High-altitude operation monitoring method and system

Abstract

A high-altitude operation monitoring method and system. The monitoring method comprises: capturing an operation region to obtain an image to be subjected to determination; then, analyzing said image to find an operator image block corresponding to an operator and find a safety rope image block group corresponding to a safety rope group; next, on the basis of the distance between the operator image block and the safety rope image block group, determining whether the operator uses the safety rope group, and generating a corresponding determination result; and finally, on the basis of the determination result, generating prompt information.

Description

High-altitude operation monitoring methods and systems Technical Field

[0001] This invention relates to a high-altitude work monitoring technology, and more particularly to a high-altitude work monitoring method and system for confirming whether high-altitude workers are properly using safety ropes. Background Technology

[0002] According to legal regulations, workers must wear safety helmets, safety belts, and other necessary protective gear when working at heights above 2 meters, or they must install safety nets or other protective measures in the workplace. However, in the current work environment, it still relies on human supervision to confirm whether these protective measures have been properly implemented. If these protective measures are not properly implemented and are not detected in time, the personal safety of workers at heights cannot be fully guaranteed.

[0003] Summary of the Invention

[0004] Therefore, one of the objectives of this invention is to provide a method and system for monitoring high-altitude operations to automatically determine whether workers are properly using safety ropes.

[0005] From one perspective, this invention proposes a method for monitoring high-altitude operations, comprising: capturing images of the work area to obtain an image to be judged; analyzing the image to be judged to find the personnel image block corresponding to the worker; analyzing the image to be judged to find the safety rope image block group corresponding to the safety rope group; generating a corresponding first judgment result based on the distance between the personnel image block and the safety rope image block group; and generating a first prompt message based on the first judgment result.

[0006] In one embodiment, the safety rope group includes at least one safety rope, and the safety rope image block group includes at least one safety rope image block corresponding one-to-one with the aforementioned safety rope. In this case, the step of generating a corresponding first judgment result based on the distance between the personnel image block and the safety rope image block group includes: calculating the distance between the personnel image block and each safety rope image block; and when all personnel-rope distances are greater than a preset length, determining that the worker is not using the safety rope group and generating the aforementioned first judgment result accordingly.

[0007] In one embodiment, when calculating the person-rope distance, the distance between the edge of the person image block and the edge of the safety rope image block corresponding to the person-rope distance is calculated.

[0008] In one embodiment, the aforementioned high-altitude operation monitoring method further includes: selecting a safety rope image block corresponding to the safety rope used by the worker; determining whether the safety rope corresponding to the selected safety rope image block provides sufficient safety based on the aspect ratio of the selected safety rope image block and generating a corresponding second judgment result; and generating a second prompt message based on the second judgment result.

[0009] In one embodiment, the aforementioned step of determining whether the safety rope corresponding to the selected safety rope image block provides sufficient safety based on the aspect ratio of the selected safety rope image block and generating a corresponding second determination result includes: calculating the horizontal angle between the corresponding safety rope and the horizontal direction based on the aspect ratio of the selected safety rope image; and when the horizontal angle is less than the safety angle, determining that the safety rope cannot provide sufficient safety and generating a corresponding second determination result.

[0010] In one embodiment, the aforementioned step of determining whether the safety rope corresponding to the selected safety rope image block provides sufficient safety based on the aspect ratio of the selected safety rope image block and generating a corresponding second judgment result further includes: when the horizontal included angle is not less than the safety angle, calculating the length of the safety rope, and when the length of the safety rope is greater than the safety length, determining that the safety rope cannot provide sufficient safety and generating a corresponding second judgment result.

[0011] In one embodiment, the aforementioned high-altitude operation monitoring method further includes: setting the range of the work area before photographing it.

[0012] From another perspective, this invention proposes a high-altitude operation monitoring system, including a photography device, an image analysis and processing device, and an alert device. The photography device is adapted to photograph the work area to generate an image to be judged; the image analysis and processing device is electrically coupled to the photography device to receive the image to be judged, analyzes the image to identify the personnel image block corresponding to the worker and the safety rope image block group corresponding to the safety rope group, and then generates a corresponding first judgment result based on the distance between the personnel image block and the safety rope image block group; the alert device is electrically coupled to the image analysis and processing device to receive the first judgment result and generate a first prompt message based on the first judgment result.

