Sleep detection alarm system and method

Abstract

The invention provides a sleep detection alarm system and method, and the system comprises an acquisition module which is used for obtaining a real-time image of a target area; the processing module is used for determining a person as a target person under the condition of determining that the person exists at a target position in the target area according to the real-time image, and acquiring body movement data of the target person through the acquisition module; according to the real-time image, determining an initial sleep detection result, and according to the body movement data, determining an auxiliary sleep detection result; according to the initial post-sleep detection result, determining a target post-sleep detection result in combination with the auxiliary post-sleep detection result; the skyeye module generates alarm push information under the condition that the target sleep detection result indicates that the target person has an illegal sleep behavior; and the terminal module outputs alarm information according to the alarm push information. The whole process does not need manual participation, the automation degree is high, the accurate target post sleep detection result can be determined, and the safety of the target area is effectively guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to a sleep-on-duty detection alarm system and method. Background Art

[0002] With the rapid development of industrialization and informatization, people are paying more and more attention to the safety of work areas. In many important work positions, such as monitoring rooms, production lines, and warehouses, on-duty personnel need to maintain high concentration and vigilance for a long time to ensure the normal operation and safety of the work area. However, long working hours often lead to fatigue, which in turn leads to violations such as sleeping on the job, posing a great threat to the safety of the work area.

[0003] Although some work areas are equipped with surveillance cameras, the monitoring process still relies on real-time observation and judgment of sleeping behavior by humans. This method requires a lot of human resources, resulting in low monitoring efficiency. In addition, it is easily affected by human factors. Staring at the monitoring screen for a long time may cause monitoring personnel to become tired and lose their attention, which will increase the risk of monitoring errors and make it difficult to ensure the safety of the work area. Summary of the invention

[0004] The present invention provides a system and method for detecting and alarming sleeping on the post, which are used to solve the defects that the prior art needs to invest a large amount of human resources, resulting in low monitoring efficiency, and is easily affected by human factors. Staring at the monitoring screen for a long time may cause monitoring personnel to be tired and their attention is reduced, which will increase the risk of monitoring errors and make it difficult to ensure the safety of the work area. The system detects the illegal sleeping on the post behavior of target personnel in the target area in real time and outputs alarm information in time through the interaction between the acquisition module, the processing module, the sky eye module and the terminal module. The whole process does not require human participation and has a high degree of automation. It can determine a relatively accurate target sleeping on the post detection result, thereby effectively ensuring the safety of the target area.

[0005] In a first aspect, the present invention provides a system for detecting and alarming sleeping on the post, comprising: an acquisition module, a processing module, a sky eye module and a terminal module, wherein the acquisition module is used to acquire a real-time image of a target area; the processing module is used to determine the person as a target person when it is determined that there is a person at a target position in the target area according to the real-time image, and obtain the body motion data of the target person through the acquisition module; determine an initial sleeping on the post detection result according to the real-time image, and determine an auxiliary sleeping on the post detection result according to the body motion data; determine a target sleeping on the post detection result according to the initial sleeping on the post detection result and in combination with the auxiliary sleeping on the post detection result; the sky eye module is used to generate an alarm push message when the target sleeping on the post detection result indicates that the target person has violated the regulations by sleeping on the post; and the terminal module is used to output an alarm message according to the alarm push message.

[0006] According to a sleep-on-the-job detection alarm system provided by the present invention, the processing module is used to determine an initial sleep-on-the-job detection result based on the real-time image, and to determine an auxiliary sleep-on-the-job detection result based on the body motion data, including: the processing module is specifically used to perform posture detection on the target person to determine the initial sleep-on-the-job detection result; and to determine the body motion detection result of the target person based on the body motion characteristics corresponding to the body motion data and a preset body motion threshold; when the body motion detection result indicates that the target person has potential sleep-on-the-job behavior, determining that the target person has potential sleep-on-the-job behavior as the auxiliary sleep-on-the-job detection result; when the body motion detection result indicates that the target person does not have potential sleep-on-the-job behavior, determining that the target person does not have potential sleep-on-the-job behavior as the auxiliary sleep-on-the-job detection result.

[0007] According to a sleeping guard detection alarm system provided by the present invention, the processing module is specifically used to perform posture detection on the target person and determine the initial sleeping guard detection result, including: the processing module is specifically used to perform posture detection on the target person, obtain multiple human body key points, and determine the current posture of the target person based on the multiple human body key points; determine the initial sleeping guard detection result based on the current posture.

[0008] According to a sleeping post detection alarm system provided by the present invention, the multiple human body key points include: shoulder key points, hip key points, knee key points, ear key points, wrist key points and nose key points, and the current posture includes: body tilt posture, neck tilt posture, sitting posture, pillow hand related posture and lying related posture; the processing module is specifically used to determine the current posture of the target person according to the multiple human body key points, including: the processing module is specifically used to determine the body tilt posture according to the shoulder key point, the hip key point and the knee key point; determine the neck tilt posture according to the ear key point and the shoulder key point; determine the sitting posture according to the hip key point and the knee key point; determine the pillow hand related posture according to the wrist key point, the nose key point and the ear key point; determine the lying related posture according to the shoulder key point and the hip key point.

