Method and equipment for monitoring personnel intrusion in operation area of scraper conveyor and medium

Abstract

The invention relates to the field of safety monitoring, and discloses a method and device for monitoring personnel intrusion in a scraper operation area and a medium, and the method comprises the steps: obtaining an image data set in a preset area at a fixed frame rate; identifying the image data set to determine a scraper area corresponding to each image and a preset trunk area of a person; overlapping the scraper areas corresponding to each frame of image in a first preset frame number before the target frame to obtain a scraper overlapping area corresponding to the target frame of image; obtaining an intrusion state mark value corresponding to the target frame image based on a personnel preset trunk area and a scraper overlapping area of the target frame image; and judging whether a person intrudes into the operation area of the scraper according to the intruding state mark value corresponding to each image in the second preset frame number. By superposing the scraper areas corresponding to the multiple frames of images, the misjudgment of the scraper areas caused by factors such as temporary shielding of personnel can be effectively avoided, so that the monitoring accuracy is improved.

Description

Technical Field

[0001] The present application relates to the field of safety monitoring, and in particular to a method, equipment and medium for monitoring the intrusion of personnel into the operating area of ​​a scraper conveyor. Background Art

[0002] In industrial production, scraper conveyors are common conveying equipment, and their operating areas are somewhat dangerous, so people are usually prohibited from entering the scraper conveyor operating area.

[0003] Currently, monitoring for intrusion into scraper conveyor operating areas primarily relies on manual inspections or simple physical safeguards. However, due to the time intervals between manual inspections, real-time monitoring is impossible, and intruders may enter during these intervals, leading to safety accidents. Simple physical safeguards cannot prevent people from climbing over or damaging the area, nor can they issue timely alerts to notify relevant personnel. Therefore, existing monitoring methods lack timeliness and accuracy, and a more efficient monitoring method is urgently needed. Summary of the Invention

[0004] In order to solve the above problems, the present application proposes a method, device and medium for monitoring personnel intrusion into the operation area of ​​a scraper, wherein the method includes:

[0005] An image data set within a preset area is acquired at a fixed frame rate; the image data set is identified to determine the scraper area and the preset torso area of ​​the person corresponding to each image; the scraper areas corresponding to each frame image within a first preset number of frames before the target frame are superimposed to obtain the scraper superposition area corresponding to the target frame image; based on the preset torso area of ​​the person and the scraper superposition area of ​​the target frame image, an intrusion status mark value corresponding to the target frame image is obtained; and through the intrusion status mark value corresponding to each image within a second preset number of frames, it is determined whether there is a person intruding into the scraper operation area.

[0006] In one example, the preset torso area of ​​the person is the foot area; the preset torso area of ​​the person corresponding to each image is determined by identifying the image data set, specifically including: identifying the image data set to determine the person area bounding box corresponding to each image; using the lower boundary of the person area bounding box as the center line of the foot area; multiplying the height of the person area bounding box by a preset ratio as the height of the foot area bounding box; using the width of the person area bounding box as the width of the foot area bounding box; and determining the foot area corresponding to each image based on the height and width of the foot area bounding box.

[0007] In one example, the scraper areas corresponding to each frame image within a first preset number of frames before the target frame are superimposed to obtain the scraper superposition area corresponding to the target frame image, specifically including: determining the first preset number of frames, and the scraper superposition area corresponding to the current frame; determining the frame number difference between the current frame and the target frame; determining the expected removal frame and the expected addition frame through the first preset number of frames and the frame number difference; removing the scraper areas corresponding to the expected removal frames from the scraper superposition area corresponding to the current frame to obtain an intermediate scraper area; adding the scraper areas corresponding to the expected addition frames to the intermediate scraper area to obtain the scraper superposition area corresponding to the target frame image.

[0008] In one example, obtaining the intrusion status mark value corresponding to the target frame image based on the preset torso area of ​​the person in the target frame image and the superimposed area of ​​the scraper specifically includes: calculating the intersection-and-union ratio of the preset torso area of ​​the person in the target frame image and the superimposed area of ​​the scraper; and determining the intrusion status mark value corresponding to the target frame image based on the intersection-and-union ratio.

