Method and system for identifying person standing under hung object and electronic equipment

Abstract

The invention provides a method and a system for identifying a person standing under a suspended object and electronic equipment, and relates to the technical field of monitoring image processing. The identification method comprises the following steps: acquiring target image information acquired by monitoring equipment; setting at least one algorithm recognition area according to the target image information; performing image recognition, depth estimation and coordinate transformation on the algorithm recognition area to obtain three-dimensional point cloud coordinates of human body key points, three-dimensional point cloud coordinates of ground contour points and three-dimensional point cloud coordinates of suspended object bottom contour points; determining whether a person is on the ground according to the three-dimensional point cloud coordinates of the human body key points and the three-dimensional point cloud coordinates of the ground contour points; determining whether the hoisted object is on the ground according to the three-dimensional point cloud coordinates of the bottom contour points of the hoisted object and the three-dimensional point cloud coordinates of the ground contour points; respectively determining three-dimensional coordinates of the human body key points and the suspended object bottom contour points in a world coordinate system; and judging whether the person stands under the hanging object according to the three-dimensional coordinates of the key points of the human body and the three-dimensional coordinates of the bottom contour points of the hanging object.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring image processing, and in particular to a method, a system and an electronic device for identifying a person standing under a suspended object. Background Art

[0002] Identification of people standing under a hoisted object is a technology used for production safety monitoring, aiming to prevent safety accidents during hoisting operations. Identification of people standing under a hoisted object is an important protective measure for government supervision of construction safety, which can effectively avoid the risk of accidents during the construction process.

[0003] The identification of people standing under the suspended object generally includes the top-down perspective identification and the horizontal angle identification. Currently, the top-down perspective identification can obtain images through the camera device and generate the circumscribed rectangle of the upper surface of the suspended object to determine whether there is a person standing under the suspended object. The horizontal angle identification generally requires the use of binocular cameras or laser radars for identification.

[0004] However, the inventors found that the effective working distance of the binocular camera is relatively short (for example, generally within 10 meters), which has certain limitations; and although the recognition accuracy of the laser radar is relatively high, there is still a problem of high cost. Summary of the invention

[0005] The present invention provides a method, system and electronic device for identifying a person standing under a suspended object, which are used to solve the problems of high cost and low recognition accuracy (short effective working distance) in the current method of identifying a person standing under a suspended object based on horizontal angle.

[0006] According to one aspect of the present invention, there is provided a method for identifying a person standing under a suspended object, the method comprising: acquiring target image information collected by a monitoring device; setting at least one algorithm identification area according to the target image information; performing image recognition, depth estimation and coordinate conversion on the algorithm identification area to obtain three-dimensional point cloud coordinates of human key points, three-dimensional point cloud coordinates of ground contour points and three-dimensional point cloud coordinates of bottom contour points of the suspended object corresponding to the human key points, ground contour points and bottom contour points of the suspended object respectively; determining whether a person is on the ground according to the three-dimensional point cloud coordinates of the human key points and the three-dimensional point cloud coordinates of the ground contour points; determining whether the suspended object is on the ground according to the three-dimensional point cloud coordinates of the bottom contour points of the suspended object and the three-dimensional point cloud coordinates of the ground contour points; when it is determined that a person is on the ground and that the suspended object is not on the ground, respectively determining the three-dimensional coordinates of the human key points and the bottom contour points of the suspended object in the world coordinate system; judging whether a person is standing under the suspended object according to the three-dimensional coordinates of the human key points and the three-dimensional coordinates of the bottom contour points of the suspended object.

[0007] According to some embodiments of the present invention, obtaining target image information collected by a monitoring device includes: obtaining initial image information collected by the monitoring device; calibrating parameters of the initial image information; and performing distortion correction on the initial image information according to parameters obtained by the parameter calibration to obtain target image information.

[0008] According to some embodiments of the present invention, image recognition, depth estimation and coordinate conversion are performed on the algorithm recognition area to obtain the three-dimensional point cloud coordinates of the human body key points, the ground contour points and the bottom contour points of the hanging object respectively corresponding to the human body key points, the ground contour points and the bottom contour points of the hanging object, including: performing human body key point detection and image segmentation on the algorithm recognition area respectively to obtain the human body key points, the ground contour points and the bottom contour points of the hanging object; based on the depth estimation, the human body key point depth values, the ground contour point depth values ​​and the bottom contour point depth values ​​corresponding to the human body key points, the ground contour points and the bottom contour points of the hanging object are obtained; the three-dimensional point cloud coordinates of the human body key points, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object are calculated according to the depth values ​​of the human body key points, the depth values ​​of the ground contour points and the depth values ​​of the bottom contour points of the hanging object.

[0009] According to some embodiments of the present invention, determining whether a person is on the ground based on the three-dimensional point cloud coordinates of the key points of the human body and the three-dimensional point cloud coordinates of the ground contour points includes: constructing a ground plane equation based on the three-dimensional point cloud coordinates of the ground contour points; determining a first distance from the three-dimensional point cloud coordinates of the key points of the human body to the ground plane equation; when the first distance is less than a first preset threshold, determining that the person is on the ground; determining whether a suspended object is on the ground based on the three-dimensional point cloud coordinates of the bottom contour points of the suspended object and the three-dimensional point cloud coordinates of the ground contour points includes: constructing a ground plane equation based on the three-dimensional point cloud coordinates of the ground contour points; determining a second distance from the three-dimensional point cloud coordinates of the bottom contour points of the suspended object to the ground plane equation; when the second distance is greater than a second preset threshold, determining that the suspended object is not on the ground.

[0010] According to some embodiments of the present invention, judging whether a person is standing under a suspended object based on the three-dimensional coordinates of key points of the human body and the three-dimensional coordinates of the bottom contour points of the suspended object includes: determining the three-dimensional coordinate range of the key points of the human body based on the three-dimensional coordinates of the key points of the human body; determining the three-dimensional coordinate range of the bottom contour points of the suspended object based on the three-dimensional coordinates of the bottom contour points of the suspended object; and judging that the person is standing under the suspended object when the three-dimensional coordinate range of the key points of the human body is within the three-dimensional coordinate range of the bottom contour points of the suspended object.

