Security detection method and device, equipment and storage medium

Abstract

The embodiment of the invention provides a security detection method and device, equipment and a storage medium, and the method comprises the steps: obtaining the target sensitivity configuration information of each detection subarea in a detection region, and transmitting the target sensitivity configuration information to a radar system, the target sensitivity configuration information is sent to the radar system, so that the radar system detects each detection partition based on the target sensitivity configuration information; receiving regional point cloud data about each detection partition sent by the radar system, and determining a target partition where a target object is located in the plurality of detection partitions based on the regional point cloud data; and obtaining a security detection model corresponding to the target partition, inputting the regional point cloud data into the security detection model for detection, and determining a security detection result of the target object. According to the embodiment of the invention, the accuracy of the security detection result can be improved.

Description

Technical Field

[0001] The present application relates to, but is not limited to, the field of security detection technology, and in particular to a security detection method, device, equipment and storage medium. Background Art

[0002] At present, radar systems are widely used to perform security detection on target objects within the detection area. However, due to the differences in activity scenes corresponding to different locations in the detection area, and the traditional security detection methods use the same detection method for each location, this leads to poor detection effects in certain specific areas, reducing the accuracy of security detection results. Summary of the invention

[0003] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0004] The embodiments of the present application provide a security detection method, apparatus, device and storage medium, which can improve the accuracy of security detection results.

[0005] To achieve the above-mentioned purpose, the first aspect of an embodiment of the present application proposes a safety detection method, including: obtaining target sensitivity configuration information of each detection partition in the detection area, and sending the target sensitivity configuration information to a radar system so that the radar system can detect each detection partition based on the target sensitivity configuration information; receiving regional point cloud data about each detection partition sent by the radar system, and determining the target partition where the target object is located in multiple detection partitions based on the regional point cloud data; obtaining a safety detection model corresponding to the target partition, inputting the regional point cloud data into the safety detection model for detection, and determining the safety detection result of the target object.

[0006] In some embodiments, the security detection result includes a first detection result and a second detection result, the first detection result is used to indicate that the target object is in an unsafe state, and the second detection result is used to indicate that the target object is in a safe state. The security detection method also includes: when the security detection result is the first detection result, generating an alarm message and sending the alarm message to the terminal; or, when the security detection result is the second detection result, receiving the regional point cloud data again.

[0007] In some embodiments, the categories of the detection partitions include a shower area and a wash area, the shower area is used for the target object to take a shower, and the wash area is used for the target object to wash. When the category of the target partition is the shower area or the wash area, the inputting of the regional point cloud data into the safety detection model for detection to determine the safety detection result of the target object includes: inputting the regional point cloud data into the safety detection model, determining the object point cloud data of the target object in the regional point cloud data based on the safety detection model, and determining the target height and target contour of the target object based on the object point cloud data, wherein the target height is the height of the top of the target object from the ground; determining the target posture of the target object based on the target height and the target contour; when the duration of the target posture in an abnormal posture is greater than a preset first time threshold, determining the safety detection result of the target object to be the first detection result, wherein the abnormal posture is that the target height is less than a preset height threshold, and the similarity between the target contour and the preset contour is greater than a preset similarity threshold.

[0008] In some embodiments, the category of the detection partition includes a toilet area, and the toilet area is used for the target object to use a toilet. When the category of the target partition is the toilet area, the regional point cloud data is input into the safety detection model for prediction to determine the safety detection result of the target object, including: inputting the regional point cloud data into the safety detection model, determining the object point cloud data of the target object in the regional point cloud map based on the safety detection model, and determining the torso displacement and breathing frequency of the target object based on the object point cloud data; when the torso displacement is greater than a preset displacement threshold or the breathing frequency is less than a preset frequency threshold, determining that the safety detection result of the target object is the first detection result.

[0009] In some embodiments, the category of the detection partition includes a bathtub area, and the bathtub area is used for the target object to use a bathtub. When the category of the target partition is the bathtub area, the regional point cloud data is input into the safety detection model for prediction to determine the safety detection result of the target object, including: inputting the regional point cloud data into the safety detection model, determining the bathtub point cloud data and the object point cloud data of the target object in the regional point cloud data based on the safety detection model, and determining the bathing time of the target object using the bathtub based on the bathtub point cloud data and the object point cloud data; when the bathing time is greater than a preset third time threshold, determining the safety detection result as the first detection result.

[0010] In some embodiments, the obtaining of target sensitivity configuration information of each detection partition in the detection area includes: in response to a region division request regarding the detection area sent by a terminal, sending a region configuration request to a radar system so that the radar system divides the detection area into multiple detection partitions, wherein the region division request is used to indicate that the category of the detection partition belongs to one of multiple region types; obtaining candidate sensitivity configuration information corresponding to each of the region types, and for each of the detection partitions, determining the target sensitivity configuration information of the detection partition in each of the candidate sensitivity configuration information based on the category of the detection partition.

[0011] In some embodiments, the obtaining of candidate sensitivity configuration information corresponding to each of the area types includes: obtaining a reference activity frequency and a reference water consumption corresponding to each of the area types, wherein the reference activity frequency is the activity frequency of the target object in the detection area corresponding to the area type, and the reference water consumption is the water consumption in the detection area corresponding to the area type; for each of the area types, the reference activity frequency and the reference water consumption are input into a sensitivity prediction model for prediction to obtain the candidate sensitivity configuration information corresponding to the area type.