[0009] According to a post sleeping detection alarm system provided by the present invention, the processing module is specifically used to determine the initial post sleeping detection result according to the current posture, including: the processing module is specifically used to determine the sleeping posture detection result of the target person according to the current posture; when the sleeping posture detection result indicates that the target person has slept on the post, determine the sleeping duration of the target person; determine the initial post sleeping detection result according to the sleeping duration; when the sleeping posture detection result indicates that the target person does not have any sleeping behavior, determine that the target person does not have any sleeping behavior as the initial post sleeping detection result.

[0010] According to a post sleeping detection alarm system provided by the present invention, the processing module is specifically used to determine the initial post sleeping detection result according to the post sleeping duration, including: the processing module is specifically used to determine that the target person has violated the regulations by sleeping on the post as the initial post sleeping detection result when the post sleeping duration is greater than a preset duration threshold; and to determine that the target person has not violated the regulations by sleeping on the post as the initial post sleeping detection result when the post sleeping duration is less than or equal to the preset duration threshold.

[0011] According to a sleeping post detection alarm system provided by the present invention, the processing module is used to determine the target sleeping post detection result based on the initial sleeping post detection result in combination with the auxiliary sleeping post detection result, including: the processing module is specifically used to determine the target sleeping post detection result as the weighted result of the initial sleeping post detection result and the auxiliary sleeping post detection result.

[0012] According to a post sleeping detection alarm system provided by the present invention, the Sky Eye module is specifically used to generate the alarm push information according to the event danger level corresponding to the illegal sleeping on duty behavior when the target sleeping on duty detection result indicates that the target person has violated the regulations by sleeping on duty.

[0013] According to a post sleeping detection alarm system provided by the present invention, the Sky Eye module is further used to record the real-time image, the sleeping time and personnel information of the target person when the target post sleeping detection result indicates that the target person has violated the regulations by sleeping on the post.

[0014] In a second aspect, the present invention further provides a method for detecting and alarming a sleeping post, which is applied to the sleeping post detection and alarm system described in any one of the first aspects, wherein the sleeping post detection and alarm system comprises: an acquisition module, a processing module, a sky eye module and a terminal module, and the method comprises: using the acquisition module to acquire a real-time image of a target area; using the processing module to determine the person as a target person when it is determined that there is a person at a target position in the target area according to the real-time image; and using the acquisition module to acquire body motion data of the target person; using the processing module to determine an initial sleeping post detection result according to the real-time image, and to determine an auxiliary sleeping post detection result according to the body motion data; using the processing module to determine a target sleeping post detection result according to the initial sleeping post detection result and the auxiliary sleeping post detection result; using the sky eye module to generate an alarm push information when the target sleeping post detection result indicates that the target person has violated the regulations by sleeping on the post; and using the terminal module to output an alarm information according to the alarm push information.

[0015] The present invention provides a system and method for detecting and alarming a sleeping post. The system comprises: an acquisition module, a processing module, a sky eye module and a terminal module, wherein the acquisition module is used to obtain a real-time image of a target area; the processing module is used to determine the person as a target person when it is determined that there is a person at a target position in the target area according to the real-time image, and obtain the body motion data of the target person through the acquisition module; the initial sleeping post detection result is determined according to the real-time image, and the auxiliary sleeping post detection result is determined according to the body motion data; the target sleeping post detection result is determined according to the initial sleeping post detection result and the auxiliary sleeping post detection result; the sky eye module is used to generate an alarm push message when the target sleeping post detection result indicates that the target person has illegal sleeping post behavior; the terminal module is used to output the alarm message according to the alarm push message. The system detects the illegal sleeping post behavior of the target person in the target area in real time and outputs the alarm message in time through the interaction among the acquisition module, the processing module, the sky eye module and the terminal module. The whole process does not require human participation, has a high degree of automation, and can determine a relatively accurate target sleeping post detection result, thereby effectively ensuring the safety of the target area. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 The invention is a schematic structural diagram of the sleeping guard detection alarm system provided by the invention.

[0018] Figure 2 It is a schematic diagram of the supine sleeping posture provided by the present invention.

[0019] Figure 3 It is a schematic diagram of the prone sleeping posture provided by the present invention.

[0020] Figure 4 This is one of the flow charts of the sleeping guard detection alarm method provided by the present invention.

[0021] Figure 5 This is the second flow chart of the sleeping guard detection alarm method provided by the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] The embodiments of the present invention are further described below.

[0024] like Figure 1 , which is a schematic diagram of the structure of the sleep-on-duty detection and alarm system provided by the present invention, may include: a collection module 101, a processing module 102, a sky eye module 103 and a terminal module 104.

[0025] The acquisition module 101 is used to acquire a real-time image of a target area.

[0026] The processing module 102 is used to determine the person as the target person when it is determined that there is a person at the target position in the target area based on the real-time image, and obtain the body motion data of the target person through the acquisition module 101; determine the initial sleeping guard detection result based on the real-time image, and determine the auxiliary sleeping guard detection result based on the body motion data; determine the target sleeping guard detection result based on the initial sleeping guard detection result combined with the auxiliary sleeping guard detection result.

[0027] The Sky Eye module 103 is used to generate an alarm push message when the target sleeping on duty detection result indicates that the target person has violated the regulations by sleeping on duty.

[0028] The terminal module 104 is used to output alarm information according to the alarm push information.