[0009] In one example, determining the intrusion status mark value corresponding to the target frame image based on the intersection-and-union ratio specifically includes: if the intersection-and-union ratio is higher than a preset intersection-and-union ratio threshold, using a first preset value as the intrusion status mark value corresponding to the target frame image; if the intersection-and-union ratio is not higher than the preset intersection-and-union ratio threshold, using a second preset value as the intrusion status mark value corresponding to the target frame image.

[0010] In one example, determining the intrusion status flag value corresponding to the target frame image based on the intersection-over-union ratio specifically includes: using the intersection-over-union ratio as the intrusion status flag value corresponding to the target frame image.

[0011] In one example, the determination of whether a person has intruded into the scraper operating area is performed by using the intrusion status mark value corresponding to each image within the second preset frame number, specifically including: determining the sum of the intrusion status mark values ​​corresponding to each image within the second preset frame number; if the ratio of the sum of the intrusion status mark values ​​to the second preset frame number is higher than the preset ratio, then a person has intruded into the scraper operating area.

[0012] In one example, after determining whether a person has intruded into the scraper operating area through the intrusion status mark value corresponding to each image within the second preset frame number, the method further includes: determining whether a person has intruded into the scraper operating area; determining a corresponding preset early warning method based on the ratio of the sum of the intrusion status mark values ​​to the second preset frame number; issuing an early warning according to the preset early warning method, and caching video frames within a third preset alarm time period before and after the early warning moment.

[0013] The present application also provides a device for monitoring personnel intrusion into a scraper operation area, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute: acquiring an image data set within a preset area at a fixed frame rate; identifying the image data set to determine the scraper area and the preset torso area of ​​the personnel corresponding to each image; superimposing the scraper areas corresponding to each frame image within a first preset number of frames before a target frame to obtain a scraper superimposed area corresponding to the target frame image; obtaining an intrusion status mark value corresponding to the target frame image based on the preset torso area of ​​the personnel and the scraper superimposed area of ​​the target frame image; judging whether there is a person intruding into the scraper operation area by the intrusion status mark value corresponding to each image within a second preset number of frames.

[0014] The present application also provides a non-volatile computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to: acquire an image data set within a preset area at a fixed frame rate; identify the image data set to determine the scraper area and the preset torso area of ​​the person corresponding to each image; superimpose the scraper areas corresponding to each frame image within a first preset number of frames before a target frame to obtain a scraper superposition area corresponding to the target frame image; obtain an intrusion status mark value corresponding to the target frame image based on the preset torso area of ​​the person and the scraper superposition area of ​​the target frame image; and determine whether there is a person intruding into the scraper operation area through the intrusion status mark value corresponding to each image within a second preset number of frames.

[0015] The method proposed in this application can achieve the following beneficial effects: by superimposing the scraper areas corresponding to multiple frames of imagery, it can effectively avoid misjudgments of the scraper areas caused by factors such as temporary obstruction by personnel, thereby improving monitoring accuracy. Calculations are performed using the intrusion status flag values ​​corresponding to multiple frames of imagery to avoid misjudgments that may occur with single-frame judgments. Furthermore, by adjusting the preset parameter values, real-time monitoring can be performed, allowing staff to respond promptly and reducing casualties and property losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0017] Figure 1 This is a flow chart of a method for monitoring personnel intrusion into a scraper operation area according to an embodiment of the present application;

[0018] Figure 2 This is a structural schematic diagram of a personnel intrusion monitoring device in the operation area of ​​a scraper conveyor according to an embodiment of the present application;

[0019] Figure 3 This is a structural schematic diagram of a personnel intrusion monitoring device in the scraper operation area in an embodiment of the present application. DETAILED DESCRIPTION

[0020] To make the purpose, technical solutions, and advantages of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the specific embodiments of this application and the corresponding drawings. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0021] The following describes in detail the technical solutions provided by various embodiments of the present application in conjunction with the accompanying drawings.

[0022] Figure 1 This is a flowchart of a method for monitoring personnel intrusion into a scraper operating area, as provided in one or more embodiments of this specification. This method can be applied to monitoring personnel intrusion into a scraper operating area. The process can be executed by a corresponding computing device (e.g., a computing device located near the scraper, or a cloud server, etc.). Certain input parameters or intermediate results in the process can be manually adjusted to help improve accuracy.