[0012] To achieve the above-mentioned purpose, the second aspect of an embodiment of the present application proposes a safety detection device, including: an acquisition module, used to acquire target sensitivity configuration information of each detection partition in the detection area, and send the target sensitivity configuration information to a radar system, so that the radar system can detect each detection partition based on the target sensitivity configuration information; a target determination module, used to receive regional point cloud data about each detection partition sent by the radar system, and determine the target partition where the target object is located in multiple detection partitions based on the regional point cloud data; a safety detection module, used to acquire a safety detection model corresponding to the target partition, input the regional point cloud data into the safety detection model for detection, and determine the safety detection result of the target object.

[0013] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, the memory stores a computer program, and the processor implements the security detection method described in the first aspect when executing the computer program.

[0014] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a storage medium, which is a computer-readable storage medium, and the storage medium stores a computer program. When the computer program is executed by a processor, it implements the security detection method described in the first aspect above.

[0015] The embodiments of the present application include at least the following beneficial effects: by acquiring the target sensitivity configuration information of each detection partition in the detection area, and then sending the target sensitivity configuration information to the radar system, so that the radar system can detect each detection partition based on the target sensitivity configuration information; therefore, during the detection process, the radar system can adapt the activity scene corresponding to the detection partition based on the specific target sensitivity configuration information, realize adaptive detection of each detection partition, and improve the data quality of the regional point cloud data generated by the radar system; then, the regional point cloud data about each detection partition sent by the radar system is received, and the target partition where the target object is located is determined in multiple detection partitions based on the regional point cloud data; then, the safety detection model corresponding to the target partition is obtained, and the regional point cloud data is input into the safety detection model for detection to determine the safety detection result of the target object, which is equivalent to when the target object is active in the activity scene corresponding to the target partition, the safety detection model adapted to the target partition can be used for safety detection; by deploying the corresponding safety detection model for the detection partition, targeted detection of the activity scene corresponding to the detection partition can be realized, and the accuracy of the safety detection result can be improved.

[0016] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or understood by practicing the present application. The purpose and other advantages of the present application can be realized and obtained by the structures specifically pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide further understanding of the technical solution of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present application and do not constitute a limitation on the technical solution of the present application.

[0018] Figure 1 An optional flow chart of a safety detection method provided in an embodiment of the present application; Figure 2 An optional flowchart for determining a security detection result provided in an embodiment of the present application; Figure 3 An optional flow chart of a safety detection model corresponding to a shower area or a washing area provided in an embodiment of the present application; Figure 4 An optional specific flow chart of a safety detection model corresponding to the toilet area provided in an embodiment of the present application; Figure 5 An optional flow chart of a safety detection model corresponding to a bathtub area provided in an embodiment of the present application; Figure 6 A schematic diagram of an optional process for obtaining target sensitivity configuration information provided in an embodiment of the present application; Figure 7 A schematic diagram of an optional process for obtaining candidate sensitivity configuration information provided in an embodiment of the present application; Figure 8 An optional initialization schematic diagram of the detection area provided in the embodiment of the present application; Fig. 9 A schematic diagram of an optional division of the detection area provided in an embodiment of the present application; Fig.10 An optional structural diagram of a safety detection device provided in an embodiment of the present application; Fig.11 A schematic diagram of an optional hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0020] It should be noted that in each specific embodiment of the present application, when it comes to the need to perform relevant processing based on data related to the characteristics of the target object such as target object attribute information or attribute information set, the permission or consent of the target object will be obtained first, and the collection, use and processing of these data will comply with relevant laws, regulations and standards. Among them, the target object can be a user. In addition, when the embodiment of the present application needs to obtain the target object attribute information, the target object's separate permission or separate consent will be obtained by means of a pop-up window or jumping to a confirmation page, etc., and after the separate permission or separate consent of the target object is clearly obtained, the necessary target object-related data for enabling the normal operation of the embodiment of the present application will be obtained.

[0021] In the description of this application, “several” means one or more, “more” means more than two, “greater than”, “less than”, “exceed”, etc. are understood to exclude the number, and “above”, “below”, “within”, etc. are understood to include the number.

[0022] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims or the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0023] To facilitate understanding of the technical solutions provided in the embodiments of the present application, some key terms used in the embodiments of the present application are explained here: Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines so that machines have the functions of perception, reasoning and decision-making.

[0024] Artificial intelligence technology is a comprehensive discipline that covers a wide range of fields, including both hardware-level and software-level technologies. Basic artificial intelligence technologies generally include sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operating / interactive systems, mechatronics and other technologies. Artificial intelligence software technologies mainly include computer vision technology, speech processing technology, natural language processing technology, as well as machine learning / deep learning, autonomous driving, smart transportation and other major directions.

[0025] Machine Learning (ML) is a multi-disciplinary subject that involves probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and other disciplines. It specializes in studying how computers simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications are spread across all areas of artificial intelligence. Machine learning and deep learning usually include artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and teaching learning.

[0026] At present, radar systems are widely used to perform security detection on target objects within the detection area. However, due to the differences in activity scenes corresponding to different locations in the detection area, and the traditional security detection methods use the same detection method for each location, this leads to poor detection effects in certain specific areas, reducing the accuracy of security detection results.

[0027] In response to the problem of low accuracy of safety detection results, the present application provides a safety detection method, apparatus, equipment and storage medium, the method comprising: obtaining target sensitivity configuration information of each detection partition in the detection area, and sending the target sensitivity configuration information to a radar system so that the radar system can detect each of the detection partitions based on the target sensitivity configuration information; receiving regional point cloud data about each of the detection partitions sent by the radar system, and determining the target partition where the target object is located in multiple detection partitions based on the regional point cloud data; obtaining a safety detection model corresponding to the target partition, inputting the regional point cloud data into the safety detection model for detection, and determining the safety detection result of the target object.