[0013] In one embodiment, the aforementioned safety rope group includes at least one safety rope, and the safety rope image block group includes at least one safety rope image block, with each safety rope image block corresponding one-to-one to the aforementioned at least one safety rope; the image analysis and processing device calculates the distance between the person image block and each safety rope image block, and when all the person-rope distances are greater than a preset length, it determines that the worker has not used the safety rope group and generates a first judgment result accordingly.

[0014] In one embodiment, the image analysis and processing device selects the safety rope image block corresponding to the safety rope used by the worker, and determines whether the safety rope corresponding to the selected safety rope image block provides sufficient safety based on the aspect ratio of the selected safety rope image block and generates a corresponding second judgment result. The reminder device receives this second judgment result and generates a second prompt message based on the second judgment result.

[0015] By employing the above-mentioned technology, the high-altitude operation monitoring method and system provided by the present invention can automatically analyze whether the personnel working on site are using safety ropes properly by analyzing images taken from the workplace. Furthermore, it can promptly issue warning signals when it is found that safety ropes are not being used or that there are problems with the way safety ropes are being used. Therefore, it can effectively reduce the manpower required to supervise the use of safety ropes, and on the other hand, it can also effectively increase the safety of workers. Attached Figure Description

[0016] Figure 1 is a schematic diagram of the architecture of a high-altitude operation monitoring system according to an embodiment of the present invention.

[0017] Figure 2 is a flowchart of a high-altitude operation monitoring method according to an embodiment of the present invention.

[0018] Figure 3 is a schematic diagram of the image to be judged and the image blocks obtained after image analysis according to an embodiment of the present invention.

[0019] Figure 4 is a flowchart of a high-altitude operation monitoring method according to an embodiment of the present invention.

[0020] Figure 5 is a detailed flowchart of step S410 according to an embodiment of the present invention.

[0021] Figure 6 is a schematic diagram of the angle between the safety rope and the horizontal direction according to an embodiment of the present invention.

[0022] Figure 7 is a schematic diagram of the architecture of an image analysis and processing apparatus according to an embodiment of the present invention.

[0023] The symbols in the attached figures are briefly explained as follows: 10: High-altitude operation monitoring system; 100: Photography device; 110: Image analysis and processing device; 110a: Edge device; 110b: Background device; 120: Reminder device; 150, 350: Working area; 310, 312: Safety rope; 320, 322: Safety rope image block; 330: Worker; 340: Personnel image block; DET1: First judgment result; DET2: Second judgment result; H1: Height of safety rope image block; IMG: Image to be judged; POS: Location information; S200~S240: Implementation steps of the high-altitude operation monitoring method according to one embodiment of the present invention; S400~S420: Implementation steps of the high-altitude operation monitoring method according to another embodiment of the present invention; S500~S550: Detailed steps in implementing step S410 according to an embodiment of the present invention; SI: Image block size; W1: Width of safety rope image block; θ: Horizontal included angle. Detailed Implementation

[0024] To ensure that the content is clearly understood by those skilled in the art, it should be noted that the term "electrically coupled" as used below means that electronic signals can be transmitted between two electrically coupled objects. Unless otherwise specified, the transmission of electronic signals can be wired or wireless, and the direction of transmission can be unidirectional or bidirectional.

[0025] Please refer to Figure 1, which is a schematic diagram of the architecture of a high-altitude operation monitoring system according to an embodiment of the present invention. In this embodiment, the high-altitude operation monitoring system 10 includes a camera 100, an image analysis and processing device 110, and an alert device 120. The camera 100 is electrically coupled to the image analysis and processing device 110 so that the electronic data captured by the camera 100 can be transmitted to the image analysis and processing device 110. The image analysis and processing device 110 is further electrically coupled to the alert device 120 so that the electronic data generated by the image analysis and processing device 110 can be transmitted to the alert device 120.

