Intelligent calling method, device and equipment based on oil pipeline monitoring and medium

Abstract

The invention provides an intelligent calling method and device based on oil pipeline monitoring, equipment and a medium. The method comprises the following steps: inputting a fused video corresponding to each rod into a target object recognition model to obtain a target object; the fused video is obtained by fusing monitoring images acquired by all image acquisition devices of each pole; the target object recognition model is obtained based on a training sample and is used for recognizing different object types; the training sample comprises an intrusion equipment sample image, a personnel sample image and an animal sample image; in all monitoring pictures of the monitoring equipment, reminding to display a target monitoring picture in which the target object appears; each monitoring picture corresponds to the unique identifier of each rod; based on the target monitoring picture, obtaining a remote calling voice; the remote calling voice is used for driving the target object; and sending the remote shouting voice to a target loudspeaker of the target rod, and controlling the target loudspeaker to play the remote shouting voice for remote shouting.

Description

Technical Field

[0001] The present invention relates to the technical field of oil pipeline monitoring, and in particular to an intelligent shouting method, device, equipment and medium based on oil pipeline monitoring. Background Art

[0002] The aviation fuel pipeline (referred to as the fuel pipeline) is the energy artery of the airport, and ensuring the safety of the aviation fuel pipeline is an important responsibility. Therefore, it is necessary to monitor the abnormal conditions of the fuel pipeline.

[0003] In the related art, it is possible to monitor the presence of dangerous objects in the monitoring area of ​​the oil pipeline.

[0004] However, this dangerous object may cause greater danger to the oil pipeline. Therefore, how to drive away the dangerous object in time becomes a technical problem to be solved urgently. Summary of the invention

[0005] The present application provides an improved intelligent shouting method, device, equipment and medium based on oil pipeline monitoring.

[0006] The present application provides an intelligent shouting method based on oil pipeline monitoring, which is used to remotely shout at the monitoring area of ​​the oil pipeline, wherein a plurality of poles are arranged at intervals in the monitoring area, and each pole is equipped with a loudspeaker and a group of image acquisition devices; the group of image acquisition devices is used to monitor the poles and the monitoring area;

[0007] The intelligent shouting method based on oil pipeline monitoring includes:

[0008] The fused video corresponding to each pole is input into the target object recognition model to obtain the target object; the fused video is obtained by fusing the monitoring images collected by all image acquisition devices of each pole; the target object recognition model is obtained based on training samples and is used to identify different object types; the training samples include sample images of intrusion equipment, sample images of personnel, and sample images of animals;

[0009] Among all monitoring screens of the monitoring device, a target monitoring screen where the target object appears is displayed; each monitoring screen corresponds to a unique identifier of each pole;

[0010] Based on the target monitoring screen, a remote shouting voice is obtained; the remote shouting voice is used to drive away the target object;

[0011] The remote shouting voice is sent to the target loudspeaker of the target pole, and the target loudspeaker is controlled to play the remote shouting voice to perform remote shouting.

[0012] Further, the acquiring of the remote shouting voice based on the target monitoring screen includes: determining the target dangerous behavior of the target object based on the target monitoring screen; acquiring, based on the dangerous behavior of the target object, the default shouting voice corresponding to the dangerous behavior of the target object from the correspondence between different dangerous behaviors and default shouting voices as the remote shouting voice;

[0013] and / or,

[0014] The acquiring of the remote shouting voice based on the target monitoring screen includes: determining the object type of the target object based on the target monitoring screen; and acquiring, based on the object type of the target object, a default shouting voice corresponding to the object type of the target object from the correspondence between different object types and default shouting voices as the remote shouting voice.

[0015] Furthermore, the method further comprises:

[0016] Prompt to display the unique identification of the target pole corresponding to the target monitoring screen;

[0017] The step of obtaining a remote voice based on the target monitoring screen includes:

[0018] receiving a start instruction of the target loudspeaker input by a user according to a unique identifier of the target rod; the start instruction is used to start the target loudspeaker;

[0019] The live voice input for the target object is received as the remote shouting voice; the live voice includes a pre-recorded voice and / or a voice input by the user on site.

[0020] Furthermore, the reminder displays the target monitoring screen where the target object appears, including:

[0021] Promptly display the target monitoring screen where the target object appears for the first time;

[0022] Before sending the remote shouting voice to the target loudspeaker of the target pole, the method further includes:

[0023] According to the target monitoring screen, obtaining a unique identifier of the target loudspeaker corresponding to the target object and the target monitoring screen;

[0024] According to the unique identifiers of the target object and the target loudspeaker, it is determined whether to continuously start and lock the target loudspeaker.

[0025] Further, the determining whether to continuously start and lock the target loudspeaker according to the unique identifier of the target object and the target loudspeaker includes:

[0026] In the case that the target object is an animal object, the target loudspeaker is activated and not locked; when the animal object moves to the next pole adjacent to the target pole, the loudspeaker of the next pole is switched on.

[0027] Further, the determining whether to continuously start and lock the target loudspeaker according to the unique identifier of the target object and the target loudspeaker includes:

[0028] In the case where the target object is a person, the first activation is maintained and the target loudspeaker is locked.

[0029] Furthermore, the reminder displays the target monitoring screen where the target object appears, including:

[0030] The target monitoring screen where the target object appears and the unique identifier of the target loudspeaker are displayed; the unique identifier of the target loudspeaker is bound to the target monitoring screen.

[0031] Furthermore, before prompting to display the target monitoring screen where the target object appears in all monitoring screens of the monitoring device, the method further includes:

[0032] Performing alarm identification for frequent alarm events of the same intrusion object to determine whether an alarm is required;

[0033] In case of need for alarm, mobilizing the image acquisition device of at least one rod to track the same said intruding object;

[0034] When the intrusion frequency of the same intruder into the monitoring area reaches the alarm threshold, it is determined as an alarm;

[0035] If the cumulative number of alarms for the same intrusion object in a continuous time period reaches the target alarm threshold of the intrusion object, an alarm is issued.