[0028] According to the scheme provided in the embodiment of the present application, by acquiring the target sensitivity configuration information of each detection partition in the detection area, and then sending the target sensitivity configuration information to the radar system, so that the radar system can detect each detection partition based on the target sensitivity configuration information, therefore, during the detection process, the radar system can adapt the activity scene corresponding to the detection partition based on the specific target sensitivity configuration information, realize adaptive detection of each detection partition, and improve the data quality of the regional point cloud data generated by the radar system, and then receive the regional point cloud data about each detection partition sent by the radar system, determine the target partition where the target object is located in multiple detection partitions based on the regional point cloud data, and then obtain the security detection model corresponding to the target partition, input the regional point cloud data into the security detection model for detection, and determine the security detection result of the target object, which is equivalent to when the target object is active in the activity scene corresponding to the target partition, can use the security detection model adapted to the target partition for security detection, by deploying the corresponding security detection model for the detection partition, to achieve targeted detection of the activity scene corresponding to the detection partition, can improve the accuracy of the security detection result.

[0029] The security detection method, device, equipment and storage medium provided in the embodiments of the present application are specifically described through the following embodiments. First, the security detection method in the embodiments of the present application is described.

[0030] The security detection method provided in the embodiment of the present application relates to the field of computer technology. The security detection method provided in the embodiment of the present application can be applied to a terminal, can be applied to a server side, or can be software running in a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or can be configured as a server cluster or a distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the security detection method, etc., but is not limited to the above forms.

[0031] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0032] The embodiments of the present application are further described below in conjunction with the accompanying drawings.

[0033] like Figure 1 As shown, Figure 1 An optional flow chart of a security detection method provided in an embodiment of the present application, the security detection method can be executed by a server, or can also be executed by a terminal, or can also be executed by a server in cooperation with a terminal, the security detection method includes but is not limited to the following steps S110 to S130: Step S110, obtaining target sensitivity configuration information of each detection partition in the detection area, and sending the target sensitivity configuration information to the radar system, so that the radar system detects each detection partition based on the target sensitivity configuration information.

[0034] Step S120: receiving regional point cloud data about each detection partition sent by the radar system, and determining a target partition where the target object is located in the multiple detection partitions based on the regional point cloud data.

[0035] Step S130, obtaining a safety detection model corresponding to the target partition, inputting the regional point cloud data into the safety detection model for detection, and determining the safety detection result of the target object.

[0036] Among them, the activity scenes corresponding to any two detection partitions are different. The activity scene refers to the specific environment in which the target object conducts activities. For example, the detection partition may include a shower area, a toilet area, a bathtub area, a washing area, etc.; different activity scenes usually have different environmental interferences and activity contents, and the target sensitivity configuration information can adapt to the environmental interference and activity content corresponding to the detection partitions.

[0037] Among them, the regional point cloud data is composed of multiple data points. The parameters of each data point include three-dimensional coordinates and energy values. Different activities of the target object correspond to different energy values ​​of the data points. The more frequent or intense the activity, the greater the energy value of the corresponding data point, and the easier it is to be detected by the radar system; at the same time, the more environmental interference corresponds to a part of the data points, the greater the environmental interference, the more corresponding data points and the greater the energy, and the more difficult it is to be distinguished by the radar system.

[0038] Based on this, by obtaining the target sensitivity configuration information of each detection partition in the detection area, and then sending the target sensitivity configuration information to the radar system, so that the radar system can detect each detection partition based on the target sensitivity configuration information, therefore, during the detection process, the radar system can adapt the activity scene corresponding to the detection partition based on the specific target sensitivity configuration information, realize adaptive detection of each detection partition, and improve the data quality of the regional point cloud data generated by the radar system, and then receive the regional point cloud data about each detection partition sent by the radar system, determine the target partition where the target object is located in multiple detection partitions based on the regional point cloud data, and then obtain the security detection model corresponding to the target partition, input the regional point cloud data into the security detection model for detection, and determine the security detection result of the target object, which is equivalent to when the target object is active in the activity scene corresponding to the target partition, it can use the security detection model adapted to the target partition for security detection, and by deploying the corresponding security detection model for the detection partition, it is possible to achieve targeted detection of the activity scene corresponding to the detection partition, which can improve the accuracy of the security detection result.

[0039] In one possible implementation, in addition to obtaining the target sensitivity configuration information of each detection partition in the detection area, the target scanning frequency and target detection range of each detection partition can also be obtained. The target scanning frequency and target detection range can be used as auxiliary detection data to further improve the detection accuracy of each detection partition in the detection area.

[0040] In addition, refer to Figure 2In one embodiment, the safety detection result includes a first detection result and a second detection result, the first detection result is used to indicate that the target object is in an unsafe state, and the second detection result is used to indicate that the target object is in a safe state. The safety detection method also includes but is not limited to the following steps: Step S210: When the security detection result is the first detection result, an alarm message is generated and sent to the terminal.

[0041] Step S220: When the safety detection result is the second detection result, the regional point cloud data is received again.

[0042] Among them, the alarm information includes first alarm information, second alarm information and third alarm information. The first alarm information is used to indicate that the target object has fallen, the second alarm information is used to indicate that the target object has fainted or is unable to move, and the third alarm information is used to indicate that the target object has been in a certain detection area for more than a predetermined time, and the target object may have an accident.

[0043] Among them, after receiving any of the above-mentioned multiple alarm information, the terminal will perform corresponding alarm behavior.

