A dangerous prompt method based on a hanger cone, a program product, an electronic device and a robot

Abstract

The application discloses a kind of dangerous prompt method based on sling cone, storage medium, program product, electronic equipment and robot, the scheme is outputted dangerous prompt when detecting that pedestrian enters target area;Target area is determined based on the coordinate position and scale value of sling cone in target image, and target image includes the sling cone of hanging suspension object and pedestrian;Scale value is determined based on at least the size of suspension object in target image.To solve the problem of false alarm and miss report in related technology, improve the safety of hoisting operation.

Description

Technical Field

[0001] This invention relates to the field of production safety technology, and in particular to a hazard warning method, storage medium, program product, electronic device and robot based on a suspended cone. Background Technology

[0002] In complex environments such as large warehouses, logistics centers, and port terminals, hoisting operations are frequent, and there are potential safety risks between the operators and the hoisted objects.

[0003] Related technologies rely on the angle formed by the line connecting the center point of the person and the center point of the suspended object with the vertical direction to determine whether a person is standing under the suspended object, thus providing a hazard warning. However, this method can only identify hazards posed by pedestrians moving horizontally and cannot identify hazards in the vertical direction, leading to the risk of false alarms and missed alarms. Summary of the Invention

[0004] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention proposes a hazard warning method, storage medium, program product, electronic device, and robot based on a suspended cone. This solution outputs a hazard warning when a pedestrian is detected entering a target area. The target area is determined based on the coordinate position and scale of the suspended cone in a target image, which includes the suspended cone and the pedestrian. The scale is determined at least based on the size of the suspended object in the target image. Thus, because this solution outputs hazard warnings based on personnel entering the target area, it can identify hazards in all directions and provide warnings, thereby solving the problems of false alarms and missed alarms in hazard identification in related technologies and improving the safety of hoisting operations.

[0005] To achieve the above objectives, this application adopts the following technical solution:

[0006] In a first aspect, this application provides a hazard warning method based on a suspended cone, comprising: outputting a hazard warning when a pedestrian is detected entering a target area; the target area is determined based on the coordinate position and scale value of the suspended cone in a target image, the target image including the suspended cone and the pedestrian; the scale value is determined at least based on the size of the suspended object in the target image.

[0007] Thus, since this solution outputs hazard warnings based on personnel entering the target area, it can identify hazards in all directions and provide warnings, thereby solving the problems of false alarms and missed alarms in hazard identification in related technologies and improving the safety of hoisting operations.

[0008] In some embodiments of this application, a safety alert is also provided when a pedestrian is detected not to have entered or left the target area.

[0009] In some embodiments of this application, the method further includes: when there are at least two hoisting cones in the hoisting scenario, assigning a unique tracking ID to each hoisting cone; uniquely binding each hoisting cone to its corresponding target area through the tracking ID, and updating the position of the corresponding target area based on the real-time coordinate position of the hoisting cone.

[0010] In some embodiments of this application, the method further includes: determining the target area based on the coordinate position information and scale value of the cone in the target image.

[0011] In some embodiments of this application, determining the target area based on the coordinate position information and scale value of the suspended cone in the target image includes: calculating the height of the suspended cone based on the coordinate position information of the suspended cone; and determining the target area based on the height of the suspended cone and the scale value.

[0012] In some embodiments of this application, the ratio value includes a first ratio value and a second ratio value. Determining the target area based on the height of the suspension cone and the ratio value includes: determining the size of the target area based on the first ratio value and the height of the suspension cone, and determining the center coordinate point of the target area based on the second ratio value and the height of the suspension cone; the X-coordinate value of the center point of the target area is equal to the X-coordinate value of the center point of the suspension cone, and the Y-coordinate value is determined based on the Y-coordinate value of the center point of the suspension cone and an increment value, wherein the increment value is determined according to the height of the suspension cone and the second ratio value; and generating the target area based on the size of the target area and the coordinates of the center point of the target area.