[0023] The analysis method involved in the embodiments of the present application can be implemented by a terminal device or a server, and the present application does not impose any special restrictions on this. For ease of understanding and description, the following embodiments are described in detail using a server as an example.

[0024] It should be noted that the server can be a single device or a system composed of multiple devices, that is, a distributed server, and this application does not make any specific restrictions on this.

[0025] like Figure 1 As shown, the embodiment of the present application provides a method for monitoring personnel intrusion into the operation area of ​​a scraper, comprising:

[0026] S101: Acquire an image dataset within a preset area at a fixed frame rate.

[0027] An image acquisition device (e.g., a high-definition camera) installed within the scraper's operating area acquires omnidirectional images of the area at a fixed frame rate f (frames per second), thereby generating an image dataset. The pre-set area can refer to the scraper's operating area, or it can be a larger area encompassing the scraper's operating area, for example, the area encompassing the scraper's operating area and the vicinity of the image acquisition device. The image dataset includes images captured by different image acquisition devices at different times. The image acquisition device's installation position and angle are optimized to ensure complete coverage of the scraper's operating area.

[0028] S102: Identify the image data set to determine the scraper area and the preset torso area of ​​the person corresponding to each image.

[0029] After obtaining the image data set within the preset area, the scraper and personnel can be detected for each image in the image data set. During monitoring, the deep neural network model can be used to quickly and accurately identify the scraper area and the preset torso area of ​​the personnel. It should be noted that the preset torso area of ​​the personnel here refers to the preset parts on the torso of the personnel, such as the head, hands, feet and other areas. This application uses the feet as an example. In particular, for the scraper area, the output is presented in the form of polygons. This output method can more accurately depict the actual area range of the scraper and provide more accurate data for subsequent operations such as intersection and union calculations.

[0030] In one embodiment, when performing image recognition, image recognition can be performed using the YOLO target detection algorithm. Specifically, it is necessary to select any image in the image data set as the target image; by segmenting the target image to obtain a preset number of image grids; by performing target detection on the preset number of image grids, to determine the image grid containing the scraper and the preset torso of the person; based on the image grid containing the scraper, determine the scraper area corresponding to the target image; based on the image grid containing the preset torso of the person, determine the preset torso area of ​​the person in the target image; by traversing any image in the image data set, to obtain the scraper area and the preset torso area of ​​the person corresponding to each image in the image data set. When using the YOLO target detection algorithm, the position of the scraper and the person, as well as the confidence c (0≤C≤1, the closer C is to 1, the more reliable the detection result) can be quickly and accurately identified. When determining the image grid containing the scraper, the grid with a scraper confidence higher than a preset threshold can be used as the image grid containing the scraper. The same applies to the image grid containing the preset torso of the person, which will not be repeated here.

[0031] In one embodiment, in addition to directly detecting the preset torso of a person in an image through the YOLO target detection algorithm as described above, the person area in the image can also be detected first, and then the person area is processed to obtain the preset torso area of ​​the person. For example, when the preset torso area of ​​the person is the foot area, during detection, the image data set can be first identified to determine the person area bounding box corresponding to each image, and then the lower boundary of the person area bounding box is used as the center line of the foot area. The height of the person area bounding box is then multiplied by a preset ratio (such as 10%) as the height of the foot area bounding box, and the width of the person area bounding box can be directly used as the width of the foot area bounding box. After obtaining the height and width of the foot area bounding box, the foot area corresponding to each image can be determined.

[0032] S103: Superimposing the scraper regions corresponding to the frame images within a first preset number of frames before the target frame to obtain the scraper superimposed region corresponding to the target frame image.

[0033] In order to accurately determine the scraper area and avoid misjudgment of the scraper area due to factors such as temporary obstruction by personnel, the scraper areas corresponding to each frame image within the first preset number of frames before the target frame can be superimposed to obtain the scraper superposition area corresponding to the target frame image. In the subsequent judgment process, the scraper superposition area corresponding to the target frame image is used as the scraper operation area for judgment. It should be noted that the "frame image" mentioned in this application refers to the image corresponding to the frame, that is, the image data set obtained in the frame.