[0036] The embodiment of the present application provides an intelligent shouting device based on oil pipeline monitoring, which is used to remotely shout at the monitoring area of ​​the oil pipeline, wherein a plurality of poles are arranged at intervals in the monitoring area, and each pole is equipped with a loudspeaker and a group of image acquisition devices; the group of image acquisition devices is used to monitor the poles and the monitoring area;

[0037] The intelligent shouting device based on oil pipeline monitoring includes:

[0038] A target object determination module is used to input the fused video corresponding to each pole into a target object recognition model to obtain the target object; the fused video is obtained by fusing the monitoring images collected by all image acquisition devices of each pole; the target object recognition model is obtained based on training samples and is used to identify different object types; the training samples include sample images of intrusion equipment, sample images of personnel, and sample images of animals;

[0039] A reminder display module, used for reminding and displaying the target monitoring screen where the target object appears in all monitoring screens of the monitoring device; each monitoring screen corresponds to the unique identifier of each pole;

[0040] A voice acquisition module, used to acquire a remote shouting voice based on the target monitoring screen; the remote shouting voice is used to drive away the target object;

[0041] The remote shouting module is used to send the remote shouting voice to the target loudspeaker of the target rod, and control the target loudspeaker to play the remote shouting voice for remote shouting.

[0042] The present application provides an electronic device, comprising one or more processors, for implementing any of the methods described above.

[0043] The present application provides a computer-readable storage medium having a program stored thereon. When the program is executed by a processor, the method described in any one of the above items is implemented.

[0044] The present application provides a computer program product, comprising a computer program / instruction, which implements any of the above methods when executed by a processor.

[0045] In some embodiments, the intelligent shouting method based on oil pipeline monitoring of the present application can quantitatively distinguish the types of intruding equipment and personnel, and timely identify the types of intruding objects of different object types. In addition, the monitoring screen with the target object is promptly reminded, and remote shouting is performed, so that the target object in the monitoring area can be driven away in time, thereby improving the safety of the oil pipeline. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a structural schematic diagram of the application of the intelligent shouting method based on oil pipeline monitoring according to an embodiment of the present application;

[0047] Figure 2 The figure shows a flow chart of an intelligent shouting method based on oil pipeline monitoring according to an embodiment of the present application;

[0048] Figure 3 The figure shows a specific flow chart of the intelligent shouting method based on oil pipeline monitoring according to an embodiment of the present application;

[0049] Figure 4 The figure shows a schematic diagram of the structure of an intelligent shouting device based on oil pipeline monitoring provided in an embodiment of the present application;

[0050] Figure 5 Shown is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0051] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of the present specification. Instead, they are merely examples of devices and methods consistent with some aspects of one or more embodiments of the present specification as detailed in the appended claims.

[0052] It should be noted that: in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the steps included in the method may be more or less than those described in this specification. In addition, a single step described in this specification may be decomposed into multiple steps for description in other embodiments; and multiple steps described in this specification may be combined into a single step for description in other embodiments.

[0053] Figure 1 Shown is a structural schematic diagram of the application of an intelligent shouting method based on oil pipeline monitoring according to an embodiment of the present application.

[0054] like Figure 1 As shown, the intelligent shouting method based on oil pipeline monitoring is applied to the intelligent shouting system based on oil pipeline monitoring in the central control room 11. The intelligent shouting system based on oil pipeline monitoring may include but is not limited to the monitoring equipment 12 in the central control room 11, and a plurality of poles 21 arranged at intervals in the monitoring area, and each pole 21 is installed with a group of image acquisition equipment 22 and a loudspeaker 23.

[0055] The loudspeaker 23 is connected to the monitoring device 12 and is used to play the remote voice. The number of loudspeakers 23 is not limited, and can be one or more (multi-directional broadcast). The loudspeaker 23 is sufficient to transmit the sound within the monitoring area.

[0056] Next, the image acquisition device 22 is used to acquire video images of the monitoring area and the pole 21 on which the image acquisition device 22 is installed, and monitor the monitoring area and the pole 21 on which the image acquisition device 22 is installed. The image acquisition device 22 may include, but is not limited to, a camera, a video camera, and other devices with photo and video recording functions. Other devices with photo and video recording functions include video recorders, etc.

[0057] In this paper, the multiple image acquisition devices 22 arranged on the same pole 21 shoot in different directions, without blind spots, but in cross-monitoring areas. In this way, monitoring of the monitoring area without blind spots is achieved.

[0058] In some examples, a group of image acquisition devices 22 may include but is not limited to more than two image acquisition devices 22. Among them, one of the above two or more image acquisition devices 22 faces the pole 21 to monitor the pole 21 itself, and the remaining image acquisition devices 22 can rotate around the pole 21 to achieve blind spot monitoring of the monitoring area. These image acquisition devices 22 can rotate to reduce the number of image acquisition devices 22.

[0059] In other examples, a group of image acquisition devices 22 may include but is not limited to more than three image acquisition devices 22. Among them, one image acquisition device 22 of the above three or more image acquisition devices 22 monitors toward the pole 21, and the other two or more image acquisition devices 22 monitor toward more than two directions of the monitoring area. These more than two directions can achieve comprehensiveness and safety of monitoring without blind spots. These image acquisition devices 22 do not need to rotate by themselves, reducing outdoor maintenance costs. At the same time, the non-rotating single image acquisition device 22 has a simple structure and relatively low cost.

[0060] The number of all the image acquisition devices 22 mentioned above is, for example, more than one hundred, for example, 105.