[0026] The present invention will be described in detail below by combining Figures 1 and 2, wherein Figure 2 is a flowchart of a high-altitude operation monitoring method according to an embodiment of the present invention. In the embodiment shown in Figures 1 and 2, the photographic device 100 first photographs the work area 150 to generate a corresponding image to be judged (IMG) (step S200). The image to be judged IMG generated by the photographic device 100 is transmitted to the image analysis and processing device 110, whereby the image analysis and processing device 110 can use existing image analysis technology to find the personnel image block corresponding to the worker and the safety rope image block group corresponding to the safety rope group from the image to be judged IMG (step S210) (step S220).

[0027] It is worth mentioning that operators can mark the work area with specific objects in advance, or they can set the scope of the work area from the acquired image before shooting, or they can adjust the position of the work area on the image analysis and processing device 110 as needed. This invention does not limit these actions. Furthermore, during the image analysis process in step S210 to obtain personnel image blocks, a filtering step based on the size of the personnel image block can be added to determine whether it is necessary to judge the safety rope status of the personnel corresponding to that personnel image block. Specifically, because the captured image may contain personnel who are located in the distance but are included in the work area due to spatial perspective, these distant personnel should be excluded when judging the safety rope status to avoid misjudgment in subsequent step S230. To achieve this effect, in step S210, the size of the found personnel image block can be compared with the preset size. If it is found that the size of the found personnel image block is smaller than the preset size, it means that the staff member corresponding to this personnel image block is likely to be a person located far away. Therefore, this personnel image block can be excluded so that it will not enter step S230 for subsequent judgment.

[0028] Please refer to Figure 3, which is a schematic diagram of the image to be judged and the image blocks obtained after image analysis according to an embodiment of the present invention. As shown in Figure 3, the area covered by the working area 350 in the image to be judged includes a worker 330 and two safety ropes 310 and 312, wherein the safety ropes 310 and 312 form the aforementioned safety rope group. After the image analysis operation of the image analysis and processing device 110, in addition to obtaining the personnel image block 340 corresponding to the worker 330 from the image to be judged shown in Figure 3, the image analysis and processing device 110 can also obtain a safety rope image block group corresponding to the safety rope group. In this embodiment, the safety rope group includes safety ropes 310 and 312, and the safety rope image block group includes a safety rope image block 320 corresponding to the safety rope 310 and a safety rope image block 322 corresponding to the safety rope 312. In one embodiment, the image analysis operation of the image analysis and processing apparatus 110 may be performed by an object detection neural network (e.g., CenterNet or CornerNet), but the present invention is not limited thereto.

[0029] After obtaining the aforementioned image blocks, the image analysis and processing device 110 can calculate the distance between the personnel image block 340 and the safety rope image block group, and determine whether the worker 330 has used the safety rope group based on the calculated distance and generate the corresponding first judgment result (step S230).

[0030] Specifically, the image analysis and processing device 110 can perform the aforementioned distance calculation operation by calculating the distance (hereinafter referred to as the person-rope distance) between the personnel image block 340 and each safety rope image block 320 and 322 in the safety rope image block group. In different embodiments, technicians can use different benchmarks to calculate the aforementioned person-rope distance depending on the actual needs, and such changes do not affect the normal operation of the technology provided by the present invention. For example, the distance between the center points of the image blocks can be selected as the person-rope distance, or the distance between the boundaries of two image blocks can be selected as the person-rope distance.

[0031] After calculating the rope distance, the image analysis and processing device 110 can determine whether the worker 330 has used the safety rope group based on the calculated rope distance and a pre-set preset length. A common and simple method is to determine that the worker 330 has not used the safety rope corresponding to that rope distance if the rope distance is greater than the preset length, and that the worker 330 has not used the safety rope group if the rope distance corresponding to each safety rope in the safety rope group is greater than the preset length. The preset length used in the determination process can be 0, that is, the worker 330 is only considered to have used the safety rope if the personnel image block and the safety rope image block are adjacent or even partially overlapped. Of course, the determination method used here is not limited to the simple determination method described above. Those skilled in the art can also use more complex methods to determine whether the worker 330 has used the safety rope group, and the present invention does not limit this.

