus probability based on the historical motion trajectory of the high-risk device. When the focus probability is greater than a first threshold, a first focus instruction is generated based on the position code of the high-risk area, so that the corresponding surveillance camera focuses on the high-risk area according to a preset focal length and feeds back surveillance video. The early warning module is used to determine the body posture of the wearer of high-risk equipment when passing through high-risk areas based on the surveillance video, and to call the corresponding service personnel urgently if the body posture is abnormal.

5. An electronic device, characterized in that: include: Memory for storing computer programs; A processor, configured to implement the method steps described in any one of claims 1 to 3 when executing a program stored in a memory.

6. A medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method steps described in any one of claims 1 to 3 are implemented.

Images