[0061] To this end, each image acquisition device 22 is connected to the monitoring device 12 in communication. The monitoring area refers to an area within a specified range including the oil pipeline. The specified range can be set according to user needs.

[0062] In other embodiments, a group of image acquisition devices 22 of each pole 21 is provided with a processor. The processor is used to process the monitoring images acquired by the image acquisition device 22. The processor is connected to the image acquisition device 22. The processor and the image acquisition device 22 are connected to the monitoring device 12 respectively.

[0063] Figure 2 The figure is a flow chart of an intelligent shouting method based on oil pipeline monitoring according to an embodiment of the present application.

[0064] like Figure 2As shown, the intelligent shouting method based on oil pipeline monitoring is used to monitor and remind the monitoring area of ​​the oil pipeline, and the intelligent shouting method based on oil pipeline monitoring may include but is not limited to the following steps 110 to 410:

[0065] Step 110, input the fused video corresponding to each pole into the target object recognition model to output the target object; the fused video is obtained by fusing the monitoring images collected by all image acquisition devices of each pole; the target object recognition model is obtained based on the training sample images and is used to identify different object types; the training sample images include sample images of intrusion devices that break into the monitoring area, sample images of personnel, and sample images of animals.

[0066] The images captured by the on-site image acquisition device are used as monitoring images. After the monitoring images captured by multiple image acquisition devices in more than two directions are fused, a full-view image without blind spots around the current pole can be displayed, thereby improving the comprehensiveness and safety of the monitoring area without blind spots.

[0067] The above-mentioned equipment sample images, personnel sample images and animal sample images are used as target objects that have entered the monitoring area. These intruding equipment, personnel and animals that have entered the monitoring area are also called object types that have entered the monitoring area. In this way, classification and identification are more efficient. These object types are potentially threatening and may endanger the oil pipeline.

[0068] The above-mentioned target object recognition model is trained by sample images of different object types to recognize different object types. Exemplarily, the target object recognition model may include a YOLO (You Only Look Once) model, a Faster R-CNN (Region-based Convolutional Neural Networks, regional convolutional neural network) model, a Mask R-CNN model (Mask Region-based Convolutional Neural Networks, mask region convolutional neural network or instance segmentation region convolutional neural network), a CenterNet (Keypoint Triplet Loss for Object Detection, center point target detection network) model, etc., one or more combinations thereof, which are not listed one by one here.

[0069] In some examples, the image acquisition device may be a camera or a video camera, and the number thereof may be, but is not limited to, three. One of the image acquisition devices monitors the pole itself, and the other two image acquisition devices are arranged in two directions facing the monitoring area with their backs facing each other, thereby avoiding the obstruction of the pole. The number of image acquisition devices is determined by the cost of the number of image acquisition devices and the requirement of monitoring without blind spots. These image acquisition devices do not need to rotate by themselves, reducing outdoor maintenance costs. At the same time, a single non-rotating image acquisition device has a simple structure and relatively low cost.

[0070] Step 210, in all monitoring screens of the monitoring device, the target monitoring screen where the target object appears is displayed; each monitoring screen corresponds to the unique identifier of each pole. For example, each monitoring screen is bound to the unique identifier of each pole. For another example, each monitoring screen corresponds to the unique identifier of each pole one by one. All these monitoring screens can be patrolled in a multi-screen small image manner, or each display window can be set to display all monitoring screens.

[0071] The alarm events in this article are used to represent events that pose a potential threat to the oil pipeline. The above alarm events include alarm events of intrusion into target objects such as equipment, personnel, and animals, and the corresponding alarm events of intrusion behavior, as well as alarm events corresponding to intrusion into equipment and personnel not following the standard operation. In this way, various factors that lead to sudden third-party construction, heavy vehicle crushing, and man-made damage in these important areas are reduced, and the great pressure on the safe operation of oil and gas pipelines is reduced. Standardized action operations are used to ensure the safe operation of oil pipelines. For example, wearing prescribed work clothes and performing prescribed behavior operations.

[0072] In the related technologies, first, the construction unit did not inform the pipeline company and privately carried out mechanical construction on the pipeline, threatening the pipeline safety. Second, the construction personnel knew the pipeline conditions and carried out rough construction during the construction process, threatening the pipeline safety. Third, the construction site monitoring was not in place, and the construction personnel did not follow the plan to damage the pipeline. In this way, the pipeline safety is threatened.

[0073] In this regard, the above-mentioned intrusion equipment may include but is not limited to third-party construction equipment and / or mechanical construction equipment. In this way, this article studies the intrusion equipment, personnel, and animals that have entered the monitoring area, and can identify these target objects. This reduces the potential threats to the oil pipeline caused by relatively frequent personnel activities and third-party construction operations. In addition, when these factors that cause potential dangers appear in the oil pipeline, a reminder is issued to reduce the great safety risks to the oil pipeline and the environment in the area.

[0074] In the related art, the central control room monitoring uses one screen, but there are hundreds of monitoring cameras. In this way, the amount of monitoring screen data is large, and it is very likely to miss the abnormal situation in the pipeline.

[0075] Compared with the related art, the reminder in this article displays the monitoring screen corresponding to the video screen where the target object is located. For example, the monitoring screen flashes. For another example, a prominent color is displayed around the monitoring screen. In this way, it is possible to promptly determine in which monitoring screen the target object appears, and display the corresponding reminder on the monitoring device. In this way, the video screen where the target object is located can be displayed, and the monitoring screen that needs more effective monitoring can be selected. This minimizes the possibility of missing abnormal situations in the pipeline.

[0076] In this way, in the above step 210, a reminder display of the monitoring screen of the central control room can be performed to remind the monitoring personnel in the central control room to timely discover potential threats to the oil pipeline.

[0077] These monitoring screens with target objects are called target monitoring screens.

[0078] Step 310, based on the target monitoring screen, obtain the remote shouting voice; the remote shouting voice is used to drive away the target object.