[0029] For example, the target area may be a work area; the target position may be a workstation; and the target person may be a staff member located at the target position in the work area.

[0030] Optionally, the terminal module 104 may be a mobile phone terminal, a personal computer (PC) terminal, or the like.

[0031] Optionally, in the process of obtaining a real-time image of the target area, the acquisition module 101 can record the target area in real time to obtain video data, and process the video data to obtain a real-time image of the target area; the acquisition module 101 can also take a real-time photo of the target area to obtain a real-time image of the target area.

[0032] Optionally, the above data processing may include: video stream decoding, image frame extraction and image preprocessing (such as color space conversion, size adjustment and noise suppression, etc.), etc.

[0033] Since there may be personnel who are not located at the target position in the real-time image of the target area, the personnel are not the target personnel at this time. The personnel may be taking normal resting behaviors at the non-target position. If these normal resting behaviors are subsequently misjudged as illegal sleeping behaviors by the processing module 102, it will cause the subsequent Sky Eye module 103 to generate erroneous alarm push information, which will further cause the terminal module 104 to report a false alarm. Therefore, the processing module 102 will only determine the personnel as the target personnel if it determines that there are personnel at the target position in the target area. In the whole process, not only can unnecessary detection calculations for non-target positions be effectively reduced, thereby improving detection efficiency, but also misjudgment of rest behaviors of personnel at non-target positions can be avoided, thereby reducing the false alarm rate.

[0034] Optionally, the processing module 102 can perform personnel position detection on the real-time image based on position-related information of the real-time image to obtain a position detection result; when the position detection result indicates that there is no person at the target position within the target area, the processing module 102 suspends the detection process until the acquisition module 101 acquires a new real-time image of the target area, and the processing module 102 starts a new round of detection process; when the position detection result indicates that there is a person at the target position within the target area, the person is determined as the target person.

[0035] Optionally, the above-mentioned location-related information can be directly determined by the processing module 102 based on the real-time image, or it can be determined by the Sky Eye module 103 based on the real-time image. The Sky Eye module 103 sends the location-related information to the processing module 102, which is not specifically limited here.

[0036] Optionally, the location-related information may include: target location information and personnel location information, etc.

[0037] Optionally, in the process of determining the position detection result, the processing module 102 can set a coordinate system based on the position-related information, and then perform a position information comparison, that is, compare the acquired personnel position information with the target position information to determine whether the coordinates corresponding to the personnel position information (referred to as: personnel coordinates) fall within the coordinate range corresponding to the target position information. If the personnel coordinates are within the coordinate range, it can be determined that the person is located at the target position within the target area.

[0038] Optionally, in the process of acquiring the body motion data of the target person, the acquisition module 101 may acquire the body motion data of the target person through a remotely connected body motion sensor, and send the body motion data to the processing module 102 .

[0039] In the embodiment of the present invention, the acquisition module 101 first records / photographs the target area in real time to obtain a real-time image of the target area, and sends the real-time image to the processing module 102; then, when the processing module 102 determines that there is a person at the target position in the target area according to the received real-time image, the person is determined as the target person, and the body motion data of the target person is obtained through the acquisition module 101. Then, the processing module 102 can determine the initial sleeping post detection result according to the real-time image, and determine the auxiliary sleeping post detection result according to the body motion data. In this way, the processing module 102 can determine the target sleeping post detection result according to the initial sleeping post detection result and the auxiliary sleeping post detection result, and send the target sleeping post detection result to the sky eye module 103; when the target sleeping post detection result indicates that the target person has violated the regulations by sleeping on the post, the sky eye module 103 generates an alarm push information, and sends the alarm push information to the terminal module 104; then, the terminal module 104 outputs corresponding alarm information according to the alarm push information to prompt the user that the target person in the target area has violated the regulations by sleeping on the post. During the whole process, through the interaction between the collection module 101, the processing module 102, the Sky Eye module 103 and the terminal module 104, the illegal sleeping behavior of the target personnel in the target area is detected in real time and the alarm information is output in time. No human intervention is required, the degree of automation is high, and a relatively accurate target sleeping detection result can be determined, thereby effectively ensuring the safety of the target area.

[0040] In some embodiments, the processing module 102 is used to determine the initial sleeping on the post detection result based on the real-time image, and to determine the auxiliary sleeping on the post detection result based on the body motion data, and may include: the processing module 102 is specifically used to perform posture detection on the target person to determine the initial sleeping on the post detection result; and determine the body motion detection result of the target person based on the body motion characteristics corresponding to the body motion data and a preset body motion threshold; when the body motion detection result indicates that the target person has potential sleeping on the post, determining the target person has potential sleeping on the post as an auxiliary sleeping on the post detection result; when the body motion detection result indicates that the target person does not have potential sleeping on the post, determining the target person does not have potential sleeping on the post as an auxiliary sleeping on the post detection result.

[0041] Optionally, the preset body movement threshold may be set by the processing module 102 before leaving the factory, or may be user-defined, which is not specifically limited here.