[0013] In some embodiments of this application, the target image is obtained by taking a top-down shot with a camera. The method further includes: when the suspended cone moves in space, binding the distance between the center point of the suspended cone and the center point of the target area and the size of the target area to the height of the suspended cone, wherein the position of the target area is dynamically changed based on the height of the suspended cone in the image.

[0014] In some embodiments of this application, the method further includes: correcting the position of the target region below the suspension cone based on a vertical offset, the vertical offset being determined at least based on the height of the suspension cone.

[0015] In some embodiments of this application, correcting the position of the target area below the suspended cone based on the vertical offset includes: if the vertical distance between the center of the suspended cone and the target area is less than or equal to a first threshold, the system increases the vertical offset so that the target area is away from the suspended object on the suspended cone on the display interface; and / or, if the vertical distance between the center of the suspended cone and the target area is greater than a second threshold, the system decreases the vertical offset so that the target area remains within a set area below the suspended object on the display interface.

[0016] In some embodiments of this application, the method further includes: confirming that the pedestrian has entered the target area when any part of the pedestrian's body is in the target area; and / or confirming that the pedestrian has not entered the target area when none of the pedestrian's body parts are in the target area.

[0017] In some embodiments of this application, the method further includes: displaying at least one of a detection box, a danger warning box, and a safety warning box for the target area.

[0018] Secondly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, causes the computer to perform the method described in the first aspect.

[0019] Thirdly, this application provides a computer program product that stores instructions which, when executed by a computer, cause the computer to perform the method described in the first aspect.

[0020] Fourthly, this application provides an electronic device, comprising: a memory having a computer program stored thereon; and a processor for executing the computer program in the memory to implement the method as described in the first aspect.

[0021] Fifthly, this application provides a robot, including: an electronic device as described in the fourth aspect;

[0022] Alternatively, a processor, the processor being configured to perform the method as described in the first aspect.

[0023] Sixthly, this application provides a hoisting system, including: a robot as described in the fifth aspect and a slide rail, wherein the robot is disposed within the slide rail and is movable within the slide rail.

[0024] The advantages and control methods of the robot, the hoisting system, and the electronic equipment compared to existing technologies are the same and will not be elaborated here.

[0025] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] To gain a more complete understanding of this application and its beneficial effects, the following description will be provided in conjunction with the accompanying drawings, wherein the same reference numerals in the following description denote the same parts.

[0028] Figure 1 This is a flowchart illustrating a hazard warning method based on a hanging cone according to an embodiment of the present invention;

[0029] Figure 2 This is a scenario illustration provided according to an embodiment of the present invention. Figure 1 ;

[0030] Figure 3 This is a scenario illustration provided according to an embodiment of the present invention. Figure 2 ;

[0031] Figure 4 This is a flowchart illustrating another tipping method based on a hanging cone provided according to an embodiment of the present invention;

[0032] Figure 5 This is a schematic diagram of the ROI hazardous area and 3D dynamic prediction process provided by an embodiment of the present invention;

[0033] Figure 6 This is a scenario illustration provided according to an embodiment of the present invention. Figure 3 ;

[0034] Figure 7 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present invention. Detailed Implementation

[0035] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the protection scope of this application.

[0036] In complex environments such as large warehouses, logistics centers, and port terminals, hoisting operations are frequent, and there are potential safety risks between the operators and the hoisted objects.

[0037] Related technologies rely on the angle formed by the line connecting the center point of the person and the center point of the suspended object with the vertical direction to determine whether a person is standing under the suspended object, thus providing a hazard warning. However, this method can only identify hazards posed by pedestrians moving horizontally and cannot identify hazards in the vertical direction, leading to the risk of false alarms and missed alarms.

[0038] To address the aforementioned technical issues, this solution outputs a hazard warning when a pedestrian is detected entering a target area. The target area is determined based on the coordinates and scale of the suspended object in a target image, which includes both the suspended object and the pedestrian. The scale is determined at least based on the size of the suspended object in the target image. Thus, because this solution outputs hazard warnings based on personnel entering the target area, it can identify hazards in all directions and provide warnings, thereby resolving the issues of false alarms and missed alarms in hazard identification in related technologies and improving the safety of hoisting operations.