[0011] According to another aspect of the present invention, the present invention provides a system for identifying people standing under a hanging object. The identification system includes an image processing module, a coordinate processing module and an information recognition module. The image processing module obtains the target image information collected by the monitoring device, and sets at least one algorithm recognition area according to the target image information; the coordinate processing module performs image recognition, depth estimation and coordinate conversion on the algorithm recognition area to obtain the three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object corresponding to the key points of the human body, the ground contour points and the bottom contour points of the hanging object respectively; the information recognition module determines whether a person is on the ground according to the three-dimensional point cloud coordinates of the key points of the human body and the three-dimensional point cloud coordinates of the ground contour points, and determines whether the hanging object is on the ground according to the three-dimensional point cloud coordinates of the bottom contour points of the hanging object and the three-dimensional point cloud coordinates of the ground contour points; the coordinate processing module determines the three-dimensional coordinates of the key points of the human body and the bottom contour points of the hanging object in the world coordinate system respectively when it is determined that a person is on the ground and the hanging object is not on the ground; the information recognition module determines whether a person is standing under the hanging object according to the three-dimensional coordinates of the key points of the human body and the three-dimensional coordinates of the bottom contour points of the hanging object.

[0012] According to some embodiments of the present invention, the coordinate processing module performs human key point detection and image segmentation on the algorithm recognition area to obtain human key points, ground contour points and bottom contour points of suspended objects; the coordinate processing module obtains the human key point depth values, ground contour point depth values ​​and bottom contour point depth values ​​corresponding to the human key points, ground contour points and bottom contour points of suspended objects based on depth estimation; the coordinate processing module calculates the three-dimensional point cloud coordinates of the human key points, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the suspended objects according to the depth values ​​of the human key points, the depth values ​​of the ground contour points and the depth values ​​of the bottom contour points of the suspended objects.

[0013] According to some embodiments of the present invention, the information recognition module constructs a ground plane equation based on the three-dimensional point cloud coordinates of the ground contour points; the information recognition module determines a first distance from the three-dimensional point cloud coordinates of the key points of the human body to the ground plane equation; the information recognition module determines that the person is on the ground when the first distance is less than a first preset threshold; and the information recognition module constructs a ground plane equation based on the three-dimensional point cloud coordinates of the ground contour points; the information recognition module determines a second distance from the three-dimensional point cloud coordinates of the bottom contour points of the hanging object to the ground plane equation; the information recognition module determines that the hanging object is not on the ground when the second distance is greater than a second preset threshold.

[0014] According to some embodiments of the present invention, the information recognition module determines the three-dimensional coordinate range of the key points of the human body based on the three-dimensional coordinates of the key points of the human body; the information recognition module determines the three-dimensional coordinate range of the bottom contour points of the suspended object based on the three-dimensional coordinates of the bottom contour points of the suspended object; the information recognition module determines that a person is standing under a suspended object when the three-dimensional coordinate range of the key points of the human body is within the three-dimensional coordinate range of the bottom contour points of the suspended object.

[0015] According to another aspect of the present invention, an electronic device is provided, which includes: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors can implement the identification method described above.

[0016] According to another aspect of the present invention, a non-volatile computer-readable storage medium is provided, wherein a computer program is stored on the storage medium, and when the computer program is executed by a processor, the identification method described above can be implemented.

[0017] The present invention performs image recognition, depth estimation and coordinate conversion on the algorithm recognition area to obtain the three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object corresponding to the key points of the human body, the ground contour points and the bottom contour points of the hanging object, and further judges whether the person is on the ground and whether the hanging object is on the ground according to the three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object. When it is determined that the person is on the ground and the hanging object is not on the ground at the same time, the three-dimensional coordinates of the key points of the human body and the bottom contour points of the hanging object can accurately judge whether the person is standing under the hanging object.

[0018] The recognition method provided by the present invention can recognize whether a person is standing under a suspended object in a horizontal viewing angle based on image processing and coordinate conversion in a single monitoring device scenario. The present invention does not need to use recognition methods such as binocular cameras or laser radars, and has the characteristics of low cost and accurate recognition of whether a person is standing under a suspended object. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 A flow chart showing an identification method according to an embodiment of the present invention;

[0021] Figure 2Another flow chart showing the identification method according to the embodiment of the present invention;

[0022] Figure 3 Another flow chart showing the identification method according to the embodiment of the present invention;

[0023] Figure 4 A schematic diagram showing key points of a human body according to an embodiment of the present invention;

[0024] Figure 5 Another flow chart showing the identification method according to the embodiment of the present invention;

[0025] Figure 6 Another flow chart showing the identification method according to the embodiment of the present invention;

[0026] Figure 7 Another flow chart showing the identification method according to the embodiment of the present invention;

[0027] Figure 8 A schematic diagram showing an identification system according to an embodiment of the present invention.

[0028] Description of reference numerals:

[0029] Recognition system 1; image processing module 10; coordinate processing module 20; information recognition module 30. DETAILED DESCRIPTION

[0030] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that the present invention will be comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0031] The described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of these specific details, or other modes, components, materials, devices, etc. may be adopted. In these cases, known structures, methods, devices, implementations, materials or operations will not be shown or described in detail.

[0032] In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0033] The terms "first", "second" and the like in the specification, claims and the above drawings of the present invention are used to distinguish different objects rather than to describe a specific order.

[0034] The technical solution of the present invention is described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0035] The full names of the English abbreviations involved in the present invention and their Chinese meanings are:

[0036] RTSP: Real Time Streaming Protocol, real-time streaming protocol;

[0037] ArUco Marker: Binary square marking method;

[0038] SSD: Single Shot MultiBox Detector, a target detection algorithm based on deep learning;

[0039] YOLO: You Only Look Once, a real-time target detection algorithm based on deep learning;

[0040] RetinaNet: a single-stage object detection model.