[0034] In one embodiment, when superimposing a scraper area, it is necessary to determine the size of a first preset number of frames, the scraper superimposition area corresponding to the current frame. Then, the frame difference between the current frame and the target frame is determined. At this time, the first preset number of frames and the frame difference can be used to determine the expected removal frame and the expected addition frame. The scraper areas corresponding to the expected removal frames are then removed from the scraper superimposition area corresponding to the current frame to obtain an intermediate scraper area. Simultaneously, the scraper areas corresponding to the expected addition frames are added to the intermediate scraper area to obtain the scraper superimposition area corresponding to the target frame image.

[0035] The specific process is as follows: Let the current frame be the nth frame, and the scraper area of ​​the current frame be M n , the historical scraper superposition area is U n-1 , the current frame scraper superposition area is U n Assuming that the first preset number of frames is N, the first preset number of frames can be understood as the maximum number of superimposed frames. When n≤N, it means that the number of frames currently processed has not yet reached the maximum number of superimposed frames. At this time, the scraper area M of the current frame is directly n Add to History Scraper Overlay Area U n-1 Un =U n-1 ∪{M n}. When n>N, and the target frame differs from the current frame by one frame, the number of frames expected to be removed and the number of frames expected to be added are both one. Specifically, the expected frame to be removed is the earliest frame corresponding to the scraper superposition area list of the current frame, and the expected frame to be added is the next frame of the current frame, that is, the target frame. In order to ensure that the length of the historical scraper superposition area list does not exceed the first preset number of frames, it is necessary to remove the scraper area M corresponding to the earliest frame. n-N+1 , and then the scraper area M corresponding to the target frame n+1 Join the list, that is, U n+1 =(U n \{M n-N+1})∪{M n+1 This multi-frame superposition method can more comprehensively and accurately depict the actual area of ​​the scraper, effectively reducing interference caused by accidental factors such as human obstruction.

[0036] S104: Based on a preset torso area of ​​the person in the target frame image and an overlapping area of ​​the scraper, obtaining an intrusion state mark value corresponding to the target frame image.

[0037] In one embodiment, the target frame image may be compared with a preset torso area of ​​the person and the scraper superimposed area to obtain an intrusion status flag value corresponding to the target frame image. The intrusion status flag value here is used to reflect the possibility of the user intruding into the preset area.

[0038] In one embodiment, when determining the intrusion status flag value, the intersection-over-union (IoU) ratio (IoU) of the preset torso region of the person in the target frame image and the superimposed region of the scraper is first calculated. The IoU reflects the degree of overlap between the person's foot region and the superimposed region of the scraper. If the IoU is large, it indicates that the overlap between the person's foot region and the scraper region is high, and the suspicion of the person intruding into the scraper region in the frame is high. If the IoU is small, it indicates that the overlap between the person's foot region and the scraper region is low, and the suspicion of the person intruding into the scraper region in the frame is low.

[0039] Specifically, when calculating the intersection-and-union ratio, let the target frame be the nth frame, and the target frame scraper superposition area be H n , the foot area of ​​the person corresponding to the target frame is represented as P n , then when calculating the intersection over union (IoU) of the target frame n When, the calculation formula is IoU n =P n ∪U n / P n ∩U n .

[0040] In one embodiment, after the IoU is calculated, the IoU corresponding to the target frame can be compared with a preset IoU threshold. n >T IoU (preset intersection-over-union ratio threshold), it indicates that the overlapping part of the personnel foot area and the scraper area exceeds the system preset range. At this time, it is determined that the personnel in the frame is suspected of breaking into the scraper area, and the intrusion status mark value S of the frame is increased. n Set to the first preset value (such as 1); otherwise, S n Set to the second preset value (such as 0), in the above case, the formula is: S n ={1,IoU n >T IoU ;0,IoU n ≤T IoU In addition to using the preset value as the intrusion status flag value corresponding to the intersection-in-union ratio, the intersection-in-union ratio value can also be directly used as the intrusion status flag value.

[0041] These two judgment methods based on intersection-over-union ratio can quantify the positional relationship between the person's feet and the scraper area, providing a clear single-frame basis for subsequent multi-frame judgment.

[0042] S105: judging whether there is a person intruding into the scraper operation area according to the intrusion status flag value corresponding to each image within the second preset number of frames.