[0079] The mobilization of the image acquisition device of at least one rod in the above step 310 may be, but is not limited to, mobilizing at least one image acquisition device of one rod or mobilizing at least one image acquisition device of multiple rods to track the same target object.

[0080] The above-mentioned object that is tracked and enters the monitoring area is called the target object. The remote shouting voice is used to shout at the target object in the monitoring area.

[0081] Step 410, sending a remote shouting voice to a target loudspeaker at a target pole, and controlling the target loudspeaker to play the remote shouting voice to perform remote shouting.

[0082] The target object is within the monitoring range of a certain pole, which is called the target pole. At the same time, the loudspeaker of the pole is mobilized, which is called the target loudspeaker.

[0083] In the embodiment of the present application, the types of intruding equipment and personnel are quantitatively distinguished, and the types of intruding objects of different object types are timely identified. In addition, the monitoring screen with the target object is promptly reminded, and a remote call is made, so that the target object in the monitoring area can be driven away in time, thereby improving the safety of the oil pipeline.

[0084] Combination Figure 2 As shown, the remote shouting voice may include, but is not limited to, a pre-set voice and / or a voice manually input by the central control room. Detailed description is as follows.

[0085] Combination Figure 2 As shown, in the embodiment of the pre-set voice, the above step 310 may include but is not limited to the following steps 1 and 2:

[0086] The first step is to determine the dangerous behavior of the target object based on the target monitoring screen.

[0087] The target objects that break into the monitoring area are potentially threatening and may endanger the oil pipeline. These target objects and their behaviors will be monitored. Some behaviors that may endanger the oil pipeline are called dangerous behaviors of the target objects. For example, the target object is holding a shovel. Another example is that the target object digs the ground.

[0088] The second step is to obtain, based on the dangerous behavior of the target object, the default shouting voice corresponding to the dangerous behavior of the target object from the corresponding relationship between different dangerous behaviors and the default shouting voice as the remote shouting voice.

[0089] The above-mentioned default shouting voice may be a pre-recorded voice for different dangerous behaviors.

[0090] Next, the correspondence between the above different dangerous behaviors and the default shouting voice is used to indicate that different default shouting voices are set for different dangerous behaviors. Example 1: The dangerous behavior may be an animal breaking into the monitoring area, and the default shouting voice is, for example, "go, go, go".

[0091] Example 2: The target object's dangerous behavior may be, for example, breaking into a surveillance area, and the default voice message may be, for example, "Please leave as soon as possible."

[0092] Example 3: The target object’s dangerous behavior may be not wearing a helmet, and the default voice message may be “Please wear a helmet”.

[0093] Furthermore, the above-mentioned step 1 may include but is not limited to: first, based on the target monitoring screen, determine the object type of the target object; second, based on the object type of the target object, from the correspondence between different object types and default shouting voices, obtain the default shouting voice corresponding to the object type of the target object as the remote shouting voice.

[0094] In the embodiment of the present application, targeted reminders can be given for different dangerous behaviors.

[0095] Figure 3 The figure shows a specific flow chart of the intelligent shouting method based on oil pipeline monitoring according to an embodiment of the present application.

[0096] In such Figure 3 In the illustrated embodiment, the intelligent shouting method based on oil pipeline monitoring may further include but is not limited to step 211, and the above step 310 may include but is not limited to the following steps 311 and 312:

[0097] Step 211, prompting and displaying the unique identifier of the target pole corresponding to the target monitoring screen.

[0098] The unique identification of the target rod is used to indicate the uniqueness of the target rod corresponding to the monitoring screen with the target object. The unique identification of the target rod can be but is not limited to the number or name of the target rod.

[0099] Step 311, receiving a target loudspeaker activation instruction input by the user according to the unique identifier of the target pole; the activation instruction is used to activate the target loudspeaker. In this way, the user can verify the situation of the monitoring area on site, so as to activate the effectiveness of the remote shouting function.

[0100] Step 312, receiving live voice input for the target object as the remote shouting voice; the live voice includes pre-recorded voice and / or voice input by the user on site.

[0101] The pre-recorded voice is, for example, a fixed voice recorded by a recording speaker. The fixed voice is played in front of a microphone in the control room as a live voice to be played on a target loudspeaker at a remote end. In this way, the user can be freed and repeated broadcasting is convenient.

[0102] In the embodiment of the present application, the existence of the target object is manually verified, thereby manually assisting in monitoring and controlling the remote shouting.

[0103] Continue to see Figure 2 and Figure 3 As shown, the reminder displaying the target monitoring screen where the target object appears in the above step 210 may include but is not limited to the following steps: reminding displaying the target monitoring screen where the target object appears for the first time.

[0104] The monitoring screen where the target object appears for the first time is used as the target monitoring screen and the reminder display is started. In this way, the target object can be found in time. When the target object in the target monitoring screen appears for the first time, it is necessary to continue tracking the target object. If the target object moves very fast, you can shout at the scene to remind the target object. When it is necessary to track the target object, control the nearest loudspeaker on the scene (that is, the loudspeaker on the pole corresponding to the monitoring screen) to turn on.

[0105] Before the above step 410, the method may also include, but is not limited to, the following steps (1) to (2):

[0106] (1) According to the target monitoring screen, a unique identifier of the target object and the target loudspeaker corresponding to the target monitoring screen is obtained.

[0107] The unique identification of the target loudspeaker is used to indicate the uniqueness of the loudspeaker on the target pole corresponding to the monitoring screen of the target object. The unique identification of the target loudspeaker may be, but is not limited to, the number or name of the target loudspeaker.

[0108] (2) Determine whether to continue to activate and lock the target loudspeaker based on the unique identifier of the target object and the target loudspeaker.