[0044] Specifically, after the terminal receives the first alarm message, the terminal will call the preset emergency contact; after the terminal receives the second alarm message, it will emit a first ringtone to wake up the target object. If the security detection result is still the first detection result within the preset wake-up time, the preset emergency contact will be called; after the terminal receives the third alarm message, it will emit a second ringtone to remind the target object that it has stayed in a certain area for too long. If the security detection result is still the first detection result within the preset target object response time, the preset emergency contact will be called.

[0045] Based on this, by inputting the regional point cloud data into the safety detection model for detection, the safety detection result is determined to be the first detection result, indicating that the target object is in an unsafe state and requires intervention, so an alarm message is generated and sent to the terminal, so that the terminal can make corresponding alarm actions according to the alarm information, preventing the target object from being in a more serious unsafe state, reducing the safety risk of the target object, and improving the safety of the target object.

[0046] In addition, refer to Figure 3 In one embodiment, the categories of the detection partitions include a shower area and a wash area. The shower area is used for the target object to take a shower, and the wash area is used for the target object to wash. When the category of the target partition is a shower area or a wash area, the regional point cloud data is input into the safety detection model for detection, and the safety detection result of the target object is determined, including but not limited to the following steps: Step S310, inputting the regional point cloud data into the safety detection model, determining the object point cloud data of the target object in the regional point cloud data based on the safety detection model, and determining the target height and target contour of the target object based on the object point cloud data.

[0047] Step S320: determining the target posture of the target object based on the target height and the target contour.

[0048] Step S330: when the duration of the target posture being in the abnormal posture is greater than a preset first time threshold, determining the safety detection result of the target object as a first detection result.

[0049] The target height is the height of the top of the target object from the ground, the abnormal posture is that the target height is less than a preset height threshold, and the similarity between the target contour and the preset contour is greater than a preset similarity threshold.

[0050] Among them, the regional point cloud data includes object point cloud data and environmental point cloud data of the target object. The object point cloud data is composed of multiple data points, so the top of the target object can be determined according to the highest data point in the object point cloud data, and then the height of the top of the target object from the ground, that is, the target height, can be further determined; the outline of the target object, that is, the target outline, can also be determined according to the outermost multiple data points of the object point cloud data; and when it is determined that the target object is in an abnormal posture, a timer can be started. When the accumulated time of the timer exceeds the first time threshold, it is determined that the target object is in an unsafe state, and the safety detection result is determined as the first safety detection result.

[0051] It should be noted that it is necessary to obtain the distance between the highest point and the lowest point of the regional point cloud data as the height of the detection area, and then multiply the height of the detection area by the preset ratio to determine the preset height threshold. For example, the abnormal posture may be a fall, and the preset ratio may be one-third, which can meet most falling postures, such as falling on flat ground or half-kneeling.

[0052] Specifically, when the target height of the target object is less than the preset height threshold, it can be indicated that the target object is located near the ground as a whole, and the preset outline can be the outline of a human body, indicating that the target object is a human body, and the entire human body is not standing or squatting normally on the ground in the shower area or washing area, so it can be judged that the posture of the human body is abnormal. In addition, the first time threshold is used to further judge whether the human body is in an abnormal posture or cannot return to a normal posture after being in an abnormal posture, and finally judge that the human body is in an unsafe state, and determine the safety detection result as the first detection result. For example, after a human body falls, the posture of the human body is judged to be an abnormal posture, and the human body cannot stand up, so the human body will continue to be in an abnormal posture. After the duration exceeds the first time threshold, it will be judged that the human body is in an unsafe state, and finally the safety detection result will be determined as the first detection result.

[0053] It should be noted that the shower area is more likely to cause the target object to fall. Therefore, the detection logic of the safety detection model corresponding to the shower area is stricter than that of the safety detection model corresponding to the wash area. Specifically, the first time threshold corresponding to the shower area is smaller than the first time threshold of the wash area. When the target object is in an abnormal posture in the shower area, the corresponding safety detection result is more easily determined as the first detection result.

[0054] It should be noted that the water flow and water vapor in the shower area are more than those in the wash area, so the interference in the shower area is more than that in the wash area. The more interference, the lower the corresponding sensitivity. At the same time, the activity frequency of the target object in the shower area is higher than that in the wash area. Correspondingly, the regional point cloud data of the shower area is easier to detect than the regional point cloud data of the wash area. Therefore, the target sensitivity configuration information corresponding to the shower area is lower than the target sensitivity configuration information corresponding to the wash area.

[0055] Based on this, by determining that the category of the target partition where the target object is located is a shower area or a wash area, and then inputting the regional point cloud data into the safety detection model, the object point cloud data of the target object is determined in the regional point cloud data based on the safety detection model, and the target height and target contour of the target object are determined based on the object point cloud data. Since the object point cloud data is composed of multiple data points, each data point has a corresponding accurate position, which can improve the accuracy of the target height and target contour. Based on the target height and target contour, the target posture of the target object is determined, and the accuracy of the target posture is improved. When the target height of the target object is less than the preset height threshold and the target wheel is When the similarity between the contour and the preset contour is greater than the preset similarity threshold, it is determined that the target posture is in an abnormal posture. The above detection method can detect a variety of abnormal postures, generalizes the application scenarios of the detection server, and when the duration of the target posture in the abnormal posture is greater than the preset first time threshold, the security detection result of the target object is determined to be the first detection result. When the target object is in an abnormal posture, it is usually unable to change quickly, but can change quickly when it is in a normal posture. This can prevent some normal postures that are very similar to abnormal postures from being judged as abnormal postures, thereby improving the robustness of the detection server, reducing the misjudgment rate, and improving the accuracy of the security detection results.