[0041] RT-DETR: Real-Time Detection Transformer, real-time end-to-end object detection model;

[0042] Deeplab: image semantic segmentation model;

[0043] Mask2Former:image segmentation model;

[0044] SegFormer: semantic segmentation model.

[0045] OpenPose: Human pose estimation technology based on deep learning;

[0046] AlphaPose - a human pose estimation library based on deep learning technology.

[0047] HRNet: High-ResolutionNet, high-resolution deep neural network.

[0048] According to one aspect of the present invention, a method for identifying people standing under a suspended object is provided, which is used to identify whether there are people standing under the suspended object, so as to avoid the risk of accidents to people during construction.

[0049] Figure 1 A flow chart of the identification method according to an embodiment of the present invention is shown. Figure 1 As shown, the identification method may include steps S100 to S700. Exemplarily, the identification method may be executed by an identification system for people standing under a suspended object with computing capability.

[0050] In step S100, the recognition system obtains target image information collected by the monitoring device.

[0051] For example, the monitoring device can be a monocular camera, which is installed at a preset monitoring position according to monitoring requirements and is approximately parallel to the ground. It is used to collect monitoring images at the lifting equipment. The monitoring device has a fixed installation position, monitoring angle and focal length, etc.

[0052] Optionally, the viewing angle between the monitoring device and the horizontal plane can be 0° to 15°. By setting the viewing angle between the monitoring device and the horizontal plane in this way, the identification of the person standing under the suspended object can be performed at a horizontal angle.

[0053] Exemplarily, the lifting equipment includes, but is not limited to, equipment such as cranes, slings, cranes and elevators that can lift objects, and the present invention is not limited to this.

[0054] Alternatively, if Figure 2 As shown, step S100 may also include steps S110-S130.

[0055] In step S110 , the recognition system obtains initial image information collected by the monitoring device.

[0056] For example, the recognition system is connected to the monocular camera for communication, obtains the RTSP video stream from the monocular camera, and decodes the RTSP video stream to obtain the initial image information at the lifting device.

[0057] Exemplarily, the initial image information may be color image information or grayscale image information.

[0058] In step S120, the recognition system performs parameter calibration on the initial image information.

[0059] In step S130, the recognition system performs distortion correction on the initial image information according to the parameters obtained by parameter calibration to obtain target image information.

[0060] For example, parameter calibration includes internal parameter calibration and external parameter calibration. For example, the recognition system first performs internal parameter calibration based on a pre-selected calibration method (such as Zhang's calibration method), and then performs distortion correction on the color image information according to the obtained internal parameter matrix and distortion coefficient.

[0061] The recognition system then performs external parameter calibration on the color image information after distortion correction, such as external parameter calibration based on ArUcoMarker, to obtain the rotation matrix R and offset matrix T of the posture of the monitoring device (such as a camera).

[0062] In step S200, the recognition system sets at least one algorithm recognition area according to the target image information.

[0063] For example, the recognition system responds to user operation instructions and sets one or more algorithm recognition areas based on target image information according to user needs. The algorithm recognition area can be a user-defined area with monitoring and recognition significance, that is, an alarm will be issued only when a person is recognized in the algorithm recognition area.

[0064] Exemplarily, the recognition system may generate a polygonal area based on coordinates of points within the algorithm recognition area.

[0065] In step S300, the recognition system performs image recognition, depth estimation and coordinate conversion on the algorithm recognition area to obtain the three-dimensional point cloud coordinates of the human body key points, the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object corresponding to the human body key points, the ground contour points and the bottom contour points of the hanging object respectively.

[0066] For example, the recognition system can segment and identify the key points of the human body, the contour points of the ground, and the contour points of the bottom of the hanging object by performing human key point detection and image segmentation on the algorithm recognition area.

[0067] Exemplarily, the key points of the human body may be the key points at the foot area of ​​the human body (such as the ankle); the ground contour points may be the contour points of the target ground area pre-selected by the user according to needs; the contour points of the bottom of the hanging object may be the contour points of the plane presented by the bottom of the hanging object.

[0068] Through depth estimation and coordinate conversion processing, the recognition system can obtain the corresponding three-dimensional point cloud coordinates of human body key points, three-dimensional point cloud coordinates of ground contour points and three-dimensional point cloud coordinates of the bottom contour points of the hanging object.

[0069] Exemplarily, the three-dimensional point cloud coordinates are (x, y, z).

[0070] Alternatively, if Figure 3 As shown, step S300 may also include steps S310-S330.

[0071] In step S310, the recognition system performs human key point detection and image segmentation on the algorithm recognition area to obtain human key points, ground contour points and bottom contour points of the hanging object.

[0072] For example, the recognition system determines the human body area image in the target image information based on image segmentation recognition technology, and detects the human body key points based on the human body segmentation technology, and then obtains the human body key point coordinates corresponding to the human body key points through the human body key point information detection technology.

[0073] And the recognition system determines the ground contour points and the bottom contour points of the hanging object in the target image information based on the image segmentation recognition technology, and then obtains the corresponding ground contour point coordinates and the bottom contour point coordinates of the hanging object.

[0074] Exemplarily, image segmentation and recognition technologies include but are not limited to SSD, YOLO, RetinaNet or RT-DETR, etc.; human body segmentation technologies include but are not limited to Deeplab, Mask2Former or SegFormer, etc., and the present invention is not limited to this.

[0075] For example, Figure 4 As shown, the ankle key points can be detected by human key point information detection technology. The human key point detection technology includes but is not limited to OpenPose, AlphaPose or HRNet, etc., and the present invention is not limited to this.

[0076] In step S320, the recognition system obtains the depth values ​​of the human body key points, the ground contour points and the bottom contour points of the suspended object corresponding to the human body key points, the ground contour points and the bottom contour points of the suspended object based on depth estimation.