[0039] The present application solution will be further described below with reference to the embodiments.

[0040] like Figure 1 The diagram shown is a flowchart of a warning method based on a hanging cone provided in an embodiment of the present invention, including: 200, acquiring a target image through a monocular camera; 201, outputting a danger warning when a pedestrian is detected entering the target area; the target area is determined based on the coordinate position and scale value of the hanging cone in the target image.

[0041] The target image mentioned above includes a suspended object and a pedestrian, and the target image is obtained by taking a top-down shot with a camera; the scale value is determined based at least on the size of the suspended object in the target image.

[0042] Thus, since this solution outputs hazard warnings based on personnel entering the target area, it can identify hazards in all directions and provide warnings, thereby solving the problems of false alarms and missed alarms in hazard identification in related technologies and improving the safety of hoisting operations.

[0043] For example, the target area mentioned above can also be referred to as the ROI danger zone.

[0044] Optionally, the method further includes: displaying at least one of a detection box for the target area, a hazard warning, and a safety warning.

[0045] For example, step 200 above may include: deploying a high-resolution monocular camera (e.g., 1920×1080 resolution, 30fps) above the hoisting operation area to ensure coverage of the entire operation range; and eliminating lens distortion through a camera calibration module to ensure the mapping accuracy between image coordinates and real spatial coordinates.

[0046] For example, the methods of outputting hazard warnings involved in step 201 above include, but are not limited to, audible and visual alarms, and interface highlighting. For instance, audible and visual alarms can be used to alert on-site workers, and interface highlighting can be used to alert monitoring personnel.

[0047] Further optionally, the above method also includes: 202. Outputting a safety alert when personnel leave the target area.

[0048] like Figure 2 The image shown is a schematic diagram of a scenario provided by an embodiment of the present invention. Figure 1 .

[0049] As shown in the diagram, the hoisting cone moves up and down, stopping at 3 meters, 2 meters, and 1 meter above the ground. A person enters and leaves the danger zone at these different stops. The system displays a warning message at each hoisting cone position, changing from "Safe" to "Danger" and then back to "Safe" upon leaving. This further verifies the boundary between the safe and dangerous zones, marking the boundary boxes and comparing stability and error magnitude. This solution dynamically adjusts the size and position of the danger zone based on the hoisting cone's current height, ensuring the continuity and stability of the danger zone.

[0050] Further optionally, the method also includes: when there are at least two lifting cones in the lifting scenario, assigning a unique tracking ID to each lifting cone; uniquely binding each lifting cone to its corresponding target area through the tracking ID, and updating the position of the corresponding target area based on the real-time coordinate position of the lifting cone.

[0051] In some embodiments, the above-mentioned multi-target tracking can be achieved through the following steps: Step S11: Assign a unique ID (ID_1, ID_2) to each crane cone using the DeepSort algorithm; Step S12: Establish a mapping table between ID and target area, and when the crane cone moves, the center point of the target area is synchronously updated to (x_1', y_1'+Δy).

[0052] This invention achieves the tracking of multiple pendants through multi-target tracking. The DeepSort multi-target tracking algorithm assigns a unique tracking ID to each pendant, enabling continuous tracking across frames. This algorithm supports scenarios where multiple pendants exist simultaneously. By recording and identifying the spatial trajectory of each pendant through the tracking module, it effectively avoids confusion between multiple pendant targets.

[0053] like Figure 3 The image shown is a schematic diagram of a scenario provided by an embodiment of the present invention. Figure 2 The diagram illustrates a hoisting scenario where multiple hoisting cones exist simultaneously, and the system can output an independent danger zone determination for each hoisting cone.

[0054] The following section will describe in detail how this solution performs hazard and safety identification based on multiple target tracking methods, specifically including the following four implementation approaches:

[0055] Method 1: As an optional implementation, the above method further includes: outputting a first danger warning when the first user enters the first target area; the first target area is determined based on the position information of the first hoist cone; and outputting a second danger warning when the second user enters the second target area; the second target area is determined based on the position information of the second hoist cone.