[0109] In the embodiment of the present application, the target loudspeaker can be continuously started and locked through the unique identification of the target object and the target loudspeaker, so as to track and remind the target object.

[0110] The above step (2) can be implemented by the following multiple optional embodiments:

[0111] In the first optional embodiment of the above step (2), when the target object is an animal object, the target loudspeaker is activated and not locked; when the animal object moves to the next pole adjacent to the target pole, the loudspeaker of the next pole is switched and activated. Since the animal moves fast, remote shouting can be carried out by switching the target loudspeaker.

[0112] In the application example of this embodiment, when the target object appears in the monitoring range of the No. 1 pole, it will appear in the first target monitoring screen for the first time. At this time, a reminder will be displayed for the first target monitoring screen. At the same time, the target loudspeaker of the first pole will be turned on as the target loudspeaker for remote shouting. Then, the target object continues to move and appears in the monitoring range of the No. 2 pole, correspondingly appearing in the second target monitoring screen, and the second target monitoring screen will continue to be tracked and reminded. At the same time, the target loudspeaker of the second pole will be turned on as the target loudspeaker for remote shouting. And so on, until the target object is tracked to leave the monitoring area, the target loudspeaker and target monitoring screen locked by the target object will be unlocked.

[0113] In the second optional embodiment of the above step (2), the target loudspeaker is kept activated for the first time and locked, and the loudspeaker of the next pole adjacent to the above target pole is not switched. Since the sound of the loudspeaker will spread to the surroundings, it is possible to switch the loudspeaker without tracking the movement of the target object, and remote shouting can also be achieved. The next pole adjacent to the above target pole refers to the monitoring range where the next pole to be reached in the moving direction of the target object is located.

[0114] Furthermore, if the target object is a person, the target loudspeaker is kept activated for the first time and locked; the unique identifier of the target loudspeaker is bound to the target object. For example, if the target object is determined to be a person and the moving speed of the target object is less than the speed threshold, the target loudspeaker is kept activated for the first time and locked.

[0115] It should be noted that the target object moving speed less than the speed threshold is used to indicate that the person is moving slowly, such as walking or running. The speed threshold can be set according to user needs. The speed threshold can be greater than 0.5 km / h and less than 20 km / h. As long as the speed threshold is used to indicate that the person is not moving with the help of external transportation, it belongs to the protection scope of the embodiment of the present application and is not limited here. For people using external transportation, this scene can be identified as a category of intrusion into the device.

[0116] Since people move slowly while animals move quickly, and the sound of the loudspeaker will spread to the surroundings, people can hear the remote calls from the central control room. Therefore, it is not necessary to switch the loudspeaker with the movement of the target object.

[0117] In another embodiment, the reminder to display the target monitoring screen where the target object appears in step 210 may include, but is not limited to, the following steps: reminding to display the target monitoring screen where the target object appears and the unique identifier of the target loudspeaker; binding the unique identifier of the target loudspeaker to the target monitoring screen. For example, the monitoring screen is bound to the loudspeaker in advance, so that it is convenient to remind the target monitoring screen where the target object appears, and also remind to display the unique identifier of the target loudspeaker. In this way, as the target object moves, the target loudspeaker can be switched to bind while reminding to display the target monitoring screen, so as to track the target object and make a remote call.

[0118] Combination Figure 1 , Figure 2 and Figure 3 As shown, before prompting to display the target monitoring screen where the target object appears in all monitoring screens of the monitoring device, the method also includes:

[0119] The alarm events in this article are used to represent events that pose a potential threat to the oil pipeline. The above alarm events include alarm events of intrusion into target objects such as equipment, personnel, and animals, and alarm events corresponding to their intrusion behaviors, as well as alarm events corresponding to intrusion into equipment and personnel not following standard actions. In this way, various factors that lead to sudden third-party construction, heavy vehicle crushing, and man-made damage in these important areas are reduced, and the great pressure on the safe operation of oil and gas pipelines is reduced. Standardized action operations are used to ensure the safe operation of oil pipelines. For example, wearing prescribed work clothes and performing prescribed behavior operations.

[0120] In the related technologies, first, the construction unit did not inform the pipeline company and privately carried out mechanical construction on the pipeline, threatening the pipeline safety. Second, the construction personnel knew the pipeline conditions and carried out rough construction during the construction process, threatening the pipeline safety. Third, the construction site monitoring was not in place, and the construction personnel did not follow the plan to damage the pipeline. In this way, the pipeline safety is threatened.

[0121] In this regard, the above-mentioned intrusion equipment may include but is not limited to third-party construction equipment and / or mechanical construction equipment. In this way, this article studies the intrusion equipment, personnel, and animals that intrude into the monitoring area, and can identify these target objects. This reduces the potential threats to the oil pipeline caused by relatively frequent personnel activities and third-party construction operations. In addition, when these potential dangerous factors appear in the oil pipeline, an alarm is issued to reduce the great safety risks to the oil pipeline and the environment in the area.

[0122] After step 210, the method may further include but is not limited to the following steps ① to ⑤:

[0123] ①. Identify the frequent occurrence of alarm events for the same target object and determine whether an alarm is needed;

[0124] ②. If no alarm is needed, return to step 110 to continue execution.

[0125] ③. When an alarm is required, mobilize the image acquisition device of at least one pole to track the same target object.

[0126] The mobilization of the image acquisition device of at least one rod in the above step ③ may be, but is not limited to, mobilizing at least one image acquisition device of one rod or mobilizing at least one image acquisition device of multiple rods to track the same target object.

[0127] ④. When the frequency of intrusion of the same target object into the monitoring area reaches the alarm threshold, it is determined as an alarm.

[0128] The above-mentioned different frequency levels are used to indicate the number of occurrences of alarm events corresponding to different object types per unit time.

[0129] In addition, the above alarm threshold is used to indicate the upper limit of the number of times an alarm is generated. The above alarm threshold can be set according to user needs or can be automatically adjusted.