[0077] For example, the recognition system performs monocular depth estimation based on the target image information and can obtain the depth values ​​of all pixels in the target image information. Then, the recognition system determines the corresponding depth values ​​of the key points of the human body, the ground contour points and the bottom contour points of the hanging objects according to the coordinates of the key points of the human body, the ground contour points and the bottom contour points of the hanging objects.

[0078] In step S330, the recognition system calculates the three-dimensional point cloud coordinates of the human body key points, the ground contour points and the bottom contour points of the hanging object according to the depth values ​​of the human body key points, the ground contour points and the bottom contour points of the hanging object.

[0079] For example, the recognition system converts the depth values ​​of human body key points, ground contour points and bottom contour points of suspended objects into three-dimensional point cloud coordinates of human body key points, ground contour points and bottom contour points of suspended objects.

[0080] Exemplarily, the calculation formula of the three-dimensional point cloud coordinates (x, y, z) can be:

[0081]

[0082] Among them, (u, ν) is the coordinate of the pixel point, z c is the depth value of the pixel (u, ν),

[0083] (u0, ν0) are the coordinates of the principal point; f x The length of the focal length in the x-axis direction is described in pixels; f y The length of the focal length in the y-axis direction is described in pixels; depth_scale is the scale factor corresponding to the depth map (the ratio of the value stored in the depth map to the actual depth).

[0084] In step S400, the recognition system determines whether there is a person on the ground based on the three-dimensional point cloud coordinates of the key points of the human body and the three-dimensional point cloud coordinates of the ground contour points.

[0085] For example, the recognition system can determine whether the recognized person is on the ground based on the distance correlation between the three-dimensional point cloud coordinates of the key points of the human body and the three-dimensional point cloud coordinates of the ground contour points.

[0086] Alternatively, if Figure 5 As shown, step S400 may also include S410-S430.

[0087] In step S410, the recognition system constructs a ground plane equation according to the three-dimensional point cloud coordinates of the ground contour points.

[0088] For example, the equation of the ground plane is A x +B y +C z +D=0.

[0089] Among them, A, B, C, and D are all coefficients.

[0090] Exemplarily, the fitting plane may be solved based on the RANSAC method, so that the coefficients A, B, C, and D may be obtained.

[0091] In step S420, the recognition system determines a first distance from the three-dimensional point cloud coordinates of the key point of the human body to the ground plane equation.

[0092] In step S430, the recognition system determines that the person is on the ground when the first distance is less than a first preset threshold.

[0093] For example, the recognition system calculates the distance (ie, the first distance) from the three-dimensional point cloud coordinates of the key points of the human body to the ground plane equation. When the first distance is less than a first preset threshold, it can be determined that the person is on the ground.

[0094] With such arrangement, the present invention can map the distance relationship between the actual key points of the human body and the actual ground through the distance from the three-dimensional point cloud coordinates of the key points of the human body to the ground plane equation, thereby determining whether there is a person on the ground.

[0095] In step S500, the recognition system determines whether the hanging object is on the ground according to the three-dimensional point cloud coordinates of the bottom contour points of the hanging object and the three-dimensional point cloud coordinates of the ground contour points.

[0096] For example, the recognition system can determine whether the hanging object is placed in a placed state or in a hoisted state based on the distance correlation between the three-dimensional point cloud coordinates of the bottom contour points of the hanging object and the three-dimensional point cloud coordinates of the ground contour points.

[0097] Alternatively, if Figure 6 As shown, step S500 may also include S510-S530.

[0098] In step S510, the recognition system constructs a ground plane equation according to the three-dimensional point cloud coordinates of the ground contour points.

[0099] The ground plane equation has been described in detail above and will not be repeated here.

[0100] In step S520, the recognition system determines a second distance from the three-dimensional point cloud coordinates of the bottom contour point of the hanging object to the ground plane equation.

[0101] In step S530, the recognition system determines that the hanging object is not on the ground when the second distance is greater than a second preset threshold.

[0102] For example, the recognition system calculates the distance from the three-dimensional point cloud coordinates of the bottom contour point of the hanging object to the ground plane equation (i.e., the second distance). When the second distance is greater than the second preset threshold, it can be determined that the hanging object is not on the ground but is in a hanging state.

[0103] With such a configuration, the present invention can map the distance relationship between the actual bottom contour points of the hanging object and the actual ground through the distance from the three-dimensional point cloud coordinates of the bottom contour points of the hanging object to the ground plane equation, thereby determining whether the hanging object is in a placed state on the ground or in a hoisted state.

[0104] In step S600, when it is determined that there is a person on the ground and that the hanging object is not on the ground, the recognition system determines the three-dimensional coordinates of the key points of the human body and the contour points of the bottom of the hanging object in the world coordinate system respectively.

[0105] For example, the recognition system converts the two-dimensional coordinates of the key points of the human body and the contour points of the ground in the pixel coordinate system into three-dimensional coordinates in the world coordinate system based on the depth values ​​of the key points of the human body and the depth values ​​of the contour points at the bottom of the hanging object.

[0106] Exemplarily, the conversion relationship between the two-dimensional coordinates of the pixel coordinate system and the three-dimensional coordinates of the world coordinate system may be:

[0107]

[0108] Among them, (u, ν) is the two-dimensional coordinate of the pixel point; (x w ,y w , z w ) is the three-dimensional coordinate of the pixel point; z c is the depth value of the pixel point (u, ν); (u0, ν0) is the coordinate of the principal point (the unit is also pixel); f is the focal length, f x The length of the focal length in the x-axis direction is described in pixels; f y is to use pixels to describe the length of the focal length in the y-axis direction; R is the rotation matrix of the camera posture (the matrix size is 3*3); T is the offset matrix of the camera posture (the matrix size is 3*1); the camera intrinsic parameter matrix size is 3*4; the camera extrinsic parameter matrix size is 4*4; d x is the width of one pixel in the X direction (in mm); d x is the width of one pixel in the Y direction of the pixel point (in mm); w Vertical ground.