[0056] In some embodiments, method one may include the following:

[0057] Step S231: Assign a unique tracking ID to each crane cone using the DeepSort algorithm;

[0058] Step S232: Generate independent ROI regions ROI_A and ROI_B for crane A and crane B respectively;

[0059] Step S233: When the first user enters ROI_A, output the first danger warning; when the second user enters ROI_B, output the second danger warning.

[0060] Method 2: As an optional implementation, the above method further includes: outputting a first safety prompt when the first user does not enter the first target area; the first target area is determined based on the position information of the first hoist cone; and outputting a second safety prompt when the second user does not enter the second target area; the second target area is determined based on the position information of the second hoist cone.

[0061] In some embodiments, method two may include:

[0062] Step S241: When the first user has not entered ROI_A, output the first security prompt;

[0063] Step S242: When the second user does not enter ROI_B, output the second security prompt.

[0064] Method 3: As an optional implementation, the above method further includes: outputting a first danger warning when the first user enters the first target area; the first target area is determined based on the position information of the first hoist cone; and outputting a second safety warning when the second user does not enter the second target area; the second target area is determined based on the position information of the second hoist cone.

[0065] In some embodiments, method three may include the following:

[0066] Step S251: When the first user enters ROI_A, output the first danger warning;

[0067] Step S252: When the second user does not enter ROI_B, output the second security prompt.

[0068] Method 4: As an optional implementation, the above method further includes: outputting a first safety warning when the first user does not enter the first target area; the first target area is determined based on the position information of the first hoist cone; and outputting a second danger warning when the second user enters the second target area; the second target area is determined based on the position information of the second hoist cone.

[0069] In some embodiments, method four may include the following:

[0070] Step S261: When the first user has not entered ROI_A, output the first security prompt;

[0071] Step S262: When the second user enters ROI_B, output the second danger warning.

[0072] In the above embodiment, the 2D bounding box of the crane cone is detected by the YOLOv11 model, and the depth information of the crane cone in the Z-axis direction is calculated by the monocular vision depth estimation algorithm. Combined with the physical size ratio of the crane cone, its actual position in 3D space is determined, and the danger zone ROI is dynamically generated.

[0073] As an optional implementation, the above method further includes: 200, determining the target area based on the coordinate position information and scale value of the cone in the target image.

[0074] For example, step 200 above includes the following: 200a1, calculating the height of the suspension cone based on the coordinate position information of the suspension cone; 200a2, determining the target area based on the height of the suspension cone and the ratio value.

[0075] Further optionally, the aforementioned ratio value includes a first ratio value and a second ratio value, and step 200a2 specifically includes the following:

[0076] A1. Determine the size of the target area based on the first proportional value and the height of the suspension cone, and determine the center coordinate point of the target area based on the second proportional value and the height of the suspension cone.

[0077] A2. Generate the target area based on the size of the target area and the coordinates of the center point of the target area.

[0078] The first ratio value mentioned above is determined based on the size of the suspended object in the target image, and the user can set it according to actual needs; the second ratio value mentioned above is a calibration value, which can be obtained by calibration based on the height h1 of the suspended cone and the distance H1 between the suspended cone and the ROI region in the image dataset. For example, the second ratio value is: α = H1 / h1.

[0079] Alternatively, in step A1 above, the X-coordinate value of the center point of the target area is equal to the X-coordinate value of the center point of the suspension cone, and the Y-coordinate value is determined based on the Y-coordinate value and the increment value of the center point of the suspension cone.

[0080] For example, the above-mentioned increment value is the distance from the center point of the suspension cone to the center point of the target area, and this increment value is determined based on the height of the suspension cone on the image and the second ratio value.