[0130] For example, the above method may also include but is not limited to the following two steps: In the first step, based on the historical intrusion frequency of the same object type, if the intrusion frequency of the object type intruding into the monitoring area is higher than the first preset frequency, the current alarm threshold is increased. In the second step, based on the historical intrusion frequency of the same object type, if the intrusion frequency of the object type intruding into the monitoring area is lower than the second preset frequency, the current alarm threshold is decreased.

[0131] Exemplarily, when the object type is a person type, the greater the frequency of the person breaking into the monitoring area, the greater the alarm threshold.

[0132] For frequent intrusions (also called frequent alarms) or events, the alarm interval can be automatically adjusted to ensure that no alarms are missed. For example, the default setting is 3 helmet alarms within 5 minutes. If in subsequent scenarios, an average of about 20 alarms occur within 5 minutes, the system will automatically adjust the alarm threshold to 15 logs within 5 minutes before an alarm is issued.

[0133] In the embodiment of the present application, the alarm interval can be automatically adjusted to ensure that no alarm is missed and to improve the effectiveness of the alarm.

[0134] ⑤. If the cumulative number of alarms for the same target object within a continuous time period reaches the target alarm threshold of the target object, an alarm will be issued.

[0135] The continuous time period may be set according to user needs. For example, the continuous time period may be one hour, several hours, half a day, or one day.

[0136] In addition, the target alarm threshold is used to indicate the upper limit of the alarm. The target alarm threshold refers to the alarm threshold of the target object. The target alarm threshold of the target object can be set according to user needs or can be automatically adjusted.

[0137] For example, the above method may also include but is not limited to that the target alarm threshold may be determined by using at least one of the following optional embodiments:

[0138] In the first embodiment of the target alarm threshold determination process, according to the type of the target object, different alarm thresholds are set for different object types to determine the target alarm threshold corresponding to the target object.

[0139] Due to the intrusion behaviors of target objects such as intrusion equipment, personnel and animals, and the failure of target objects, intrusion equipment and personnel to operate in accordance with the standard, the potential risks to the oil pipeline are different. Therefore, different alarm thresholds are set for different object types to distinguish different object types and make more targeted alarms.

[0140] For example, animals are less likely to break into the monitoring area, but they are more dangerous. Therefore, the alarm threshold corresponding to the type of animal is lower than the alarm threshold corresponding to the type of intruder equipment and personnel. In this way, the sensitivity of the type of animal can be used to detect the potential danger caused by the animal in time.

[0141] In the second embodiment of the target alarm threshold determination process, different alarm thresholds are set for different object types. First, if the current time period is determined to be a peak time period, the target alarm threshold of the target object is increased to the alarm threshold corresponding to the peak time period.

[0142] Secondly, if it is determined that the current time period is a non-peak time period, the target alarm threshold of the target object is adjusted down to the alarm threshold corresponding to the non-peak time period; the alarm threshold corresponding to the non-peak time period is less than the alarm threshold corresponding to the peak time period;

[0143] Among them, the alarm threshold corresponding to the peak time period and the alarm threshold for the non-peak time period are determined by the following steps: counting the time period when the alarm frequency occurs, and determining the alarm threshold for the peak time period and the alarm threshold for the non-peak time period corresponding to the alarm frequency.

[0144] The time period when the alarm frequency increases sharply (for example, the alarm frequency increases to the maximum) is called the peak time period. The other time periods except the peak time period are called non-peak time periods.

[0145] In the application example of the second embodiment described above, the same or different image acquisition devices are bound with corresponding AI algorithms, and each camera + algorithm is a unique object. The alarm threshold can be set for the alarm event of the target camera + algorithm corresponding to the target object.

[0146] For example, in an image acquisition device with a human intrusion algorithm, it is preset that 5 human intrusions within 3 minutes will be regarded as an alarm. However, after the later trial operation, it was found that this alarm threshold was affected by the time period, and the alarm accuracy rate varied greatly. For example, during the peak time period from 8:00 to 9:00 in the morning, it is more reasonable to set this alarm threshold to around 8 to 10, and it is more reasonable to set the alarm threshold corresponding to the non-peak time period to around 3 to 6. In this way, we can configure this alarm threshold at any time in the background to improve the alarm accuracy in different time periods.

[0147] Combined with the application examples of the first and second embodiments, the daytime scene is the peak time period, mainly focusing on monitoring construction machinery, and it is more reasonable to set the corresponding alarm threshold for construction machinery equipment to about 8 to 10. The nighttime scene is the non-peak time period, mainly focusing on monitoring vans, motorcycles, oil drums, and personnel stay, and it is more reasonable to set the corresponding alarm threshold for vans, motorcycles, oil drums, and personnel stay to about 3 to 6. In this way, the daytime scene and the night scene are monitored separately, so as to conduct comprehensive intelligent diagnosis and analysis to increase the accuracy of identifying external force damage to pipeline behavior.

[0148] In the embodiment of the present application, the target alarm threshold is adjusted with different time periods. Thus, increasing the target alarm threshold during the peak time period can reduce the alarm frequency, thereby improving the effectiveness of monitoring and reminders. At the same time, decreasing the target alarm threshold during the peak time period can increase the alarm frequency, thereby improving the effectiveness of monitoring and reminders. At the same time, comprehensive intelligent diagnosis and analysis are performed to increase the accuracy of identifying external force damage to pipeline behavior.

[0149] The alarm in step ⑤ of this article can be implemented by at least one of the following optional methods. In one optional method, the monitoring area is controlled to perform an audible and visual alarm. In this way, it is used to remind the scene in the monitoring area. In another optional method, the monitoring screen of the central control room is alarmed. In this way, it is used to remind the monitoring personnel in the central control room to timely discover potential threats to the oil pipeline.