[0032] As those skilled in the art know, the relative positions of the target object (worker or safety rope) and the corresponding image block (personnel image block or safety rope image block) may vary depending on the type of image analysis method selected. Therefore, although in the embodiment shown in Figure 3, the positional relationship between the target object and the analyzed image block corresponding to it is represented by the smallest rectangle enclosing the target object, different image analysis methods may yield image blocks of different sizes or shapes for the same target object shown in Figure 3. Those skilled in the art can easily combine the technology provided by this invention with the results obtained from different image analysis methods, based on the technical spirit described in the previous embodiments, to enable the entire system to operate normally; therefore, further complex explanations are omitted here.

[0033] As shown in Figure 3, the personnel image block 340 and the safety rope image block 322 partially overlap. Therefore, in this embodiment, the worker 330 will be judged to be using the safety rope 312.

[0034] Referring again to Figures 1 and 2, after determining that the worker has used a certain safety rope, the first judgment result DET1 output by the image analysis and processing device 110 can keep the reminder device 120 in normal working order, or the image analysis and processing device 110 can directly not generate or output this first judgment result DET1 to the reminder device 120 to reduce unnecessary hardware computing consumption and network traffic. Conversely, if it is determined that the worker has not used any safety rope in the safety rope group, the image analysis and processing device 110 can generate a second first judgment result DET1. Once this second first judgment result DET1 is received, the reminder device 120 can generate a first prompt message based on the received second first judgment result DET1 (step S240). The first prompt message here can be expressed in different ways according to different needs, such as displaying specific text, adding special marks to the display screen, or generating specific sound effects, etc., and the present invention does not limit this.

[0035] It is worth noting that, in order to avoid occasional misjudgments inappropriately triggering the prompt message, the aforementioned first prompt message can be set to be triggered only when more than a predetermined number of second first judgment results DET1 are received within a predetermined time period. In other words, in one embodiment, the designer can set triggering logic on the reminder device 120 to trigger the prompt message only when more than a predetermined number of second first judgment results DET1 are received within a predetermined time period; or, in another embodiment, the designer can set triggering logic in the image analysis and processing device 110 to send the second first judgment result DET1 to the reminder device 120 to trigger the prompt message only when the number of times the result of determining that the worker did not use the safety rope group exceeds a predetermined number within a predetermined time period.

[0036] Furthermore, after determining that a worker used a certain safety rope, the image analysis and processing device 110 can further confirm whether the safety rope used by the worker can provide sufficient safety.

[0037] Please refer to Figure 4, which is a flowchart of a high-altitude operation monitoring method according to an embodiment of the present invention. The operation process of steps S200 to S240 in this embodiment is largely the same as that shown in Figure 2, and will not be repeated here. The difference from the embodiment shown in Figure 2 is that after step S230 determines that a worker has used a safety rope, the first judgment result DET1 output by the image analysis and processing device 110 can not only keep the alerting device 120 in normal operation, but also trigger step S400 for further operation; in another approach, after step S230 determines that a worker has used a safety rope, the image analysis and processing device 110 may directly not generate or output the first judgment result DET1 to the alerting device 120, but still allow the process to proceed to step S400 for further operation.

[0038] It should be noted that regardless of which method is used to bring the process to step S400, it indicates that the image analysis and processing device 110 has determined that the worker has indeed used a safety rope. Therefore, in step S400, the image analysis and processing device 110 selects the safety rope image block corresponding to the used safety rope, and then further determines whether the safety rope can provide sufficient safety based on the aspect ratio of the selected safety rope image block, generating a corresponding second determination result (step S410).

[0039] Please refer to Figure 5, which is a detailed flowchart of step S410 according to an embodiment of the present invention. In this embodiment, the aspect ratio of the selected safety rope image block is first calculated (step S500). Then, the horizontal angle between the safety rope and the horizontal direction corresponding to this safety rope image block can be obtained using an inverse trigonometric function based on the calculated aspect ratio (step S505). As shown in Figure 6, when the width of the safety rope image block 610 corresponding to the safety rope 600 is W1 and the height is H1, the horizontal angle θ between the safety rope 600 and the horizontal direction can be calculated based on the ratio of W1 to H1 and the inverse trigonometric function.