[0081] For example, establish an image coordinate system xy with the top left corner of the image as the origin. The corresponding coordinate points on the diagonal of the suspended cone are A[x1,y1] and B[x2,y2], respectively. The coordinates of the corresponding center point O1 of the suspended cone are [1 / 2(x1+x2),1 / 2(y1+y2)], and the height of the suspended cone is hoist. Height =y2-y1, the second ratio is α=H1 / h1, based on the second ratio Rh and the height of the suspension cone hoist Height The coordinates of the center point O2 of the determined target area are [1 / 2(x1+x2), 1 / 2(y1+y2)+hoist] Height *α].

[0082] When the first ratio value mentioned above includes: b and c, the height of the suspension cone is hoist Height， The corresponding target region's height and width are ROI. width =b*hoist Height ROI height =c*hoist Height This allows us to determine the size of the target area.

[0083] Further optionally, the above method also includes: when the suspension cone moves in space, binding the distance between the center point of the suspension cone and the center point of the target area and the size of the target area to the height of the suspension cone, wherein the position of the target area is dynamically changed based on the height of the suspension cone in the image.

[0084] Specifically, the danger zone of the hoisted object is not fixed, but dynamically generated based on the movement trend of the hoist cone in space. The algorithm uses the height of the hoist cone as a scale to estimate the depth in space, and binds the distance between the center point of the hoist cone and the center point of the ROI and the size of the ROI to the height of the hoist cone in a certain proportion. The ROI window can slide on the ground according to the distance of the hoist cone in space.

[0085] Further optionally, the above method further includes: 203, correcting the position of the target area below the pylon based on the vertical offset, wherein the vertical offset is determined at least based on the height of the pylon.

[0086] For example, step 203 above specifically includes the following:

[0087] 203a. If the vertical distance (Δy) between the center of the sling and the target area is less than or equal to the first threshold, the system increases the vertical offset so that the target area is away from the suspended object on the sling on the display interface; and / or, 203b. If the vertical distance (Δy) between the center of the sling and the target area is greater than the second threshold, the system decreases the vertical offset so that the target area is kept within the set area below the suspended object on the display interface.

[0088] Specifically, the aforementioned vertical offset can be calculated using the following formula: Δy = hoistHeight * α + δ, where: α is a static estimate used to prevent jitter in the detection factor when calculating Δy; δ is the correction amount obtained by interpolation iteration based on the vertical offset difference between the target center in the current frame and the previous frame, i.e., the Δy of the previous frame; hoistHeight is the height of the hoisting cone, which can be determined based on the coordinates of the hoisting cone. The addition of this correction amount ensures that the ROI region can be "locked" in a relatively stable position below the suspended object, so that even if the suspended object moves slightly up and down, the ROI region remains stable.

[0089] To ensure that the ROI area is neither too close to the hoisted object nor too far from its actual drop range, the system sets a two-way safety limit on the offset Δy: if the vertical distance between the hoisting center and the ROI center is less than the set minimum safety clearance (based on the hoisting cone height hoist), the ROI area will be subject to a minimum safety clearance. height High ROI height If the decision is made jointly, the system will automatically increase Δy, at which point Δy = d. min Keep the ROI away from the hoisting object to create a necessary safety buffer zone; if this distance exceeds the set maximum safety clearance (to prevent the ROI from drifting too far), the system will automatically shrink Δy, at which point Δy = d. max Keep the ROI within a reasonable area below the hoisting object.

[0090] Among them, the minimum safe distance (i.e., the lower limit of the buffer zone) d mentioned above min =(ROI) height / 2+hoist height / 2)*mintimes, where min times is a set value representing a multiple, which can be set by the user as needed; maximum safety distance (i.e., the upper limit for anti-drift) d max =hoist height *α, where α is the static proportional calculation value.

[0091] Optionally, the above methods also include: 200A, determining whether personnel have entered the target area.

[0092] For example, step 200A above specifically includes the following: when any part of a person's body is within the target area, it is confirmed that the person has entered the target area; when all parts of a person's body are not within the target area, it is confirmed that the person has not entered the target area.