[0056] In addition, refer to Figure 4 In one embodiment, the category of the detection partition includes a toilet area, and the toilet area is used for the target object to use the toilet. When the category of the target partition is the toilet area, the regional point cloud data is input into the safety detection model for prediction to determine the safety detection result of the target object, including but not limited to the following steps: Step S410, inputting the regional point cloud data into the safety detection model, determining the object point cloud data of the target object in the regional point cloud map based on the safety detection model, and determining the torso displacement and breathing frequency of the target object based on the object point cloud data.

[0057] Step S420: When the trunk displacement is greater than a preset displacement threshold or the breathing frequency is less than a preset frequency threshold, the safety detection result of the target object is determined to be the first detection result.

[0058] The trunk displacement of the target object is used to indicate the movement of the trunk of the target object. Usually, the target object in the toilet area will only move the trunk slightly.

[0059] It should be noted that the object point cloud data of the target object includes torso point cloud data. The torso point cloud data corresponding to the two previous and subsequent time frames are obtained, and the displacement is extracted based on the change in the center of gravity position of the two torso point cloud data, which is determined as the torso displacement.

[0060] It should be noted that the target object's breathing frequency can be calculated based on the torso point cloud data. When the target object breathes, the torso will rise and fall. The torso point cloud data corresponding to multiple time frames are obtained, and the torso contour corresponding to the torso point cloud data is extracted. The change frequency of the torso is determined based on the multiple torso contours, and the breathing frequency of the target object can be determined.

[0061] Among them, if the trunk displacement is greater than the preset displacement threshold, it means that the target object may have slipped off the toilet. For example, the surface of the toilet is relatively smooth and the target object does not sit firmly on the toilet. In this case, the target object is likely to slip off the toilet, causing danger; if the breathing frequency is less than the preset frequency threshold, it means that the target object has almost stopped breathing. For example, the target object's respiratory tract is originally blocked by foreign objects, and breathing is not smooth. When using the toilet, it is necessary to hold the breath, which may cause respiratory arrest. The target object is in a more urgent and dangerous situation. The safety detection result of the target object is set as the first detection result, and an alarm can be immediately sounded to call for help.

[0062] It should be noted that there is less water interference in the toilet area. At the same time, the target object has a lower activity frequency and a smaller activity amplitude, so the target sensitivity configuration information corresponding to the toilet area is larger, so that the radar system can improve the detection accuracy and capture more difficult parameters, such as the target object's breathing rate.

[0063] Based on this, by determining that the category of the target partition where the target object is located is the toilet area, and then inputting the regional point cloud data into the safety detection model, the object point cloud data of the target object is determined in the regional point cloud data based on the safety detection model, and the torso displacement and breathing frequency of the target object are determined based on the object point cloud data. Since the object point cloud data is composed of multiple data points, each data point has a corresponding accurate position, which can improve the accuracy of the point cloud data corresponding to the torso, and then improve the accuracy of the torso displacement and breathing frequency. The torso displacement greater than the preset displacement threshold can be judged as an abnormal situation. After the abnormal situation occurs, the safety detection result of the target object is determined to be the first detection result, which can detect the danger of the target object slipping into the toilet. At the same time, the breathing frequency less than the preset frequency threshold can be judged as an abnormal situation. After the abnormal situation occurs, the safety detection result of the target object is determined to be the first detection result, which can detect the danger of respiratory arrest of the target object. A variety of methods for judging abnormal situations are set for the toilet area, which improves the robustness of the detection server, reduces the misjudgment rate, and improves the accuracy of the safety detection results.

[0064] In addition, refer to Figure 5 In one embodiment, the category of the detection partition includes a bathtub area, and the bathtub area is used for the target object to use a bathtub. When the category of the target partition is the bathtub area, the regional point cloud data is input into the safety detection model for prediction to determine the safety detection result of the target object, including but not limited to the following steps: Step S510, inputting the regional point cloud data into the safety detection model, determining the bathtub point cloud data and the object point cloud data of the target object in the regional point cloud data based on the safety detection model, and determining the bathing time of the target object using the bathtub based on the bathtub point cloud data and the object point cloud data.

[0065] Step S520: when the bathing time is greater than a preset third time threshold, the safety detection result is determined as the first detection result.

[0066] Among them, the regional point cloud data also includes bathtub point cloud data and object point cloud data. Since there is too much water interference in the bathtub, the target sensitivity configuration information corresponding to the bathtub area is reduced as much as possible so that the radar system can detect the approximate bathtub point cloud data and object point cloud data.

[0067] Among them, when the object point cloud data and the bathtub point cloud data are adhered, it means that the target object is using the bathtub, and when the object point cloud data and the bathtub point cloud data are not adhered, it means that the target object is not using the bathtub. The bathing time of the target object using the bathtub can be determined by determining the state change process of the object point cloud data and the bathtub point cloud data and the corresponding time. For example, the state change process of the object point cloud data and the bathtub point cloud data is: first non-adhesion, then adhesion, and finally change back to non-adhesion. The time point when the state changes from non-adhesion to adhesion is the starting time point, and the time point when the state changes from adhesion to non-adhesion is the final time point. The starting time point and the final time point are obtained, and the bathing time of the target object using the bathtub is obtained by subtracting the final time point from the starting time point.

[0068] It should be noted that the safety detection model corresponding to the bathtub area will increase the recognition and filtering capabilities of water surface reflection signals, improve anti-interference capabilities, and make the detected area point cloud data more realistic.