[0042] In the embodiment of the present invention, the processing module 102 can perform posture detection on the target person to determine the initial post-sleeping detection result, and the processing module 102 can compare the body motion characteristics with the preset body motion threshold to obtain the body motion detection result of the target person; if the body motion detection result indicates that the target person has potential post-sleeping behavior, it means that the target person may not be in an active state at this time, and the processing module 102 determines that the target person has potential post-sleeping behavior as an auxiliary post-sleeping detection result; if the body motion detection result indicates that the target person does not have potential post-sleeping behavior, it means that the target person may be in an active state at this time, and the processing module 102 determines that the target person does not have potential post-sleeping behavior as an auxiliary post-sleeping detection result. In order to subsequently determine the target post-sleeping detection result.

[0043] In some embodiments, the processing module 102 is specifically used to perform posture detection on the target person and determine the initial sleeping guard detection result, which may include: the processing module 102 is specifically used to perform posture detection on the target person, obtain multiple human body key points, and determine the current posture of the target person based on the multiple human body key points; determine the initial sleeping guard detection result based on the current posture.

[0044] Among them, multiple human key points cover the main joints and feature points of the human body, providing a basis for analyzing human posture and movement.

[0045] In the embodiment of the present invention, after determining the target person, the processing module 102 performs posture detection on the target person to obtain multiple human body key points; then, the processing module 102 analyzes the multiple human body key points to determine the current posture of the target person, and then determines the initial sleeping guard detection result. In the whole process, the processing module 102 can determine the current posture of the target person more accurately based on the multiple human body key points, so that the determined initial sleeping guard detection result is also more accurate.

[0046] In some embodiments, multiple human body key points may include: shoulder key points, hip key points, knee key points, ear key points, wrist key points and nose key points, and the current posture may include: body tilt posture, neck tilt posture, sitting posture, pillow-hand related posture and lying related posture; processing module 102, specifically used to determine the current posture of the target person according to multiple human body key points, may include: processing module 102, specifically used to determine the body tilt posture according to shoulder key points, hip key points and knee key points; determine the neck tilt posture according to ear key points and shoulder key points; determine the sitting posture according to hip key points and knee key points; determine the pillow-hand related posture according to wrist key points, nose key points and ear key points; determine the lying related posture according to shoulder key points and hip key points.

[0047] Among them, the shoulder key points may include: left shoulder key point and right shoulder key point; the hip key points may include: left hip key point and right hip key point; the knee key points may include: left knee key point and right knee key point; the ear key points may include: left ear key point and right ear key point; the wrist key points may include: left wrist key point and right wrist key point.

[0048] Taking the key points of the human body on the left as an example, the processing module 102 specifically includes the following steps 1 to 5 in the process of determining the current posture of the target person.

[0049] Step 1: Determine the first connecting line between the left shoulder key point and the left hip key point and the second connecting line between the left knee key point and the left hip key point, and determine whether the first angle between the first connecting line and the second connecting line is greater than a first preset angle (such as 90 degrees). If greater than, it means that the current posture of the target person is a backward leaning posture; if less than, it means that the current posture of the target person is a forward leaning posture; if equal, it means that the current posture of the target person is a sitting posture. Among them, the backward leaning posture, the forward leaning posture and the sitting posture are all body tilting postures.

[0050] Step 2: Determine the third line between the left ear key point and the left shoulder key point, the second angle between the third line and the vertical direction, and the third angle between the third line and the horizontal direction. If the second angle is less than the second preset angle (such as -45 degrees) or the third angle is greater than the first preset angle (such as 90 degrees), it means that the current posture of the target person is the neck backward posture; if the second angle is equal to the third preset angle (such as 0 degrees) or the third angle is equal to the first preset angle (such as 90 degrees), it means that the current posture of the target person is the natural neck posture (neck is in neutral position); if the second angle is greater than the fourth preset angle (such as 45 degrees) or the third angle is less than the first preset angle (such as 90 degrees), it means that the current posture of the target person is the neck forward posture. Among them, the neck backward posture, the neck natural posture and the neck forward posture all belong to the neck tilt posture.

[0051] Step 3: Determine the fourth line between the left hip key point and the left knee key point, and determine whether the fourth angle between the fourth line and the horizontal direction is less than the fifth preset angle (such as 70 degrees). If less than, it means that the current posture of the target person is sitting; if greater than or equal to, it means that the current posture of the target person may be standing. Among them, the sitting posture and the standing posture are both sitting postures.

[0052] Step 4: Determine the first distance between the left wrist key point and the nose key point, and the second distance between the left ear key point and the nose key point, and determine whether the first distance is less than or equal to the second distance. If so, it means that the current posture of the target person is the hand-rest posture; if greater than, it means that the current posture of the target person is the non-hand-rest posture. Among them, the hand-rest posture and the non-hand-rest posture are both hand-rest-related postures.

[0053] Step 5: Determine the fifth line between the left shoulder key point and the left hip key point, and determine whether the fifth angle between the fifth line and the horizontal direction is greater than the first preset angle (such as 90 degrees). If it is greater than, it means that the current posture of the target person is a lying posture; if it is less than or equal to, it means that the current posture of the target person is a non-lying posture. Among them, the lying posture and the non-lying posture are both lying-related postures.

[0054] It should be noted that the timing of the above steps 1 to 5 is not limited.

[0055] The processing module 102 can also determine the current posture of the target person according to the key points of the human body on the right side, which is similar to the above steps 1 to 5 and will not be described in detail here.