[0093] Specifically, step 200A above includes the following:

[0094] Step S311: Track the movement of the worker in real time using a key point detection model or a target detection model; Step S312: Calculate the coordinates of the key points on the worker's body; Step S313: When the coordinates of any key point are within the danger zone (ROI), determine that the worker has entered the target area; Step S314: When the coordinates of all key points are not within the danger zone, determine that the worker has not entered the target area.

[0095] In the above embodiments, by accurately tracking the key points of the workers, the misjudgment problem based on the geometric center in traditional methods is avoided, especially when the workers are tilted or partially enter the danger zone, the judgment can still be made accurately.

[0096] Preferably, this solution can use the bottom key points of the personnel (left bottom corner, right bottom corner, center bottom point) to determine whether the personnel have entered the target area by spatial inclusion judgment.

[0097] The following will describe the entire implementation process of this invention in detail with reference to practical applications. This invention proposes the aforementioned alerting method, deployed on a suspended rail inspection robot platform. Based on images acquired from different angles, it can dynamically identify the crane cone and workers, and determine whether personnel are in a danger zone based on real-time prediction results, thereby improving the safety of lifting operations. For example... Figure 4 As shown, the present invention includes the following steps:

[0098] 101. Image Acquisition.

[0099] The RGB camera mounted on the rail-mounted inspection robot acquires video frame images of the hoisting operation area.

[0100] 102. Target detection.

[0101] Two YOLO models are loaded respectively to perform real-time detection of the crane cone and the workers, and output the coordinates of the rectangular bounding box.

[0102] 103. Target Tracking (Multi-Cone Tracking):

[0103] The DeepSort multi-target tracking algorithm assigns a unique tracking ID to each pendant, enabling continuous tracking across frames. This algorithm supports scenarios where multiple pendants exist simultaneously. By recording and identifying the spatial trajectory of each pendant through the tracking module, it effectively avoids confusion between multiple pendant targets.

[0104] Specifically, for each confirmed hoisting target, the system records its position, center point, and vertical correction displacement relative to the previous frame in the current frame. This correction displacement is calculated based on the interpolation of the difference between the center points of the preceding and following frames, and scaled by combining the target height to obtain a more robust center displacement trend.

[0105] 104. Prediction and correction of dangerous areas.

[0106] like Figure 5 The diagram shown is a schematic diagram of the ROI hazardous area and 3D dynamic prediction process provided by the present invention.

[0107] The danger zone of a hoisted object is not fixed, but dynamically generated based on its spatial height and movement trend. For each hoisted target, the system calculates the standard aspect ratio of the danger zone based on its height, and determines the initial center offset ratio α by combining empirical parameters.

[0108] Let the coordinates of the hoisting cone be [x1, y1, x2, y2], the center point be hoistCenter = ((x1+x2) / 2, (y1+y2) / 2), and the height be hoistHeight = y2-y1. Accordingly, the ROI width and height calculation formulas are: ROIwidth = hoistHeight * rw, ROIheight = hoistHeight * rh; where rw and rh are the ratios in the width and height directions, which can be set according to user needs. The vertical offset of the center point is: Δy = hoistHeight * α + δ, where α is a static estimate used to prevent jitter in the detection factor when calculating Δy. δ is the correction amount obtained by interpolation iteration based on the vertical offset difference between the target center in the current frame and the previous frame, i.e., the Δy of the previous frame. The addition of this correction amount allows the ROI region to be "locked" in a relatively stable position below the hoisted object, so that even if the hoisted object moves slightly up and down, the ROI region remains stable.

[0109] To ensure that the danger zone (ROI) is neither too close to the hoisted object nor too far from its actual drop range, the system sets a two-way safety limit on the offset Δy.

[0110] If the vertical distance between the hoisting center and the ROI center is less than the set minimum safety clearance (determined by both the hoisting height and the ROI height), the system will automatically increase Δy, at which point Δy = d.min Keep the ROI away from the hoisting object to create a necessary safety buffer zone.