[0109] In step S700, the recognition system determines whether a person is standing under the suspended object based on the three-dimensional coordinates of the key points of the human body and the contour points of the bottom of the suspended object.

[0110] For example, when it is determined that a person is on the ground and that a suspended object is not on the ground, it is necessary to further determine whether the identified person is standing under the suspended object.

[0111] Alternatively, if Figure 7 As shown, step S700 may also include steps S710-S730.

[0112] In step S710, the recognition system determines the three-dimensional coordinate range of the key points of the human body according to the three-dimensional coordinates of the key points of the human body.

[0113] In step S720, the recognition system determines the three-dimensional coordinate range of the bottom contour point of the hanging object according to the three-dimensional coordinates of the bottom contour point of the hanging object.

[0114] In step S730, when the three-dimensional coordinate range of the key points of the human body is within the three-dimensional coordinate range of the bottom contour points of the hanging object, the recognition system determines that the person is standing under the hanging object.

[0115] For example, the three-dimensional coordinate range of the key points of the human body is (x foot ,y foot ), the three-dimensional coordinate range of the bottom contour point of the hanging object is (x bottom ,y bottom ). w is vertical to the ground, if (x foot ,y foot ) is in (x bottom ,y bottom ), it means that the key point of the human body is below the bottom contour point of the suspended object in the vertical direction, that is, the person is standing under the suspended object.

[0116] With such a configuration, the present invention can further map the range relationship between the actual key points of the human body and the actual bottom contour points of the hanging object (i.e., determine whether the person is in the range below the hanging object) based on the relationship between the three-dimensional coordinate range of the key points of the human body and the three-dimensional coordinate range of the bottom contour points of the hanging object when judging that a person is on the ground and the hanging object is not on the ground, so as to determine whether the person is standing under the hanging object. In this way, when judging whether the person is standing under the hanging object, an alarm prompt is issued to avoid safety accidents.

[0117] The present invention performs image recognition, depth estimation and coordinate conversion on the algorithm recognition area to obtain the three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object corresponding to the key points of the human body, the ground contour points and the bottom contour points of the hanging object, and further judges whether the person is on the ground and whether the hanging object is on the ground according to the three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object. When it is determined that the person is on the ground and the hanging object is not on the ground at the same time, the three-dimensional coordinates of the key points of the human body and the bottom contour points of the hanging object can accurately judge whether the person is standing under the hanging object.

[0118] The recognition method provided by the present invention can identify whether a person is standing under a suspended object in a horizontal viewing angle based on image processing and coordinate conversion in a single monitoring device scenario. The present invention does not need to use recognition methods such as binocular cameras or laser radars, and is not limited by the working distance of the monitoring device (such as the limited working distance of binocular cameras is short). The present invention has the characteristics of low cost and can accurately identify whether a person is standing under a suspended object.

[0119] According to another aspect of the present invention, a method for identifying people standing under a suspended object is provided, which is used to identify whether there is a person standing under the suspended object, so as to avoid the risk of accidents to people during construction.

[0120] like Figure 8As shown, the recognition system 1 includes an image processing module 10 , a coordinate processing module 20 and an information recognition module 30 .

[0121] According to an exemplary embodiment, the image processing module 10 obtains target image information collected by a monitoring device.

[0122] For example, the monitoring device can be a monocular camera, which is installed at a preset monitoring position according to monitoring requirements and is approximately parallel to the ground. It is used to collect monitoring images at the lifting equipment. The monitoring device has a fixed installation position, monitoring angle and focal length, etc.

[0123] Optionally, the viewing angle between the monitoring device and the horizontal plane can be 0° to 15°. By setting the viewing angle between the monitoring device and the horizontal plane in this way, the identification of the person standing under the suspended object can be performed at a horizontal angle.

[0124] Exemplarily, the lifting equipment includes, but is not limited to, equipment such as cranes, slings, cranes and elevators that can lift objects, and the present invention is not limited to this.

[0125] Optionally, the image processing module 10 obtains initial image information collected by the monitoring device.

[0126] For example, the image processing module 10 is connected to the monocular camera for communication, obtains the RTSP video stream from the monocular camera, and decodes the RTSP video stream to obtain the initial image information at the lifting device.

[0127] Exemplarily, the initial image information may be color image information or grayscale image information.

[0128] The image processing module 10 performs parameter calibration on the initial image information.

[0129] The image processing module 10 performs distortion correction on the initial image information according to the parameters obtained by parameter calibration to obtain target image information.

[0130] For example, parameter calibration includes internal parameter calibration and external parameter calibration. For example, the image processing module 10 first performs internal parameter calibration based on a pre-selected calibration method (such as Zhang's calibration method), and then performs distortion correction on the color image information according to the obtained internal parameter matrix and distortion coefficient.

[0131] Then, the image processing module 10 performs external parameter calibration on the color image information after distortion correction, such as performing external parameter calibration based on ArUco Marker, to obtain the rotation matrix R and offset matrix T of the posture of the monitoring device (such as a camera).

[0132] According to an example embodiment, the image processing module 10 sets at least one algorithm recognition area according to target image information.

[0133] For example, the image processing module 10 responds to the user's operation instruction and sets one or more algorithm recognition areas based on the target image information according to the user's needs. The algorithm recognition area can be a user-defined area with monitoring and recognition significance, that is, an alarm prompt will be issued only when a person is recognized in the algorithm recognition area.

[0134] Exemplarily, the image processing module 10 may generate a polygonal area based on the coordinates of points within the algorithm recognition area.

[0135] According to an example embodiment, the coordinate processing module 20 performs image recognition, depth estimation and coordinate conversion on the algorithm identification area to obtain the three-dimensional point cloud coordinates of the human body key points, the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object corresponding to the human body key points, the ground contour points and the bottom contour points of the hanging object respectively.

[0136] For example, the coordinate processing module 20 can segment and identify the key points of the human body, the contour points of the ground, and the contour points of the bottom of the hanging object by performing human key point detection and image segmentation on the algorithm recognition area.