[0040] After calculating the horizontal angle in step S505, the process proceeds to step S510 to determine whether this horizontal angle is less than the safe angle. Since the horizontal angle decreases as the safety rope is stretched further (i.e., the worker moves further away from the rope's anchor point) while working at the same height, a horizontal angle less than the safe angle likely indicates that the safety rope is stretched too long and cannot provide sufficient tension to prevent a fall. Therefore, if step S510 determines that the safety rope is temporarily unable to provide sufficient safety to prevent a fall, this embodiment concludes that the safety rope is temporarily unable to provide sufficient protection to prevent a fall, thus generating the first second determination result DET2 (step S520).

[0041] On the other hand, when the judgment result of step S510 is negative, in some embodiments it can be determined that the safety rope can provide sufficient safety and subsequent operations can proceed. Alternatively, in this embodiment, the length of the safety rope will be further calculated to ensure that the safety rope can provide sufficient safety. In this embodiment, when the judgment result of step S510 is negative, the process will proceed to step S530 to determine whether the length of the safety rope is greater than the preset safety length. The judgment criterion here requires the actual length of the safety rope, which can be obtained in many ways. For example, it can be calculated based on the ratio between the size of the safety rope in the image and other objects of known size, or it can be marked on the safety rope to show the size it has been pulled out. Once the judgment result of step S530 is positive, based on the reason that the safety rope may not be able to provide sufficient tension to prevent the worker from falling to the ground due to excessive pulling length, the process will proceed to step S540 to generate a second judgment result DET2. Conversely, when the judgment result of step S530 is negative, the process will proceed to step S550 to generate a third judgment result DET2.

[0042] Please refer back to Figure 4. After the second judgment result to be generated in step S410 is determined by steps S500 to S550 as described above, the reminder device 120 can generate a corresponding second prompt message based on the received second judgment result (step S420). For example, when receiving the first or second second judgment result DET2, the reminder device 120 can use the second prompt message to express concern that the safety rope is stretched too much; conversely, when receiving the third second judgment result DET2, the reminder device 120 can remain inactive or use the second prompt message to express that the safety rope is still within the normal range of use. Of course, the second prompt message can also be triggered only when the same second judgment result is generated more than a preset number of times within a preset time period. The specific method is similar to that described earlier when triggering the first prompt message, and will not be repeated here.

[0043] Referring also to FIG7, in the embodiment shown in FIG7, the image analysis and processing apparatus 110 includes an edge device 110a and a back-end device 110b. The edge device 110a may be electrically coupled to the back-end device 110b via wired or wireless means. The edge device 110a receives the image to be judged (IMG) from the imaging device 100 and performs image analysis to obtain the personnel image block 340 and the safety rope image blocks 320 / 322. Then, it transmits the position information (e.g., pixel coordinates) POS of the personnel image block 340 and the safety rope image blocks 320 / 322 in the image to be judged, or the image block size SI of the safety rope image blocks 320 / 322, to the back-end device 110b. Next, the back-end device 110b can calculate the distance between the personnel image block 340 and the safety rope image block group based on the location information POS, and determine whether the worker 330 has used the safety rope group based on the calculated distance and generate a corresponding first judgment result DET1. Alternatively, it can calculate whether the safety rope can provide sufficient tension to prevent falling based on the image block size SI and generate a corresponding second judgment result DET2.

[0044] In summary, the high-altitude operation monitoring method and system provided by this invention can automatically analyze whether the personnel working on site are using safety ropes appropriately by analyzing images taken from the workplace. Furthermore, it can promptly issue warning signals when it detects that safety ropes are not being used or that problems are occurring in the way safety ropes are being used. Therefore, it can effectively reduce the manpower required to supervise the use of safety ropes and can also effectively increase the safety of workers.