[0056] In addition, the multiple human body key points may include eye key points (left eye key points and right eye key points), elbow key points (left elbow key points and right elbow key points), ankle key points (left ankle key points and right ankle key points), etc. The processing module 102 may also determine the current posture of the target person in combination with these human body key points, which will not be described in detail here.

[0057] In the embodiment of the present invention, the processing module 102 can determine a relatively accurate current posture of the target person by analyzing a plurality of key points of the human body, so as to subsequently determine a relatively accurate initial sleeping guard detection result.

[0058] In some embodiments, the processing module 102 is specifically used to determine the initial sleeping on duty detection result according to the current posture, and may include: the processing module 102 is specifically used to determine the sleeping posture detection result of the target person according to the current posture; when the sleeping posture detection result indicates that the target person has slept on duty, determine the sleeping duration of the target person; determine the initial sleeping on duty detection result according to the sleeping duration; when the sleeping posture detection result indicates that the target person does not have any sleeping on duty, determine that the target person does not have any sleeping on duty as the initial sleeping on duty detection result.

[0059] Specifically, if the current posture of the target person includes: body leaning back posture, neck leaning back posture and sitting posture, it means that the target person may be in a supine sleeping posture at this time, and the processing module 102 determines that the target person has a sleeping posture as the sleeping posture detection result; on the contrary, if the target person is not in a supine sleeping posture, the target person does not have a sleeping posture as the sleeping posture detection result. For example, Figure 2 , which is a schematic diagram of the supine sleeping posture provided by the present invention. Figure 2 Points 1 to 17 are the 17 key points of the human body.

[0060] If the current posture of the target person includes: sitting posture, hand on pillow posture and neck forward posture, it means that the target person may be in a prone sleeping posture, and the processing module 102 determines that the target person has a sleeping posture as the sleeping posture detection result; on the contrary, if the target person is not in a prone sleeping posture, the target person does not have a sleeping posture as the sleeping posture detection result. For example, Figure 3 FIG. 2 is a schematic diagram of the prone sleeping posture provided by the present invention.

[0061] If the current posture of the target person is a lying posture, it means that the target person may be in a lying sleeping posture at this time, and the processing module 102 determines that the target person has a sleeping behavior as the sleeping posture detection result; conversely, if the target person is not in a lying sleeping posture, the target person does not have a sleeping behavior as the sleeping posture detection result.

[0062] Next, the processing module 102 further determines the duration of the target person's sleep on the post when the sleeping posture detection result indicates that the target person has a sleep on the post behavior. Specifically, based on a plurality of continuous real-time images of the target area acquired by the acquisition module 101, the processing module 102 determines the start frame and the end frame of the sleep on the post behavior, and calculates the duration of the target person's sleep on the post in combination with the image frame rate or timestamp. Then, the initial sleep on the post detection result is determined based on the duration of the sleep on the post.

[0063] When the sleeping posture detection result indicates that the target person does not have the behavior of sleeping on the post, the processing module 102 directly determines that the target person does not have the behavior of sleeping on the post as the initial sleeping on the post detection result.

[0064] In the embodiment of the present invention, the processing module 102 determines the target person's sleeping posture detection result according to the current posture of the target person; if the sleeping posture detection result indicates that the target person has sleeping on the post, the target person's sleeping time is determined, and then the initial sleeping on the post detection result is determined according to the sleeping time; if the sleeping posture detection result indicates that the target person does not have sleeping on the post, the target person's non-sleeping on the post is directly determined as the initial sleeping on the post detection result, so as to determine the target sleeping on the post detection result in combination with the auxiliary sleeping on the post detection result.

[0065] In some embodiments, the processing module 102 is specifically used to determine the initial sleeping on the post detection result based on the sleeping time, and may include: the processing module 102 is specifically used to determine that the target person has violated the regulations by sleeping on the post as the initial sleeping on the post detection result when the sleeping time is greater than a preset time threshold; when the sleeping time is less than or equal to the preset time threshold, determine that the target person has not violated the regulations by sleeping on the post as the initial sleeping on the post detection result.

[0066] Optionally, the preset duration threshold may be set by the processing module 102 before leaving the factory, or may be user-defined, which is not specifically limited here.

[0067] In an embodiment of the present invention, if the duration of sleeping on duty is greater than the preset duration threshold, it means that the target person's sleeping on duty behavior has violated the regulations, and the processing module 102 determines that the target person has violated the regulations by sleeping on duty as the initial sleeping on duty detection result; if the duration of sleeping on duty is less than or equal to the preset duration threshold, it means that the duration of the target person's sleeping on duty behavior meets the regulations, for example, the target person may take a temporary lunch break within the prescribed duration, that is, the target person's sleeping on duty behavior may not violate the regulations, and the processing module 102 determines that the target person has not violated the regulations by sleeping on duty as the initial sleeping on duty detection result. Throughout the process, by comparing the duration of sleeping on duty with the preset duration threshold, it is possible to avoid the processing module 102 from misjudging the non-violation sleeping on duty behavior as the violation sleeping on duty behavior, thereby reducing the false alarm rate.

[0068] In some embodiments, the processing module 102 determines the target sleeping post detection result based on the initial sleeping post detection result in combination with the auxiliary sleeping post detection result, which may include: the processing module 102 is specifically used to determine the weighted result of the initial sleeping post detection result and the auxiliary sleeping post detection result as the target sleeping post detection result.