[0137] Exemplarily, the key points of the human body may be the key points at the foot area of ​​the human body (such as the ankle); the ground contour points may be the contour points of the target ground area pre-selected by the user according to needs; the contour points of the bottom of the hanging object may be the contour points of the plane presented by the bottom of the hanging object.

[0138] The coordinate processing module 20 can obtain the corresponding three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object through depth estimation processing and coordinate conversion processing.

[0139] Exemplarily, the three-dimensional point cloud coordinates are (x, y, z).

[0140] Optionally, the coordinate processing module 20 performs arbitrary key point detection and image segmentation on the algorithm recognition area to obtain human key points, ground contour points and bottom contour points of the hanging object.

[0141] For example, the coordinate processing module 20 determines the human body area image in the target image information based on image segmentation and recognition technology, and detects the human body key points based on the human body segmentation technology, and then obtains the human body key point coordinates corresponding to the human body key points through the human body key point information detection technology.

[0142] And the coordinate processing module 20 determines the ground contour points and the bottom contour points of the hanging object in the target image information based on the image segmentation and recognition technology, and then obtains the corresponding ground contour point coordinates and the bottom contour point coordinates of the hanging object.

[0143] Exemplarily, image segmentation and recognition technologies include but are not limited to SSD, YOLO, RetinaNet or RT-DETR, etc.; human body segmentation technologies include but are not limited to Deeplab, Mask2Former or SegFormer, etc., and the present invention is not limited to this.

[0144] For example, Figure 4 As shown, the ankle key points can be detected by human key point information detection technology. The human key point detection technology includes but is not limited to OpenPose, AlphaPose or HRNet, etc., and the present invention is not limited to this.

[0145] The coordinate processing module 20 obtains the depth values ​​of the key points of the human body, the ground contour points and the bottom contour points of the suspended object corresponding to the key points of the human body, the ground contour points and the bottom contour points of the suspended object based on the depth estimation.

[0146] For example, the coordinate processing module 20 performs monocular depth estimation based on the target image information, and can obtain the depth values ​​of all pixels in the target image information. Then, the coordinate processing module 20 determines the corresponding depth values ​​of the key points of the human body, the depth values ​​of the ground contour points, and the depth values ​​of the bottom contour points of the hanging object according to the coordinates of the key points of the human body, the coordinates of the ground contour points, and the coordinates of the bottom contour points of the hanging object.

[0147] The coordinate processing module 20 calculates the three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object according to the depth values ​​of the key points of the human body, the depth values ​​of the ground contour points and the depth values ​​of the bottom contour points of the hanging object.

[0148] For example, the coordinate processing module 20 converts the depth values ​​of human key points, ground contour points and bottom contour points of suspended objects into three-dimensional point cloud coordinates of human key points, ground contour points and bottom contour points of suspended objects.

[0149] Exemplarily, the calculation formula of the three-dimensional point cloud coordinates (x, y, z) can be:

[0150]

[0151] Among them, (u, ν) is the coordinate of the pixel point, z c is the depth value of the pixel (u, ν),

[0152] (u0, ν0) are the coordinates of the principal point; f x The length of the focal length in the x-axis direction is described in pixels; f y The length of the focal length in the y-axis direction is described in pixels; depth_scale is the scale factor corresponding to the depth map (the ratio of the value stored in the depth map to the actual depth).

[0153] According to an example embodiment, the information recognition module 30 determines whether a person is on the ground based on the three-dimensional point cloud coordinates of the key points of the human body and the three-dimensional point cloud coordinates of the ground contour points.

[0154] For example, the information recognition module 30 can determine whether the recognized person is on the ground based on the distance correlation between the three-dimensional point cloud coordinates of the key points of the human body and the three-dimensional point cloud coordinates of the ground contour points.

[0155] Optionally, the information recognition module 30 constructs a ground plane equation according to the three-dimensional point cloud coordinates of the ground contour points.

[0156] For example, the equation of the ground plane is A x +B y +C z +D=0.

[0157] Among them, A, B, C, and D are all coefficients.

[0158] Exemplarily, the fitting plane may be solved based on the RANSAC method, so that the coefficients A, B, C, and D may be obtained.

[0159] The information recognition module 30 determines a first distance from the three-dimensional point cloud coordinates of the key points of the human body to the ground plane equation.

[0160] The information recognition module 30 determines that the person is on the ground when the first distance is less than a first preset threshold.

[0161] For example, the information recognition module 30 calculates the distance (ie, the first distance) from the three-dimensional point cloud coordinates of the key points of the human body to the ground plane equation. When the first distance is less than a first preset threshold, it can be determined that the person is on the ground.

[0162] With such arrangement, the present invention can map the distance relationship between the actual key points of the human body and the actual ground through the distance from the three-dimensional point cloud coordinates of the key points of the human body to the ground plane equation, thereby determining whether there is a person on the ground.

[0163] The information recognition module 30 determines whether the hanging object is on the ground according to the three-dimensional point cloud coordinates of the bottom contour points of the hanging object and the three-dimensional point cloud coordinates of the ground contour points.

[0164] For example, the information recognition module 30 can determine whether the hanging object is placed in a placed state or in a hoisted state based on the distance correlation between the three-dimensional point cloud coordinates of the bottom contour points of the hanging object and the three-dimensional point cloud coordinates of the ground contour points.

[0165] Optionally, the information recognition module 30 constructs a ground plane equation according to the three-dimensional point cloud coordinates of the ground contour points.

[0166] The ground plane equation has been described in detail above and will not be repeated here.

[0167] The information recognition module 30 determines a second distance from the three-dimensional point cloud coordinates of the bottom contour point of the hanging object to the ground plane equation.

[0168] When the second distance is greater than a second preset threshold, the information recognition module 30 determines that the hanging object is not on the ground.

[0169] For example, the information identification module 30 calculates the distance (i.e., the second distance) from the three-dimensional point cloud coordinates of the bottom contour point of the suspended object to the ground plane equation. When the second distance is greater than the second preset threshold, it can be determined that the suspended object is not on the ground but is in a suspended state.