[0111] If this distance exceeds the set maximum safety margin (to prevent ROI drift from becoming too far), the system will automatically shrink Δy, at which point Δy = d. max Keep the ROI within a reasonable area below the hoisting object.

[0112] Finally, output the ROI result for the current frame, including the ROI box coordinates [x1, y1, x2, y2] and the correction amount δ.

[0113] Through this mechanism, the system ensures that the ROI area stably and accurately covers the potentially dangerous space below the hoisted object, thus making safety risks controllable during dynamic hoisting.

[0114] 105. Hazard identification (relationship between workers and ROI).

[0115] like Figure 6 The image shown is a schematic diagram of a scenario provided by an embodiment of the present invention. Figure 3 .

[0116] Using geometric relationships, spatial inclusion determination is performed on the bottom key points (left bottom corner, right bottom corner, center bottom point) of each worker;

[0117] If the bottom left or right corners fall into the ROI area, it is judged as "danger" (personnel are in the danger zone); otherwise, it is "safety".

[0118] 106. Results Output and Visualization.

[0119] All detection boxes and hazard determination results are output in a standardized LabelMe format. At the same time, borders, connecting lines and text information are drawn in the image for intuitive display.

[0120] The solution provided by the embodiments of the present invention has the following advantages:

[0121] (1) DeepSort supports simultaneous tracking of multiple pendant targets and independent ROI generation, which is suitable for multi-target scenarios;

[0122] (2) Introducing an inter-frame interpolation correction mechanism enables the ROI region to adaptively lock below the hoisting, improving the stability of hazard zone identification;

[0123] (3) Spatial geometric discrimination method accurately identifies the risk level of workers;

[0124] (4) Multi-target tracking combined with interpolation correction enhances cross-frame continuity and adapts to dynamic scene changes;

[0125] (5) Deployed on a rail-mounted robot platform, effectively avoiding ground obstruction and enabling safe inspection with a wider field of view.

[0126] Figure 7 This is a schematic diagram of the structure of an electronic device according to an embodiment of this specification. The electronic device 700 includes a processor 701 with one or more processing cores, a memory 702 with one or more computer-readable storage media, and a computer program stored on the memory 702 and executable on the processor. The processor 701 and the memory 702 are electrically connected.

[0127] The processor 701 is the control center of the electronic device 700. It connects various parts of the electronic device 700 via various interfaces and lines. By running or loading software programs and / or units stored in the memory 702, and by calling data stored in the memory 702, it executes various functions and processes data of the electronic device 700, thereby providing overall monitoring of the electronic device 700. The processor 701 can be a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a Network Processor (NP), etc., and can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0128] In this embodiment of the application, the processor 701 in the electronic device 700 loads the computer program corresponding to the process of one or more applications into the memory 702 according to the method or steps of the above embodiment, and the processor 701 runs the applications stored in the memory 702 to execute the above method.

[0129] According to an embodiment of the present invention, the electronic device outputs a hazard warning when personnel enter a target area by executing the above-described method; the target area is identified based on the position information of the hoisting cone. Thus, because this solution outputs a hazard warning based on personnel entering the target area, it is possible to identify hazards present in various directions and provide warnings, thereby solving the problems of false alarms and missed alarms in hazard identification in related technologies and improving the safety of hoisting operations.

[0130] Embodiments of the present invention also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, enables the computer to perform the methods described above. For example, the computer-readable storage medium may be the aforementioned memory including program instructions, which may be executed by a processor of an electronic device to implement or execute the methods, steps, and logic diagrams disclosed in the embodiments of this application.

[0131] This invention also provides a computer program product storing instructions that, when executed by a computer, cause the computer to perform the methods described above. For example, when executed by a computer, the instructions implement or execute the methods, steps, and logic diagrams disclosed in the embodiments of this application.

[0132] Embodiments of the present invention also provide a robot that includes the system described above, or an electronic device, or a processor, the processor being used to perform the methods described above.