[0150] In the embodiment of the present application, the types of intruding devices and personnel are quantitatively distinguished, and the alarm events of intrusions of different object types are identified in a timely manner. In addition, the monitoring alarm efficiency is ensured to be above 9%, the monitoring identification accuracy is above 95%, and the alarm efficiency can be continuously improved to reduce the amount of useless data. At the same time, after the alarm is triggered, it is convenient to focus on displaying the monitoring screen for subsequent manual or automatic verification and remote shouting.

[0151] Based on the same application concept as the above method, the embodiment of the present application also provides an intelligent shouting device based on oil pipeline monitoring, such as Figure 4 As shown, the intelligent shouting device based on oil pipeline monitoring is used to remotely shout at the monitoring area of ​​the oil pipeline. A plurality of poles are arranged at intervals in the monitoring area, and each pole is equipped with a group of image acquisition equipment and a loudspeaker; a group of image acquisition equipment is used to monitor the poles and the above-mentioned monitoring area.

[0152] Based on this, the intelligent shouting device based on oil pipeline monitoring can include the following modules:

[0153] The target object determination module 71 is used to input the fused video corresponding to each pole into the target object recognition model to obtain the target object; the fused video is obtained by fusing the monitoring images collected by all the image acquisition devices of each pole; the target object recognition model is obtained based on the training samples and is used to identify different object types; the training samples include sample images of intrusion equipment, sample images of personnel and sample images of animals;

[0154] The reminder display module 72 is used to remind and display the target monitoring screen where the target object appears in all monitoring screens of the monitoring device; each monitoring screen corresponds to the unique identifier of each pole;

[0155] The voice acquisition module 73 is used to acquire the remote shouting voice based on the target monitoring screen; the remote shouting voice is used to drive away the target object;

[0156] The remote shouting module 74 is used to send a remote shouting voice to the target loudspeaker at the target pole, and control the target loudspeaker to play the remote shouting voice for remote shouting.

[0157] As an embodiment, the above device further comprises: an identification reminder display module, which is used to remind and display the unique identification of the target pole corresponding to the target monitoring screen;

[0158] The voice acquisition module 73 is specifically used to receive the start-up instruction of the target loudspeaker input by the user according to the unique identifier of the target pole; the start-up instruction is used to turn on the target loudspeaker; and, receive the on-site voice input for the target object as the remote shouting voice; the on-site voice includes a pre-recorded voice and / or a voice input by the user on-site.

[0159] As an embodiment, the above device further includes: a reminder display module 72, which is used to remind and display the target monitoring screen when the target object appears for the first time;

[0160] The above-mentioned device further comprises: an identification acquisition module, which is used to acquire the unique identification of the target loudspeaker corresponding to the target object and the target monitoring screen according to the target monitoring screen before sending the remote shouting voice to the target loudspeaker of the target pole;

[0161] The continuous locking determination module is used to determine whether to continuously start and lock the target loudspeaker according to the unique identifiers of the target object and the target loudspeaker.

[0162] As an embodiment, the above device also includes:

[0163] The alarm identification module is used to identify the frequent occurrence of alarm events of the same target object before reminding to display the target monitoring screen where the target object appears in all monitoring screens of the monitoring device, and determine whether an alarm is needed;

[0164] A tracking module, for mobilizing at least one rod image acquisition device to track the same target object when an alarm is required;

[0165] An alarm determination module is used to determine an alarm when the intrusion frequency of the same target object into the monitoring area reaches an alarm threshold;

[0166] The alarm module is used to generate an alarm if the cumulative number of alarms for the same target object within a continuous time period reaches the target alarm threshold of the target object.

[0167] The implementation process of the functions and effects of each module in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and the same technical effect can be achieved, which will not be repeated here.

[0168] An embodiment of the present application provides an electronic device, including the above-mentioned intelligent shouting device based on oil pipeline monitoring.

[0169] The intelligent shouting method based on oil pipeline monitoring in the embodiment of the present application is applied to an electronic device. The electronic device may be a monitoring device. The monitoring device may include a PC (Personal Computer) terminal device. The PC terminal device may include but is not limited to a desktop computer, a tablet computer or a laptop computer.

[0170] Figure 5 Shown is a schematic diagram of the structure of an electronic device 90 provided in an embodiment of the present application.

[0171] like Figure 5 As shown, the electronic device 90 includes one or more processors 91 for implementing the intelligent shouting method based on oil pipeline monitoring as described above.

[0172] In some embodiments, the electronic device 90 may include a storage medium 99. For example, the computer-readable storage medium may store a program that can be called by the processor 91, and may include a non-volatile storage medium. In some embodiments, the electronic device 90 may include a memory 98 and an interface 97. In some embodiments, the electronic device 90 may also include other hardware according to actual applications.

[0173] The computer-readable storage medium of the embodiment of the present application stores a program thereon, which, when executed by the processor 91, is used to implement the intelligent shouting method based on oil pipeline monitoring as described above.

[0174] The present application provides a computer program product, comprising a computer program / instruction, which implements any of the above methods when executed by a processor.

[0175] The present application also provides a computer program, which is stored in a computer-readable storage medium, such as Figure 5 The computer program is stored in the storage medium 99, and when the processor executes the computer program, it causes the processor 91 to execute the method described above.

[0176] The present application may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer-readable 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, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0177] The above description is only a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this specification should be included in the scope of protection of this specification.

[0178] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the phrase "includes a ..." defines an element, and does not exclude the presence of other identical elements in the process, method, commodity or device including the element.