[0069] In an embodiment of the present invention, the processing module 102 can assign different weights according to the importance and credibility of each detection result by weighted processing of the initial sleeping guard detection result and the auxiliary sleeping guard detection result, so as to more reasonably reflect the impact of different detection results on the final judgment and obtain a more comprehensive and accurate weighted result. The processing module 102 determines the weighted result as the target sleeping guard detection result, which is relatively accurate, thereby effectively ensuring the safety of the target area.

[0070] It should be noted that in the actual detection process, it may be affected by various external factors (such as equipment failure, environmental interference and personal habits, etc.), resulting in errors or deviations in the detection results. The processing module 102 can reduce the impact of these errors and interferences on the final judgment to a certain extent and improve the reliability of the target sleeping post detection results by weighted processing of the initial sleeping post detection results and the auxiliary sleeping post detection results.

[0071] In some embodiments, the Sky Eye module 103 is specifically used to generate an alarm push message according to the event danger level corresponding to the illegal sleeping on duty behavior when the target sleeping on duty detection result indicates that the target person has violated the regulations by sleeping on duty.

[0072] Exemplarily, it is assumed that there are three target areas, namely: target area A, target area B and target area C, and target area A is an ordinary area, target area B is a generally important area, and target area C is a very important area. For target area A, if the target sleeping post detection result indicates that target person a in target area A has violated the regulations by sleeping on the post, and the event danger level corresponding to the illegal sleeping post behavior is a low danger level, then the Sky Eye module 103 generates a first-level alarm push message; for target area B, if the target sleeping post detection result indicates that target person b in target area B has violated the regulations by sleeping on the post, and the event danger level corresponding to the illegal sleeping post behavior is a medium danger level, then the Sky Eye module 103 generates a second-level alarm push message; for target area C, if the target sleeping post detection result indicates that target person c in target area C has violated the regulations by sleeping on the post, and the event danger level corresponding to the illegal sleeping post behavior is a high danger level, then the Sky Eye module 103 generates a third-level alarm push message.

[0073] In this way, the terminal module 104 can output corresponding alarm information according to the alarm push information of different levels received. Exemplarily, if the terminal module 104 receives the first-level alarm push information, the terminal module 104 can determine the first-level alarm information, and output the first-level alarm information to the display interface to prompt the user that the event danger level corresponding to the illegal sleeping on the post is a low danger level; if the terminal module 104 receives the second-level alarm push information, the terminal module 104 can determine the second-level alarm information, and output the second-level alarm information to the display interface, and at the same time, a sound alarm or a light alarm can be performed to prompt the user that the event danger level corresponding to the illegal sleeping on the post is a medium danger level; if the terminal module 104 receives the third-level alarm push information, the terminal module 104 can determine the third-level alarm information, and output the third-level alarm information to the display interface, and at the same time, a sound alarm and a light alarm can be performed to prompt the user that the event danger level corresponding to the illegal sleeping on the post is a high danger level. Allow users to take corresponding measures according to different danger levels.

[0074] In an embodiment of the present invention, when the target sleeping on duty detection result indicates that the target person in the target area has violated the regulations by sleeping on duty, the Sky Eye module 103 generates an alarm push information according to the event danger level corresponding to the illegal sleeping on duty behavior, so that the subsequent terminal module 104 can output corresponding alarm information according to the received alarm push information of different levels, and promptly remind the user to take corresponding measures.

[0075] In some embodiments, the Sky Eye module 103 is also used to record real-time images, the target personnel's sleeping time and personnel information when the target sleeping on duty detection result indicates that the target personnel has violated the regulations by sleeping on duty.

[0076] In an embodiment of the present invention, if the target sleeping on duty detection result indicates that the target personnel in the target area have violated the regulations by sleeping on duty, the Sky Eye module 103 will record the real-time image, the target personnel's sleeping time and personnel information for subsequent accountability.

[0077] The following is a description of the post sleeping detection alarm method provided by the present invention. The post sleeping detection alarm method described below and the post sleeping detection alarm system described above can be referenced to each other.

[0078] like Figure 4 As shown, it is one of the flow charts of the sleep guard detection alarm method provided by the present invention, which is applied to Figure 1 The sleep-on-the-job detection and alarm system shown includes: a collection module, a processing module, a sky eye module and a terminal module. The method includes the following steps 401-406.

[0079] Step 401: Use an acquisition module to obtain a real-time image of a target area.

[0080] Step 402: using the processing module, when it is determined based on the real-time image that there is a person at the target position in the target area, the person is determined as a target person; and using the acquisition module, the body motion data of the target person is acquired.

[0081] Step 403: using the processing module, determine the initial sleeping guard detection result according to the real-time image, and determine the auxiliary sleeping guard detection result according to the body movement data.

[0082] Optionally, using a processing module to determine an initial sleeping on the post detection result based on real-time images, and determining an auxiliary sleeping on the post detection result based on body motion data, may include: using the processing module to perform posture detection on the target person to determine an initial sleeping on the post detection result; and using the processing module to determine the body motion detection result of the target person based on body motion features corresponding to the body motion data and a preset body motion threshold; when the body motion detection result indicates that the target person has potential sleeping on the post, using the processing module to determine the target person has potential sleeping on the post as an auxiliary sleeping on the post detection result; when the body motion detection result indicates that the target person does not have potential sleeping on the post, using the processing module to determine the target person does not have potential sleeping on the post as an auxiliary sleeping on the post detection result.