[0170] With such a configuration, the present invention can map the distance relationship between the actual bottom contour points of the hanging object and the actual ground through the distance from the three-dimensional point cloud coordinates of the bottom contour points of the hanging object to the ground plane equation, thereby determining whether the hanging object is in a placed state on the ground or in a hoisted state.

[0171] According to an exemplary embodiment, when it is determined that a person is on the ground and that a hanging object is not on the ground, the coordinate processing module 20 respectively determines the three-dimensional coordinates of the key points of the human body and the bottom contour points of the hanging object in the world coordinate system.

[0172] For example, the coordinate processing module 20 converts the two-dimensional coordinates of the human key points and the ground contour points in the pixel coordinate system into three-dimensional coordinates in the world coordinate system based on the depth values ​​of the human key points and the depth values ​​of the bottom contour points of the hanging objects.

[0173] Exemplarily, the conversion relationship between the two-dimensional coordinates of the pixel coordinate system and the three-dimensional coordinates of the world coordinate system may be:

[0174]

[0175] Among them, (u, ν) is the two-dimensional coordinate of the pixel point; (x w ,y w , z w ) is the three-dimensional coordinate of the pixel point; z c is the depth value of the pixel point (u, ν); (u0, ν0) is the coordinate of the principal point (the unit is also pixel); f is the focal length, f x The length of the focal length in the x-axis direction is described in pixels; f y is to use pixels to describe the length of the focal length in the y-axis direction; R is the rotation matrix of the camera posture (the matrix size is 3*3); T is the offset matrix of the camera posture (the matrix size is 3*1); the camera intrinsic parameter matrix size is 3*4; the camera extrinsic parameter matrix size is 4*4; d x is the width of one pixel in the X direction (in mm); d x is the width of one pixel in the Y direction of the pixel point (in mm);w Vertical ground.

[0176] According to an exemplary embodiment, the information recognition module 30 determines whether a person is standing under the hanging object based on the three-dimensional coordinates of the key points of the human body and the contour points of the bottom of the hanging object.

[0177] For example, when it is determined that a person is on the ground and that a suspended object is not on the ground, it is necessary to further determine whether the identified person is standing under the suspended object.

[0178] Optionally, the information recognition module 30 determines the three-dimensional coordinate range of the key points of the human body according to the three-dimensional coordinates of the key points of the human body.

[0179] The information recognition module 30 determines the three-dimensional coordinate range of the bottom contour point of the hanging object according to the three-dimensional coordinates of the bottom contour point of the hanging object.

[0180] When the three-dimensional coordinate range of the key points of the human body is within the three-dimensional coordinate range of the contour points at the bottom of the hanging object, the information recognition module 30 determines that the person is standing under the hanging object.

[0181] For example, the three-dimensional coordinate range of the key points of the human body is (x foot ,y foot ), the three-dimensional coordinate range of the bottom contour point of the hanging object is (x bottom ,y bottom ). w is vertical to the ground, if (x foot ,y foot ) is in (x bottom ,y bottom ), it means that the key point of the human body is below the bottom contour point of the suspended object in the vertical direction, that is, the person is standing under the suspended object.

[0182] With such a configuration, the present invention can further map the range relationship between the actual key points of the human body and the actual bottom contour points of the hanging object (i.e., determine whether the person is in the range below the hanging object) based on the relationship between the three-dimensional coordinate range of the key points of the human body and the three-dimensional coordinate range of the bottom contour points of the hanging object when judging that a person is on the ground and the hanging object is not on the ground, so as to determine whether the person is standing under the hanging object. In this way, when judging whether the person is standing under the hanging object, an alarm prompt is issued to avoid safety accidents.

[0183] The present invention performs image recognition, depth estimation and coordinate conversion on the algorithm recognition area to obtain the three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object corresponding to the key points of the human body, the ground contour points and the bottom contour points of the hanging object, and further judges whether the person is on the ground and whether the hanging object is on the ground according to the three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object. When it is determined that the person is on the ground and the hanging object is not on the ground at the same time, the three-dimensional coordinates of the key points of the human body and the bottom contour points of the hanging object can accurately judge whether the person is standing under the hanging object.

[0184] The recognition method provided by the present invention can identify whether a person is standing under a suspended object in a horizontal viewing angle based on image processing and coordinate conversion in a single monitoring device scenario. The present invention does not need to use recognition methods such as binocular cameras or laser radars, and is not limited by the working distance of the monitoring device (such as the limited working distance of binocular cameras is short). The present invention has the characteristics of low cost and can accurately identify whether a person is standing under a suspended object.

[0185] According to another aspect of the present invention, an electronic device is provided, which includes: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors can implement the identification method described above.

[0186] According to another aspect of the present invention, a non-volatile computer-readable storage medium is provided, wherein a computer program is stored on the storage medium, and when the computer program is executed by a processor, the identification method described above can be implemented.

[0187] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions of the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for identifying a person standing under a suspended object, characterized in that: include: Obtain target image information collected by monitoring equipment; Setting at least one algorithm recognition area according to the target image information; Performing image recognition, depth estimation and coordinate conversion on the algorithm recognition area to obtain the three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object corresponding to the key points of the human body, the ground contour points and the bottom contour points of the hanging object respectively; Determine whether there is a person on the ground according to the three-dimensional point cloud coordinates of the key points of the human body and the three-dimensional point cloud coordinates of the ground contour points; Determine whether the hanging object is on the ground according to the three-dimensional point cloud coordinates of the bottom contour points of the hanging object and the three-dimensional point cloud coordinates of the ground contour points; When it is determined that a person is on the ground and the hanging object is not on the ground, the three-dimensional coordinates of the key points of the human body and the contour points of the bottom of the hanging object in the world coordinate system are determined respectively; Whether a person is standing under the hanging object is determined based on the three-dimensional coordinates of the key points of the human body and the three-dimensional coordinates of the contour points at the bottom of the hanging object.