[0043] In order to avoid possible misjudgments in single-frame judgment, when making judgments, it will be determined whether there is a person breaking into the scraper operation area based on the intrusion status mark value of each image corresponding to each frame within the second preset frame number before the frame number corresponding to the judgment moment.

[0044] In one embodiment, the sum of the intrusion status mark values ​​corresponding to each image within a second preset frame number can be determined. If the ratio of the sum of the intrusion status mark values ​​to the second preset frame number is higher than the preset ratio, it is considered that a person has intruded into the scraper operation area.

[0045] Specifically, a queue Q of length K can be maintained at this time n , used to store the intrusion status mark of consecutive K frames. When n≤K, the queue Q n What is stored is the intrusion status mark from the 1st frame to the nth frame, that is, Q n =[S1,…,S n ]; when n>K, queue Q n It will be updated to store the intrusion status mark from the n-K+1 frame to the n frame, that is, Q n =[S n-K+1 ,…,S n]. Such a queue update mechanism ensures that the statistical analysis is always performed on the latest K consecutive frames.

[0046] Next, calculate the sum of all intrusion status marks in the queue T n , the calculation formula is:

[0047]

[0048] T n It reflects the total number of frames in which people are suspected of breaking in among the consecutive K frames. Then, the intrusion ratio R is calculated. n , the formula is R n =T n / K, which indicates the ratio of the number of frames in which people are suspected of breaking in to the total number of frames in the continuous K frames. n >R tg (preset ratio), it means that in the continuous K frames, the situation of suspected intrusion by personnel is relatively frequent, exceeding the threshold set by the system. At this time, it is determined that there are indeed personnel intruding into the scraper operation area.

[0049] In one embodiment, after determining that a person has intruded into the scraper operation area, the current alarm count status can be drawn on the annotated image to intuitively display the statistics of the person's intrusion status in multiple consecutive frames, making it easier for monitoring personnel to monitor abnormal situations in real time. When an alarm is issued, the corresponding preset warning method can be determined based on the ratio of the sum of the intrusion status mark values ​​to the second preset number of frames, such as immediately issuing an audible and visual alarm signal to attract the attention of on-site personnel and monitoring personnel. At the same time, video frames t1 seconds before the alarm and t2 seconds after the alarm will be cached (the cached frames are calculated based on the frame rate f, and are f×t1 and f×t2 respectively). These cached video frames can provide detailed process records for subsequent accident analysis. In addition, the current alarm image can be saved as intuitive evidence of the person's intrusion.

[0050] It can be understood that the above-mentioned preset number of frames can refer to either continuous frames or frames separated by a certain number of frames, with an interval of three frames as the premise. If the preset number of frames is 5 frames at this time, and the target frame is the nth frame, then the preset number of frames before the target frame are the n-4th frame, the n-8th frame, the n-12th frame, the n-16th frame, and the n-20th frame respectively.

[0051] In one embodiment, during the above-mentioned calculation process, collaborative data processing can be performed based on multiple pre-deployed computing nodes. These computing nodes work together to improve the efficiency and speed of data processing, and the computing nodes communicate and share data through the network.

[0052] This application has the following technical effects: Compared with traditional scraper personnel monitoring that relies on manual inspections and simple physical protection, manual inspections have time intervals and are prone to omissions, and physical protection cannot prevent illegal entry and does not provide real-time feedback. This application uses a multi-frame area superposition algorithm to fuse multiple frames of information to accurately outline the scraper area, avoid occlusion misjudgment, and improve recognition accuracy; personnel intrusion judgment combines intersection-over-union calculation and multi-frame statistics to reduce the misjudgment rate and make the judgment more reliable.

[0053] Theoretically, this application's intrusion detection is more real-time and efficient. The target detection module, based on the YOLO algorithm, can quickly identify targets and analyze data in real time. Once an intrusion is detected, an immediate alarm can be issued, allowing staff to respond promptly, reducing casualties and property losses.