[0069] Based on this, by determining that the category of the target partition where the target object is located is the bathtub area, and then inputting the regional point cloud data into the safety detection model, the bathtub point cloud data and the object point cloud data of the target object are determined in the regional point cloud data based on the safety detection model, and the bathing time of the target object using the bathtub is determined based on the bathtub point cloud data and the object point cloud data. Since the object point cloud data is composed of multiple data points, each data point has a corresponding accurate position, and the bathtub area has corresponding target sensitivity configuration information, it can resist interference to a large extent, improve the accuracy of the bathtub point cloud data and the object point cloud data, and thus improve the accuracy of the bathing time. Bathing time greater than the preset third time threshold can be judged as an abnormal situation. After the abnormal situation occurs, the safety detection result of the target object is determined to be the first detection result. The method of setting the third time threshold is given to perform safety detection on the bathtub area, and the bathtub area with greater interference is detected, which can reduce the misjudgment rate and improve the accuracy of the safety detection results.

[0070] In addition, refer to Figure 6 In one embodiment, obtaining target sensitivity configuration information of each detection zone in the detection area includes but is not limited to the following steps: Step S610, in response to the area division request regarding the detection area sent by the terminal, sending an area configuration request to the radar system, so that the radar system divides the detection area into a plurality of detection partitions.

[0071] Step S620, obtaining candidate sensitivity configuration information corresponding to each area type, and for each detection partition, determining target sensitivity configuration information of the detection partition in each candidate sensitivity configuration information based on the category of the detection partition.

[0072] The area division request is used to indicate that the category of the detection partition belongs to one of multiple area types.

[0073] The detection area is an area that can be scanned by the radar system, and the shape may be a square, a circle, or a triangle, which is not limited in this embodiment.

[0074] Based on this, by responding to the area division request regarding the detection area sent by the terminal, the area configuration request is sent to the radar system, so that the radar system divides the detection area into multiple detection partitions, and obtains the candidate sensitivity configuration information corresponding to each area type. For each detection partition, based on the category of the detection partition, the target sensitivity configuration information of the detection partition is determined in each candidate sensitivity configuration information, and each detection partition is finely divided into regions, and the target sensitivity configuration information corresponding to each detection partition is determined in a targeted manner, which is conducive to targeted processing of specific detection areas and improving the accuracy of safety detection results.

[0075] In one possible implementation, reference Figure 8 The terminal divides the detection area into multiple grids. The user selects some grids as a detection partition through the terminal and sets the corresponding category. After the user sets the detection partition, the terminal encodes the detection area division method into a specific format of area division request according to the user map setting protocol, and then sends it to the detection server.

[0076] Specifically, the content of the user map setting protocol is as follows: the area division request includes N bytes, the first byte to the sixth byte carry relevant information of the communication process, the seventh byte to the N-2 byte carry relevant information of the detection partition, the N-1 byte is the first verification code, and the N byte is the second verification code.

[0077] For example, assuming that N is 84, the first byte is the identification code with a fixed value of 0x55, the second byte is the identification code with a fixed value of 0xAA, the third byte is the length with a fixed value of 84, the fourth byte is a reserved byte with a default content of 0x00, the fifth byte is the source address with a value of 0xAB, the sixth byte is the command word with a value of 0x30, the seventh byte is the map data length, the eighth byte is the map data width, the ninth byte is the map data scale, the tenth byte is the map version number, and the maximum value of the tenth byte is 255; Then, the detection area can be divided into multiple grids, and the detection partition can contain one or more grids. From the eleventh byte to the eighty-second byte, the hexadecimal values ​​determined by the first 4 bits and the last 4 bits of each byte are used to characterize the category of the detection partition where the corresponding grid is located. For example, the hexadecimal value corresponding to the edge wall is 1, the hexadecimal value corresponding to the shower area is F, the hexadecimal value corresponding to the wash area is 4, the hexadecimal value corresponding to the bathtub area is B, the hexadecimal value corresponding to the entrance and exit is 2, and the hexadecimal value corresponding to the toilet area is The hexadecimal value corresponding to the corridor area is 0. Assuming the eleventh byte is 11, it means that the categories of the detection partitions where the first and second grids are located are both edge walls. Assuming the twenty-first byte is FF, it means that the categories of the detection partitions where the 21st and 22nd grids are located are both shower areas. Therefore, each byte can correspond to two grids. The eleventh byte to the eighty-second byte correspond to a total of 144 grids. The 144 grids can be marked with the corresponding 144 area identifiers in order. The 144 area identifiers are 0 to 143 respectively. Then, the eighty-third byte is CRC8, which indicates the CRC8 check from the first byte to the eighty-second byte, and the eighty-fourth byte is CheckSum, which indicates the checksum from the first byte to the eighty-third byte, and the checksum is inverted and added 1 as the result.

[0078] refer to Fig. 9 The terminal divides the detection area into shower area, washing area, toilet area and bathtub area. The washing area includes the washbasin, and the radar is located in the center of the detection area.

[0079] The detection area in the figure includes 144 grids. The grid in the upper left corner is the first grid, and the corresponding starting number is 0. The numbers increase by 1 from left to right and by 12 from top to bottom. The last grid is numbered 143. Each grid has a corresponding number. For example, the grids in the shower area are numbered 19, 20, 21, 31, 32, 33, 43, 44, 45, the grids in the wash area are numbered 60, 61, 72, 73, and the grids in the toilet area are numbered 126, 127, 138, 149, 150, 151, 152, 153, 154, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191 39. The bathtub area includes grid numbers 80, 81, 92, 93, 104, and 105. The entrance and exit include grid numbers 120, 121, 122, 132, 133, and 134. The edge walls include grid numbers 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 22, 23, 34, 35, 46, 47, 58, 59, 70, 71, 82, 83, 94, 95, 106, 107, 118, 119, 130, 131, 142, and 143.