Claims

1. A method of aerial work monitoring, adapted to confirm whether a worker is utilizing a safety line set, characterized in that, comprises: capturing a work area to obtain a to-be-judged image; analyzing the to-be-judged image to find a staff image block corresponding to the staff; analyzing the to-be-judged image to find a safety rope image block group corresponding to the safety rope group; and generating a first judgment result corresponding to the distance between the staff image block and the safety rope image block group; and generating first prompt information according to the first judgment result. The safety rope group comprises at least one safety rope, the safety rope image block group comprises at least one safety rope image block, the at least one safety rope image block corresponds to the at least one safety rope one-to-one, and the step of generating the first judgment result corresponding to the distance between the staff image block and the safety rope image block group comprises:

2. The aerial work monitoring method of claim 1, wherein, calculating a staff-rope distance between the staff image block and each of the at least one safety rope image block; and when all the staff-rope distances are greater than a preset length, judging that the staff does not use the safety rope group and generating the first judgment result corresponding thereto. When calculating the staff-rope distance, the distance between the edge of the staff image block and the edge of the at least one safety rope image block corresponding to the staff-rope distance is calculated.

3. The aerial work monitoring method of claim 2, wherein, 4. The aerial work monitoring method of claim 2, further comprising: selecting the safety rope image block corresponding to the safety rope used by the staff from the at least one safety rope; judging whether the safety rope corresponding to the selected safety rope image block provides sufficient safety according to the aspect ratio of the selected safety rope image block and generating a second judgment result corresponding thereto; and generating second prompt information according to the second judgment result. The step of judging whether the safety rope corresponding to the selected safety rope image block provides sufficient safety according to the aspect ratio of the selected safety rope image block and generating the second judgment result corresponding thereto comprises: calculating a horizontal included angle between the corresponding safety rope and the horizontal direction according to the aspect ratio of the selected safety rope image; and 5. The aerial work monitoring method of claim 4, wherein, when the horizontal included angle is less than a safety angle, judging that the safety rope cannot provide sufficient safety and generating the second judgment result corresponding thereto. The step of judging whether the safety rope corresponding to the selected safety rope image block provides sufficient safety according to the aspect ratio of the selected safety rope image block and generating the second judgment result corresponding thereto further comprises: when the horizontal included angle is not less than the safety angle, calculating the length of the safety rope, and when the length of the safety rope is greater than a safety length, judging that the safety rope cannot provide sufficient safety and generating the second judgment result corresponding thereto.

7. The aerial work monitoring method of claim 1, further comprising:

6. The aerial work monitoring method of claim 5, wherein, setting the range of the work area before capturing the work area. comprises: a camera adapted to capture a work area to generate a to-be-judged image; ​ 8. An aerial work monitoring system adapted to confirm whether a worker is utilising a safety line set, characterised in that, ​ ​ An image analysis and processing device is electrically coupled to the camera device to receive the image to be judged, analyzes the image to be judged to find a personnel image block corresponding to the worker and to find a safety rope image block set corresponding to the safety rope set, and generates a first judgment result according to the distance between the personnel image block and the safety rope image block set; and A reminding device is electrically coupled to the image analysis and processing device to receive the first judgment result and generate a first prompt information according to the first judgment result.

9. The aerial monitoring system of claim 8, wherein, The safety rope set includes at least one safety rope, the safety rope image block set includes at least one safety rope image block, the at least one safety rope image block one-to-one corresponds to the at least one safety rope, the image analysis and processing device calculates a person-rope distance between the personnel image block and each of the at least one safety rope image block, and judges that the worker does not use the safety rope set when all the person-rope distances are greater than a preset length, and correspondingly generates the first judgment result.

10. The aerial monitoring system of claim 9, wherein, The image analysis and processing device selects the safety rope image block corresponding to the safety rope used by the worker from the at least one safety rope, and generates a second judgment result according to whether the safety rope corresponding to the selected safety rope image block provides sufficient safety according to the aspect ratio of the selected safety rope image block, and the reminding device receives the second judgment result and generates a second prompt information according to the second judgment result.

Images