[0054] To verify the effect, this application selected on-site scraper data to conduct a series of experiments, set up complex scenes such as personnel occlusion, sudden changes in light, and dust interference, and used the same equipment to ensure consistent conditions. The results showed that the traditional method had a false alarm rate of up to 54% due to personnel occlusion, while the technical solution of this application, with its multi-frame area superposition algorithm, had a false alarm rate of only 5%. In cases of sudden changes in light, dust interference, etc., the traditional method had a false alarm rate of 37%, while the technical solution of this application had a false alarm rate as low as 6% through a multi-frame intrusion status statistical judgment mechanism. The data strongly proves that the technical solution of this application has obvious advantages in accuracy and reliability, and can provide more reliable security for the scraper area.

[0055] like Figure 2 As shown, an embodiment of the present application further provides a device for monitoring human intrusion into an operating area of ​​a scraper, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to:

[0056] An image data set within a preset area is acquired at a fixed frame rate; the image data set is identified to determine the scraper area and the preset torso area of ​​the person corresponding to each image; the scraper areas corresponding to each frame image within a first preset number of frames before the target frame are superimposed to obtain the scraper superposition area corresponding to the target frame image; based on the preset torso area of ​​the person and the scraper superposition area of ​​the target frame image, an intrusion status mark value corresponding to the target frame image is obtained; and through the intrusion status mark value corresponding to each image within a second preset number of frames, it is determined whether there is a person intruding into the scraper operation area.

[0057] like Figure 3 As shown, the embodiment of the present application further provides a device for monitoring personnel intrusion into the operating area of ​​a scraper, comprising:

[0058] The image acquisition module 301 acquires an image data set within a preset area at a fixed frame rate.

[0059] The image recognition module 302 identifies the image data set to determine the scraper area and the preset torso area of ​​the person corresponding to each image.

[0060] The region superposition module 303 superimposes the scraper regions corresponding to the frame images within a first preset number of frames before the target frame to obtain the scraper superposition region corresponding to the target frame image.

[0061] The state marking module 304 obtains an intrusion state marking value corresponding to the target frame image based on a preset torso area of ​​the person in the target frame image and an overlapping area of ​​the scraper.

[0062] The event determination module 305 determines whether a person has intruded into the operating area of ​​the scraper conveyor according to the intrusion status flag value corresponding to each image within the second preset number of frames.

[0063] The embodiment of the present application further provides a non-volatile computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to:

[0064] An image data set within a preset area is acquired at a fixed frame rate; the image data set is identified to determine the scraper area and the preset torso area of ​​the person corresponding to each image; the scraper areas corresponding to each frame image within a first preset number of frames before the target frame are superimposed to obtain the scraper superposition area corresponding to the target frame image; based on the preset torso area of ​​the person and the scraper superposition area of ​​the target frame image, an intrusion status mark value corresponding to the target frame image is obtained; and through the intrusion status mark value corresponding to each image within a second preset number of frames, it is determined whether there is a person intruding into the scraper operation area.

[0065] The various embodiments in this application are described in a progressive manner. Similar portions between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the device and medium embodiments are generally similar to the method embodiments, so their descriptions are relatively simple. For relevant portions, refer to the descriptions of the method embodiments.

[0066] The devices and media provided in the embodiments of the present application correspond one-to-one to the methods. Therefore, the devices and media also have similar beneficial technical effects to their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0067] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0068] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0069] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0070] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0071] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0072] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0073] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0074] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0075] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A method for monitoring personnel intrusion into a scraper operation area, characterized in that: include: Acquire an image dataset within a preset area at a fixed frame rate; By identifying the image data set, the scraper area and the preset torso area of ​​the person corresponding to each image are determined; Superimposing the scraper machine areas corresponding to the frame images within a first preset number of frames before the target frame to obtain the scraper machine superimposed area corresponding to the target frame image; Obtaining an intrusion state flag value corresponding to the target frame image based on a preset torso area of ​​the person in the target frame image and an overlapping area of ​​the scraper; It is determined whether a person has intruded into the operating area of ​​the scraper conveyor by using the intrusion status mark value corresponding to each image within the second preset number of frames.

2. The method according to claim 1, characterized in that The person presets the torso area as the foot area; Identifying the image dataset to determine a preset torso region of a person corresponding to each image specifically includes: Identifying the image dataset to determine a person region bounding box corresponding to each image; Taking the lower boundary of the personnel area boundary box as the center line of the foot area; Multiplying the height of the personnel area bounding box by a preset ratio to obtain the height of the foot area bounding box; The width of the bounding box of the person area is used as the width of the bounding box of the foot area; The foot region corresponding to each image is determined based on the height and width of the foot region bounding box.