[0080] Specifically, the encoding corresponding to the area division request is: 55 aa 54 00 ba 31 0c 0c 32 00 11 1111 11 11 11 00 00 00 0F FF 11 00 00 00 0F FF 11 00 00 00 0F FF 11 00 00 00 0000 11 44 ​​00 00 00 00 11 44 ​​00 00 00 BB 11 00 00 00 00 BB 11 00 00 00 00 BB 1100 00 00 00 00 11 22 20 00 55 00 11 33 13.

[0081] Among them, obtaining the candidate sensitivity configuration information corresponding to each area type is equivalent to obtaining the key-value pairs of multiple area types and corresponding candidate sensitivity configuration information. As long as the area type of these key-value pairs is compared with the category of a certain detection partition, the corresponding candidate sensitivity configuration information can be found, and this candidate sensitivity configuration information is used as the target sensitivity configuration information of this detection partition. For example, the candidate sensitivity configuration information corresponding to the shower area is 20, the candidate sensitivity configuration information corresponding to the wash area is 50, the candidate sensitivity configuration information corresponding to the toilet area is 90, and the candidate sensitivity configuration information corresponding to the bathtub area is 5.

[0082] In addition, refer to Figure 7 In one embodiment, obtaining candidate sensitivity configuration information corresponding to each area type includes but is not limited to the following steps: Step S710, obtaining reference activity frequencies and reference water consumption corresponding to each area type.

[0083] Step S720: For each area type, the reference activity frequency and the reference water consumption are input into the sensitivity prediction model for prediction to obtain candidate sensitivity configuration information corresponding to the area type.

[0084] The reference activity frequency is the activity frequency of the target object in the detection area corresponding to the area type, and the reference water consumption is the water consumption in the detection area corresponding to the area type.

[0085] It should be noted that the reference activity frequency and reference water consumption contain information about the difficulty of detection by the radar system.

[0086] Specifically, for a certain area type, when the reference activity frequency is high, it means that the target object is more active in the detection area corresponding to this area type, and the radar system can more easily detect the target object, and the corresponding candidate sensitivity configuration information tends to be a smaller value; when the reference water consumption is high, it means that there is more water flow or water vapor in the detection area corresponding to this area type, which causes greater interference to the radar system, and the corresponding candidate sensitivity configuration information tends to be a smaller value.

[0087] Based on this, by obtaining the reference activity frequency and reference water consumption corresponding to each area type, it is equivalent to obtaining the variable factors that affect the target sensitivity configuration information. Since the reference activity frequency corresponding to a certain area type represents the activity frequency of the target object in the detection area corresponding to this area type, it can affect the candidate sensitivity configuration information corresponding to this area type from one dimension, and the reference water consumption represents the interference degree of the detection area corresponding to this area type, which can affect the candidate sensitivity configuration information corresponding to this area type from another dimension. Therefore, the reference activity frequency and reference water consumption both have a guiding role in determining the candidate sensitivity configuration information. For each area type, the reference activity frequency and reference water consumption are input into the sensitivity prediction model for prediction, so that the sensitivity prediction model can simultaneously pay attention to the radar detection difficulty information contained in the reference activity frequency and the reference water consumption, thereby accurately generating dynamically changing candidate sensitivity configuration information in each iteration round, thereby improving the accuracy of the candidate sensitivity configuration information.

[0088] In a possible implementation, the sensitivity prediction model is a pre-trained regression model, and the training steps of the sensitivity prediction model are as follows: First, obtain reference activity frequency samples and reference water consumption samples as well as corresponding candidate sensitivity configuration information labels, extract the characteristics of the reference activity frequency samples as reference activity frequency characteristics, extract the characteristics of the reference water consumption samples as reference water consumption characteristics, input the reference activity frequency characteristics and the reference water consumption characteristics into the coefficient prediction model for regression processing, and obtain the reference candidate sensitivity configuration information; then, determine the model loss according to the reference candidate sensitivity configuration information and the candidate sensitivity configuration information labels, and train the coefficient prediction model based on the model loss, so that the coefficient prediction model can learn and understand the intrinsic relationship between the reference activity frequency, reference water consumption and candidate sensitivity configuration information, and then predict accurate candidate sensitivity configuration information by combining the reference activity frequency and reference water consumption in the inference stage.

[0089] In one possible implementation, the user or installation engineer may conduct an on-site inspection in the detection area, and after comprehensively considering the factors of each area, directly set the candidate sensitivity configuration information corresponding to each area type on the terminal, and then send it to the detection server.

[0090] In addition, refer to Fig.10 The present application also provides a safety detection device 1000, comprising: An acquisition module 1010 is used to acquire target sensitivity configuration information of each detection partition in the detection area, and send the target sensitivity configuration information to the radar system so that the radar system can detect each detection partition based on the target sensitivity configuration information; The target determination module 1020 is used to receive the regional point cloud data about each detection partition sent by the radar system, and determine the target partition where the target object is located in the multiple detection partitions based on the regional point cloud data; The safety detection module 1030 is used to obtain a safety detection model corresponding to the target partition, input the regional point cloud data into the safety detection model for detection, and determine the safety detection result of the target object.

[0091] It can be understood that the specific implementation of the safety detection device 1000 is basically the same as the specific embodiment of the above-mentioned safety detection method, and will not be repeated here.

[0092] In addition, refer to Fig.11 , Fig.11 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes: The processor 1101 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application; The memory 1102 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1102 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1102, and the processor 1101 is called to execute the security detection method of the embodiment of the present application.