2. The identification method according to claim 1, characterized in that: The obtaining of target image information collected by the monitoring device includes: Acquiring initial image information collected by the monitoring device; Performing parameter calibration on the initial image information; The initial image information is subjected to distortion correction according to the parameters obtained by parameter calibration to obtain the target image information.

3. The identification method according to claim 1, characterized in that: The image recognition, depth estimation and coordinate conversion are performed on the algorithm recognition area to obtain the three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object corresponding to the key points of the human body, the ground contour points and the bottom contour points of the hanging object respectively, including: Performing human key point detection and image segmentation on the algorithm recognition area to obtain human key points, ground contour points and bottom contour points of hanging objects; Based on the depth estimation, the depth values ​​of the key points of the human body, the depth values ​​of the ground contour points and the depth values ​​of the bottom contour points of the hanging object corresponding to the key points of the human body, the ground contour points and the bottom contour points of the hanging object are obtained; The three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object are calculated according to the depth values ​​of the key points of the human body, the depth values ​​of the ground contour points and the depth values ​​of the bottom contour points of the hanging object.

4. The identification method according to claim 1, characterized in that: Determining whether there is a person on the ground according to the three-dimensional point cloud coordinates of the human body key points and the three-dimensional point cloud coordinates of the ground contour points comprises: Constructing a ground plane equation according to the three-dimensional point cloud coordinates of the ground contour points; Determine a first distance from the three-dimensional point cloud coordinates of the key point of the human body to the ground plane equation; When the first distance is less than a first preset threshold, it is determined that the person is on the ground; The step of determining whether the hanging object is on the ground according to the three-dimensional point cloud coordinates of the bottom contour points of the hanging object and the three-dimensional point cloud coordinates of the ground contour points comprises: Constructing a ground plane equation according to the three-dimensional point cloud coordinates of the ground contour points; Determine a second distance from the three-dimensional point cloud coordinates of the bottom contour point of the hanging object to the ground plane equation; When the second distance is greater than a second preset threshold, it is determined that the hanging object is not on the ground.

5. The identification method according to claim 1, characterized in that: The determining whether a person is standing under a hanging object according to the three-dimensional coordinates of the key points of the human body and the three-dimensional coordinates of the contour points at the bottom of the hanging object comprises: Determine the three-dimensional coordinate range of the key point of the human body according to the three-dimensional coordinate of the key point of the human body; Determine the three-dimensional coordinate range of the bottom contour point of the hanging object according to the three-dimensional coordinates of the bottom contour point of the hanging object; When the three-dimensional coordinate range of the key points of the human body is within the three-dimensional coordinate range of the bottom contour points of the hanging object, it is determined that the person is standing under the hanging object.

6. A system for identifying people standing under a suspended object, characterized in that: include: An image processing module, which obtains target image information collected by the monitoring device and sets at least one algorithm recognition area according to the target image information; A coordinate processing module performs image recognition, depth estimation and coordinate conversion on the algorithm recognition area to obtain the three-dimensional point cloud coordinates of the key points of the human body, the three-dimensional point cloud coordinates of the ground contour points and the three-dimensional point cloud coordinates of the bottom contour points of the hanging object corresponding to the key points of the human body, the ground contour points and the bottom contour points of the hanging object respectively; An information recognition module determines whether a person is on the ground according to the three-dimensional point cloud coordinates of the key points of the human body and the three-dimensional point cloud coordinates of the ground contour points, and determines whether the hanging object is on the ground according to the three-dimensional point cloud coordinates of the bottom contour points of the hanging object and the three-dimensional point cloud coordinates of the ground contour points; The coordinate processing module determines the three-dimensional coordinates of the key points of the human body and the bottom contour points of the hanging object in the world coordinate system respectively when it is determined that a person is on the ground and the hanging object is not on the ground; The information recognition module determines whether a person is standing under the hanging object based on the three-dimensional coordinates of the key points of the human body and the three-dimensional coordinates of the contour points at the bottom of the hanging object.

7. The identification system according to claim 6, characterized in that: The coordinate processing module performs human key point detection and image segmentation on the algorithm recognition area to obtain human key points, ground contour points and bottom contour points of hanging objects; The coordinate processing module obtains the depth values ​​of the key points of the human body, the depth values ​​of the ground contour points and the depth values ​​of the bottom contour points of the hanging object corresponding to the key points of the human body, the ground contour points and the bottom contour points of the hanging object based on depth estimation; The coordinate processing module calculates the three-dimensional point cloud coordinates of the human body key points, the ground contour points and the bottom contour points of the hanging object according to the depth values ​​of the human body key points, the ground contour points and the bottom contour points of the hanging object.

8. The identification system according to claim 6, characterized in that: The information recognition module constructs a ground plane equation according to the three-dimensional point cloud coordinates of the ground contour points; The information recognition module determines a first distance from the three-dimensional point cloud coordinates of the key point of the human body to the ground plane equation; The information recognition module determines that the person is on the ground when the first distance is less than a first preset threshold; as well as The information recognition module constructs a ground plane equation according to the three-dimensional point cloud coordinates of the ground contour points; The information recognition module determines a second distance from the three-dimensional point cloud coordinates of the bottom contour point of the hanging object to the ground plane equation; The information recognition module determines that the hanging object is not on the ground when the second distance is greater than a second preset threshold.

9. The identification system according to claim 6, characterized in that: The information recognition module determines the three-dimensional coordinate range of the key point of the human body according to the three-dimensional coordinate of the key point of the human body; The information recognition module determines the three-dimensional coordinate range of the bottom contour point of the hanging object according to the three-dimensional coordinates of the bottom contour point of the hanging object; The information recognition module determines that a person is standing under the suspended object when the three-dimensional coordinate range of the key point of the human body is within the three-dimensional coordinate range of the bottom contour point of the suspended object.

10. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the identification method as described in any one of claims 1-5.

Images