3. The method according to claim 1, characterized in that The step of superimposing the scraper regions corresponding to the respective frame images within a first preset number of frames before the target frame to obtain the scraper superimposed region corresponding to the target frame image specifically includes: Determining the first preset frame number and the scraper superposition area corresponding to the current frame; Determining a frame number difference between the current frame and the target frame; Determining the expected removal frames and the expected addition frames according to the difference between the first preset number of frames and the number of frames; removing the scraper machine areas corresponding to the expected removal frames from the scraper machine superposition area corresponding to the current frame to obtain an intermediate scraper machine area; The scraper machine areas corresponding to the expected added frames are added to the intermediate scraper machine area to obtain the scraper machine superposition area corresponding to the target frame image.

4. The method according to claim 1, wherein The obtaining of the intrusion status mark value corresponding to the target frame image based on the preset torso area of ​​the person in the target frame image and the superimposed area of ​​the scraper specifically includes: Calculating the intersection-over-union ratio of a preset torso area of ​​the person in the target frame image and an overlapping area of ​​the scraper; Based on the intersection-over-union ratio, an intrusion status flag value corresponding to the target frame image is determined.

5. The method according to claim 4, characterized in that Determining the intrusion status flag value corresponding to the target frame image based on the intersection-to-union ratio specifically includes: If the IoU is higher than a preset IoU threshold, taking a first preset value as an intrusion status flag value corresponding to the target frame image; If the IoU is not higher than the preset IoU threshold, the second preset value is used as the intrusion status flag value corresponding to the target frame image.

6. The method according to claim 4, characterized in that Determining the intrusion status flag value corresponding to the target frame image based on the intersection-to-union ratio specifically includes: The intersection-over-union ratio is used as an intrusion status flag value corresponding to the target frame image.

7. The method according to claim 1, characterized in that The step of judging whether a person has entered the scraper operation area by using the intrusion status flag value corresponding to each image within the second preset number of frames specifically includes: determining a sum of intrusion status flag values ​​corresponding to each image within the second preset number of frames; If the ratio of the sum of the intrusion status mark values ​​to the second preset number of frames is higher than the preset ratio, a person has intruded into the scraper operation area.

8. The method according to claim 7, characterized in that After determining whether a person has intruded into the scraper operation area based on the intrusion status flag value corresponding to each image within the second preset number of frames, the method further includes: Determine whether there are people entering the scraper operation area; determining a corresponding preset warning mode based on a ratio of the sum of the intrusion status flag values ​​to the second preset number of frames; An early warning is performed according to the preset early warning method, and video frames within a third preset alarm time period before and after the early warning moment are cached.

9. A device for monitoring personnel intrusion into the operating area of ​​a scraper, characterized in that: include: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, wherein the instructions are executed by the at least one processor to enable the at least one processor to perform: Acquire an image dataset within a preset area at a fixed frame rate; By identifying the image data set, the scraper area and the preset torso area of ​​the person corresponding to each image are determined; Superimposing the scraper machine areas corresponding to the frame images within a first preset number of frames before the target frame to obtain the scraper machine superimposed area corresponding to the target frame image; Obtaining an intrusion state flag value corresponding to the target frame image based on a preset torso area of ​​the person in the target frame image and an overlapping area of ​​the scraper; It is determined whether a person has intruded into the operating area of ​​the scraper conveyor by using the intrusion status mark value corresponding to each image within the second preset number of frames.

10. A non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions are configured to: Acquire an image dataset within a preset area at a fixed frame rate; By identifying the image data set, the scraper area and the preset torso area of ​​the person corresponding to each image are determined; Superimposing the scraper machine areas corresponding to the frame images within a first preset number of frames before the target frame to obtain the scraper machine superimposed area corresponding to the target frame image; Obtaining an intrusion state flag value corresponding to the target frame image based on a preset torso area of ​​the person in the target frame image and an overlapping area of ​​the scraper; It is determined whether a person has intruded into the operating area of ​​the scraper conveyor by using the intrusion status mark value corresponding to each image within the second preset number of frames.

Images