Claims

1. An intelligent shouting method based on oil pipeline monitoring, characterized in that: Used to remotely announce to the monitoring area of ​​the oil pipeline, wherein a plurality of poles are arranged at intervals in the monitoring area, and each pole is equipped with a loudspeaker and a set of image acquisition equipment; The set of image acquisition devices is used to monitor the pole and monitor the monitoring area; The intelligent shouting method based on oil pipeline monitoring includes: The fused video corresponding to each pole is input into the target object recognition model to obtain the target object; the fused video is obtained by fusing the monitoring images collected by all image acquisition devices of each pole; the target object recognition model is obtained based on training samples and is used to identify different object types; the training samples include sample images of intrusion equipment, sample images of personnel, and sample images of animals; Among all monitoring screens of the monitoring device, a target monitoring screen where the target object appears is displayed; each monitoring screen corresponds to a unique identifier of each pole; Based on the target monitoring screen, a remote shouting voice is obtained; the remote shouting voice is used to drive away the target object; The remote shouting voice is sent to the target loudspeaker of the target pole, and the target loudspeaker is controlled to play the remote shouting voice to perform remote shouting.

2. The intelligent shouting method based on oil pipeline monitoring according to claim 1 is characterized in that: The step of obtaining a remote shouting voice based on the target monitoring screen includes: determining a target dangerous behavior of the target object based on the target monitoring screen; and obtaining a default shouting voice corresponding to the dangerous behavior of the target object from a correspondence between different dangerous behaviors and default shouting voices based on the dangerous behavior of the target object as the remote shouting voice; and / or, The acquiring of the remote shouting voice based on the target monitoring screen includes: determining the object type of the target object based on the target monitoring screen; and acquiring, based on the object type of the target object, a default shouting voice corresponding to the object type of the target object from the correspondence between different object types and default shouting voices as the remote shouting voice.

3. The intelligent shouting method based on oil pipeline monitoring according to claim 1 is characterized in that: The method further comprises: Prompt to display the unique identification of the target pole corresponding to the target monitoring screen; The step of obtaining a remote voice based on the target monitoring screen includes: receiving a start instruction of the target loudspeaker input by a user according to a unique identifier of the target rod; the start instruction is used to start the target loudspeaker; The live voice input for the target object is received as the remote shouting voice; the live voice includes a pre-recorded voice and / or a voice input by the user on site.

4. The intelligent shouting method based on oil pipeline monitoring according to claim 1 is characterized in that: The reminder displays the target monitoring screen where the target object appears, including: Promptly display the target monitoring screen where the target object appears for the first time; Before sending the remote shouting voice to the target loudspeaker of the target pole, the method further includes: According to the target monitoring screen, obtaining a unique identifier of the target loudspeaker corresponding to the target object and the target monitoring screen; According to the unique identifiers of the target object and the target loudspeaker, it is determined whether to continuously start and lock the target loudspeaker.

5. The intelligent shouting method based on oil pipeline monitoring according to claim 4 is characterized in that: The determining whether to continuously start and lock the target loudspeaker according to the unique identifier of the target object and the target loudspeaker comprises: In the case that the target object is an animal object, the target loudspeaker is activated and not locked; when the animal object moves to the next pole adjacent to the target pole, the loudspeaker of the next pole is switched on.

6. The intelligent shouting method based on oil pipeline monitoring according to claim 4 is characterized in that: The determining whether to continuously start and lock the target loudspeaker according to the unique identifier of the target object and the target loudspeaker comprises: In the case where the target object is a person, the first activation is maintained and the target loudspeaker is locked.

7. The intelligent shouting method based on oil pipeline monitoring according to claim 1 is characterized in that: The reminder displays the target monitoring screen where the target object appears, including: The target monitoring screen where the target object appears and the unique identifier of the target loudspeaker are displayed; the unique identifier of the target loudspeaker is bound to the target monitoring screen.

8. The intelligent shouting method based on oil pipeline monitoring according to claim 1 is characterized in that: Before prompting to display the target monitoring screen where the target object appears in all monitoring screens of the monitoring device, the method further includes: Performing alarm identification for frequent alarm events of the same intrusion object to determine whether an alarm is required; In case of need for alarm, mobilizing the image acquisition device of at least one rod to track the same said intruding object; When the intrusion frequency of the same intruder into the monitoring area reaches the alarm threshold, it is determined as an alarm; If the cumulative number of alarms for the same intrusion object in a continuous time period reaches the target alarm threshold of the intrusion object, an alarm is issued.

9. An intelligent shouting device based on oil pipeline monitoring, characterized in that: Used to remotely announce to the monitoring area of ​​the oil pipeline, wherein a plurality of poles are arranged at intervals in the monitoring area, and each pole is equipped with a loudspeaker and a set of image acquisition equipment; The set of image acquisition devices is used to monitor the pole and monitor the monitoring area; The intelligent shouting device based on oil pipeline monitoring includes: A target object determination module is used to input the fused video corresponding to each pole into a target object recognition model to obtain the target object; the fused video is obtained by fusing the monitoring images collected by all image acquisition devices of each pole; the target object recognition model is obtained based on training samples and is used to identify different object types; the training samples include sample images of intrusion equipment, sample images of personnel, and sample images of animals; A reminder display module, used for reminding and displaying the target monitoring screen where the target object appears in all monitoring screens of the monitoring device; each monitoring screen corresponds to the unique identifier of each pole; A voice acquisition module, used to acquire a remote shouting voice based on the target monitoring screen; the remote shouting voice is used to drive away the target object; The remote shouting module is used to send the remote shouting voice to the target loudspeaker of the target rod, and control the target loudspeaker to play the remote shouting voice for remote shouting.

10. An electronic device, characterized in that: It includes one or more processors, used to implement the intelligent shouting method based on oil pipeline monitoring as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that: A program is stored thereon, and when the program is executed by the processor, the intelligent shouting method based on oil pipeline monitoring as described in any one of claims 1-8 is implemented.

12. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the intelligent shouting method based on oil pipeline monitoring described in any one of claims 1-8 is implemented.

Images