[0133] The robot according to an embodiment of the present invention includes the aforementioned electronic device or processor. The processor is used to execute the aforementioned method by outputting a hazard warning when personnel enter a target area; the target area is identified based on the position information of the hoisting cone. Thus, because this solution outputs a hazard warning based on personnel entering a target area, it is possible to identify hazards present in various directions and provide warnings, thereby solving the problems of false alarms and missed alarms in hazard identification in related technologies and improving the safety of hoisting operations.

[0134] The above-described embodiments are only used to illustrate the technical solutions of applying the above methods to robots, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the method can also be used in other intelligent devices, etc., without causing the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0135] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0136] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," "optional example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0137] The embodiments, implementation methods, and related technical features of this application can be combined and substituted for each other without conflict.

[0138] The above are merely preferred embodiments of this application and are not intended to limit this application in any way. Although the descriptions of each embodiment in this application have different focuses, and the parts not described in detail in a certain embodiment can be referred to the relevant embodiments of other embodiments, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of this application without departing from the content of the technical solution of this application shall still fall within the scope of the technical solution of this application.

Claims

1. A hazard warning method based on a hanging cone, characterized in that, include: When a pedestrian is detected entering the target area, a danger warning is output; The target area is determined based on the coordinate position and scale of the suspended cone in the target image, which includes the suspended cone and pedestrians. The scale value is determined based at least on the size of the suspended object in the target image.

2. The method according to claim 1, characterized in that, The method further includes: When there are at least two hoisting cones in a hoisting scenario, assign a unique tracking ID to each hoisting cone; Each of the suspended cones is uniquely bound to its corresponding target area using the tracking ID, and the position of the corresponding target area is updated based on the real-time coordinates of the suspended cones.

3. The method according to claim 1, characterized in that, Also includes: The height of the suspension cone is calculated based on its coordinate position information; The target area is determined based on the height and proportion of the suspended cone.

4. The method according to claim 3, characterized in that, The ratio value includes a first ratio value and a second ratio value. Determining the target area based on the height of the suspension cone and the ratio value includes: The size of the target area is determined based on a first proportional value and the height of the suspension cone, and the center coordinate point of the target area is determined based on a second proportional value and the height of the suspension cone; the X coordinate value of the center point of the target area is equal to the X coordinate value of the center point of the suspension cone, and the Y coordinate value is determined based on the Y coordinate value of the center point of the suspension cone and an incremental value, wherein the incremental value is determined according to the height of the suspension cone and the second proportional value; The target region is generated based on the size of the target region and the coordinates of the center point of the target region.

5. The method according to claim 1, characterized in that, The target image is obtained by taking a top-down view with a camera, and the method further includes: When the suspended cone moves in space, the distance between the center point of the suspended cone and the center point of the target area, and the size of the target area, are linked to the height of the suspended cone. The position of the target area is based on the dynamic changes in the height of the suspended cone in the image; and / or, The position of the target area below the suspended cone is corrected based on a vertical offset, which is determined at least based on the height of the suspended cone.

6. The method according to claim 5, characterized in that, The step of correcting the position of the target area below the suspension cone based on the vertical offset includes: If the vertical distance between the center of the suspension cone and the target area is less than or equal to a first threshold, the system increases the vertical offset so that the target area on the display interface is moved away from the suspended object on the suspension cone; and / or, If the vertical distance between the center of the suspension cone and the target area is greater than the second threshold, the system reduces the vertical offset so that the target area remains within a set area below the suspended object on the display interface.

7. The method according to claim 1, characterized in that, The method further includes: When any part of a pedestrian's body is within the target area, the pedestrian is confirmed to have entered the target area; and / or, If all parts of a pedestrian's body are not within the target area, then it is confirmed that the pedestrian has not entered the target area.

8. An electronic device, characterized in that, include: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the method of any one of claims 1 to 7.

9. A robot, characterized in that, include: The electronic device according to claim 8; Alternatively, a processor, said processor being configured to perform the method of any one of claims 1-7.

10. A hoisting system, comprising: The robot and slide rail of claim 9, wherein the robot is disposed within the slide rail and is movable within the slide rail.

Images