[0083] Optionally, using a processing module to perform posture detection on a target person and determine an initial sleeping guard detection result may include: using a processing module to perform posture detection on a target person to obtain a plurality of human body key points, and using the processing module to determine a current posture of the target person based on the plurality of human body key points; using the processing module to determine an initial sleeping guard detection result based on the current posture.

[0084] Optionally, multiple human body key points may include: shoulder key points, hip key points, knee key points, ear key points, wrist key points and nose key points, and the current posture may include: body tilt posture, neck tilt posture, sitting posture, pillow-hand related posture and lying related posture; using the processing module to determine the current posture of the target person according to multiple human body key points may include: using the processing module to determine the body tilt posture according to the shoulder key points, hip key points and knee key points; using the processing module to determine the neck tilt posture according to the ear key points and shoulder key points; using the processing module to determine the sitting posture according to the hip key points and knee key points; using the processing module to determine the pillow-hand related posture according to the wrist key points, nose key points and ear key points; using the processing module to determine the lying related posture according to the shoulder key points and hip key points.

[0085] Optionally, using a processing module to determine an initial sleeping-on-the-job detection result according to a current posture may include: using a processing module to determine a sleeping posture detection result of a target person according to the current posture; when the sleeping posture detection result indicates that the target person has slept on the job, using the processing module to determine the sleeping duration of the target person; using the processing module to determine an initial sleeping-on-the-job detection result according to the sleeping duration; when the sleeping posture detection result indicates that the target person has not slept on the job, using the processing module to determine the initial sleeping-on-the-job detection result as the target person has not slept on the job.

[0086] Optionally, a processing module is used to determine an initial sleeping on duty detection result based on the sleeping duration, which may include: when the sleeping duration is greater than a preset duration threshold, using the processing module to determine that the target person has violated the regulations by sleeping on duty as the initial sleeping on duty detection result; when the sleeping duration is less than or equal to the preset duration threshold, using the processing module to determine that the target person has not violated the regulations by sleeping on duty as the initial sleeping on duty detection result.

[0087] Step 404: using the processing module, according to the initial sleeping post detection result and in combination with the auxiliary sleeping post detection result, determine the target sleeping post detection result.

[0088] Optionally, using a processing module to determine a target sleeping post detection result based on an initial sleeping post detection result in combination with an auxiliary sleeping post detection result may include: using the processing module to determine a weighted result of the initial sleeping post detection result and the auxiliary sleeping post detection result as the target sleeping post detection result.

[0089] Step 405: using the Sky Eye module, when the target sleeping on duty detection result indicates that the target person has violated the regulations by sleeping on duty, an alarm push message is generated.

[0090] Optionally, utilizing the Sky Eye module, when the target sleeping on the post detection result indicates that the target person has violated the regulations by sleeping on the post, generating an alarm push message, may include: utilizing the Sky Eye module, when the target sleeping on the post detection result indicates that the target person has violated the regulations by sleeping on the post, generating an alarm push message according to the event danger level corresponding to the illegal sleeping on the post behavior.

[0091] Optionally, the method may also include: utilizing the Sky Eye module to record real-time images, the target personnel's sleeping time and personnel information when the target sleeping on duty detection result indicates that the target personnel has violated the regulations by sleeping on duty.

[0092] Step 406: Utilize the terminal module to output alarm information according to the alarm push information.

[0093] Take the key points of the human body on the left as an example. Figure 5 The figure shows the second flow chart of the sleeping guard detection alarm method provided by the present invention. Figure 5 The process shown is the same as the process described above for the sleeping guard detection alarm system to determine the target sleeping guard detection result and output alarm information, and will not be described in detail here.

[0094] In an embodiment of the present invention, a real-time image of a target area is acquired by using an acquisition module; a processing module is used to determine the existence of a person at a target position in the target area according to the real-time image, and the person is determined as a target person; and the acquisition module is used to acquire the body motion data of the target person; a processing module is used to determine the initial sleeping post detection result according to the real-time image, and the auxiliary sleeping post detection result is determined according to the body motion data; a processing module is used to determine the target sleeping post detection result according to the initial sleeping post detection result and the auxiliary sleeping post detection result; a sky eye module is used to generate an alarm push message when the target sleeping post detection result indicates that the target person has violated the regulations by sleeping on the post; and a terminal module is used to output the alarm message according to the alarm push message. This method can detect the illegal sleeping post behavior of the target person in the target area in real time and output the alarm message in time. The whole process does not require human participation, has a high degree of automation, and can determine a relatively accurate target sleeping post detection result, thereby effectively ensuring the safety of the target area.

[0095] The system embodiment described above is merely illustrative, wherein the units described 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 they may be 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. Those of ordinary skill in the art may understand and implement it without creative work.