[0093] Input / output interface 1103, used to implement information input and output; The communication interface 1104 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.); A bus 1105 that transmits information between various components of the device (e.g., the processor 1101, the memory 1102, the input / output interface 1103, and the communication interface 1104); The processor 1101 , the memory 1102 , the input / output interface 1103 and the communication interface 1104 are connected to each other in communication within the device via the bus 1105 .

[0094] An embodiment of the present application also provides a storage medium, which is a computer-readable storage medium used for computer-readable storage. The storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the above-mentioned security detection method.

[0095] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0096] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0097] It can be understood by those skilled in the art that Figures 1 to 11 The technical solutions shown in the figure do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than those shown in the figure, or a combination of certain steps, or different steps.

[0098] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0099] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0100] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0101] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0102] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0103] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0104] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0105] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0106] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A safety detection method, characterized in that: include: Acquire target sensitivity configuration information of each detection partition in the detection area, and send the target sensitivity configuration information to the radar system, so that the radar system detects each detection partition based on the target sensitivity configuration information; receiving regional point cloud data about each of the detection partitions sent by the radar system, and determining a target partition where a target object is located in the plurality of detection partitions based on the regional point cloud data; A safety detection model corresponding to the target partition is obtained, the regional point cloud data is input into the safety detection model for detection, and a safety detection result of the target object is determined.

2. The safety detection method according to claim 1, characterized in that: The safety detection result includes a first detection result and a second detection result, the first detection result is used to indicate that the target object is in an unsafe state, and the second detection result is used to indicate that the target object is in a safe state. The safety detection method further includes: When the security detection result is the first detection result, generating alarm information, and sending the alarm information to the terminal; Alternatively, when the safety detection result is the second detection result, the regional point cloud data is received again.

3. The safety detection method according to claim 2, characterized in that: The categories of the detection partitions include a shower area and a wash area, the shower area is used for the target object to take a shower, and the wash area is used for the target object to wash. When the category of the target partition is the shower area or the wash area, the inputting of the regional point cloud data into the safety detection model for detection to determine the safety detection result of the target object includes: Inputting the regional point cloud data into the safety detection model, determining the object point cloud data of the target object in the regional point cloud data based on the safety detection model, and determining the target height and target contour of the target object based on the object point cloud data, wherein the target height is the height of the top of the target object from the ground; Determining a target posture of the target object based on the target height and the target profile; When the duration of the target posture being in an abnormal posture is greater than a preset first time threshold, the safety detection result of the target object is determined to be the first detection result, wherein the abnormal posture is that the target height is less than a preset height threshold, and the similarity between the target contour and the preset contour is greater than a preset similarity threshold.

4. The safety detection method according to claim 2, characterized in that: The category of the detection partition includes a toilet area, and the toilet area is used for the target object to use the toilet. When the category of the target partition is the toilet area, the regional point cloud data is input into the safety detection model for prediction to determine the safety detection result of the target object, including: Inputting the regional point cloud data into the safety detection model, determining the object point cloud data of the target object in the regional point cloud map based on the safety detection model, and determining the trunk displacement and breathing frequency of the target object based on the object point cloud data; When the trunk displacement is greater than a preset displacement threshold or the breathing frequency is less than a preset frequency threshold, it is determined that the safety detection result of the target object is the first detection result.

5. The safety detection method according to claim 2, characterized in that: The category of the detection partition includes a bathtub area, and the bathtub area is used for the target object to use a bathtub. When the category of the target partition is the bathtub area, the regional point cloud data is input into the safety detection model for prediction to determine the safety detection result of the target object, including: Inputting the regional point cloud data into the safety detection model, determining bathtub point cloud data and object point cloud data of the target object in the regional point cloud data based on the safety detection model, and determining the bathing time of the target object using the bathtub based on the bathtub point cloud data and the object point cloud data; When the bathing time is greater than a preset third time threshold, the safety detection result is determined as the first detection result.

6. The safety detection method according to claim 1, characterized in that: The step of obtaining target sensitivity configuration information of each detection zone in the detection area includes: In response to a region division request regarding a detection region sent by a terminal, sending a region configuration request to a radar system so that the radar system divides the detection region into a plurality of detection partitions, wherein the region division request is used to indicate that the category of the detection partition belongs to one of a plurality of region types; The candidate sensitivity configuration information corresponding to each of the area types is obtained, and for each of the detection partitions, based on the category of the detection partition, the target sensitivity configuration information of the detection partition is determined in each of the candidate sensitivity configuration information.

7. The safety detection method according to claim 6, characterized in that: The obtaining of candidate sensitivity configuration information corresponding to each of the area types includes: Obtaining reference activity frequencies and reference water consumption corresponding to each of the area types, wherein the reference activity frequency is the activity frequency of the target object in the detection area corresponding to the area type, and the reference water consumption is the water consumption in the detection area corresponding to the area type; For each of the area types, the reference activity frequency and the reference water consumption are input into a sensitivity prediction model for prediction to obtain candidate sensitivity configuration information corresponding to the area type.

8. A safety detection device, characterized in that: include: An acquisition module, used for acquiring target sensitivity configuration information of each detection partition in the detection area, and sending the target sensitivity configuration information to the radar system, so that the radar system can detect each detection partition based on the target sensitivity configuration information; a target determination module, configured to receive regional point cloud data about each of the detection partitions sent by the radar system, and determine a target partition where a target object is located in the plurality of detection partitions based on the regional point cloud data; The safety detection module is used to obtain the safety detection model corresponding to the target partition, input the regional point cloud data into the safety detection model for detection, and determine the safety detection result of the target object.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the security detection method described in any one of claims 1 to 7 when executing the computer program.

10. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the security detection method according to any one of claims 1 to 7 is implemented.

Images