[0096] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0097] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A sleeping post detection alarm system, characterized in that: include: Acquisition module, processing module, Sky Eye module and terminal module, among which, The acquisition module is used to obtain a real-time image of the target area; The processing module is used to determine the person as a target person when it is determined that there is a person at the target position in the target area according to the real-time image, and obtain the body motion data of the target person through the acquisition module; determine the initial sleeping guard detection result according to the real-time image, and determine the auxiliary sleeping guard detection result according to the body motion data; determine the target sleeping guard detection result according to the initial sleeping guard detection result and the auxiliary sleeping guard detection result; The Sky Eye module is used to generate an alarm push message when the target sleeping on duty detection result indicates that the target person has violated the regulations by sleeping on duty; The terminal module is used to output alarm information according to the alarm push information.

2. The sleeping post detection alarm system according to claim 1, characterized in that: The processing module is used to determine the initial sleep-on-the-job detection result according to the real-time image, and to determine the auxiliary sleep-on-the-job detection result according to the body motion data, including: The processing module is specifically used to perform posture detection on the target person to determine the initial sleeping-on-the-job detection result; and determine the body motion detection result of the target person based on the body motion features corresponding to the body motion data and in combination with a preset body motion threshold; when the body motion detection result indicates that the target person has potential sleeping-on-the-job behavior, the presence of the target person's potential sleeping-on-the-job behavior is determined as the auxiliary sleeping-on-the-job detection result; when the body motion detection result indicates that the target person does not have potential sleeping-on-the-job behavior, the absence of the target person's potential sleeping-on-the-job behavior is determined as the auxiliary sleeping-on-the-job detection result.

3. The sleeping post detection alarm system according to claim 2, characterized in that: The processing module is specifically used to perform posture detection on the target person and determine the initial sleeping post detection result, including: The processing module is specifically used to perform posture detection on the target person, obtain multiple human body key points, and determine the current posture of the target person based on the multiple human body key points; and determine the initial sleeping position detection result based on the current posture.

4. The sleeping post detection alarm system according to claim 3, characterized in that: The multiple human body key points include: shoulder key points, hip key points, knee key points, ear key points, wrist key points and nose key points; the current posture includes: body tilt posture, neck tilt posture, sitting posture, pillow hand related posture and lying related posture; the processing module is specifically used to determine the current posture of the target person according to the multiple human body key points, including: The processing module is specifically used to determine the body tilt posture according to the shoulder key point, the hip key point and the knee key point; determine the neck tilt posture according to the ear key point and the shoulder key point; determine the sitting posture according to the hip key point and the knee key point; determine the pillow hand related posture according to the wrist key point, the nose key point and the ear key point; determine the lying related posture according to the shoulder key point and the hip key point.

5. The sleeping post detection alarm system according to claim 3, characterized in that: The processing module is specifically used to determine the initial sleeping post detection result according to the current posture, including: The processing module is specifically used to determine the sleeping posture detection result of the target person according to the current posture; when the sleeping posture detection result indicates that the target person has slept on the post, determine the sleeping duration of the target person; determine the initial sleeping on the post detection result according to the sleeping duration; when the sleeping posture detection result indicates that the target person has not slept on the post, determine that the target person has not slept on the post as the initial sleeping on the post detection result.

6. The sleeping post detection alarm system according to claim 5, characterized in that: The processing module is specifically used to determine the initial sleep-on-the-job detection result according to the sleep-on-the-job duration, including: The processing module is specifically used to determine that the target person has violated the regulations by sleeping on the job as the initial sleeping on the job detection result when the sleeping time is greater than a preset time threshold; and to determine that the target person has not violated the regulations by sleeping on the job as the initial sleeping on the job detection result when the sleeping time is less than or equal to the preset time threshold.

7. The sleeping post detection alarm system according to any one of claims 1 to 6, characterized in that: The processing module is used to determine the target sleeping post detection result according to the initial sleeping post detection result in combination with the auxiliary sleeping post detection result, including: The processing module is specifically used to determine a weighted result of the initial sleeping post detection result and the auxiliary sleeping post detection result as the target sleeping post detection result.

8. The sleeping post detection alarm system according to any one of claims 1 to 6, characterized in that: The Sky Eye module is specifically used to generate the alarm push information according to the event danger level corresponding to the illegal sleeping on duty behavior when the target sleeping on duty detection result indicates that the target person has violated the regulations by sleeping on duty.

9. The sleeping post detection alarm system according to any one of claims 1 to 6, characterized in that: The Sky Eye module is also used to record the real-time image, the sleeping time and personnel information of the target person when the target sleeping on duty detection result indicates that the target person has violated the regulations by sleeping on duty.

10. A method for detecting and alarming a person sleeping on the post, characterized in that: The method is applied to the sleep-on-the-job detection and alarm system according to any one of claims 1 to 9, wherein the sleep-on-the-job detection and alarm system comprises: a collection module, a processing module, a sky eye module and a terminal module, and the method comprises: Using the acquisition module, a real-time image of the target area is acquired; Using the processing module, when it is determined according to the real-time image that there is a person at the target position in the target area, the person is determined as a target person; and using the acquisition module, the body motion data of the target person is acquired; Using the processing module, determine an initial sleep-on-the-post detection result according to the real-time image, and determine an auxiliary sleep-on-the-post detection result according to the body movement data; Determine the target sleeping post detection result by using the processing module according to the initial sleeping post detection result and in combination with the auxiliary sleeping post detection result; Using the Sky Eye module, when the target sleeping on duty detection result indicates that the target person has violated the regulations by sleeping on duty, an alarm push message is generated; The terminal module is used to output alarm information according to the alarm push information.

Images