Alarm method, device and equipment in electrified scene, medium and product

Abstract

The invention provides an alarm method and device in a live-line scene, equipment, a medium and a product. The method comprises the steps that if it is determined that an object exists in a preset range of the electrified equipment, category information of the object is determined; determining feature information of the object according to the category information of the object; wherein the feature information represents the current state of the object; generating alarm information according to the feature information of the object; wherein the alarm information is used for prompting the object to leave the live scene. According to the method provided by the invention, potential safety hazards in a live-line scene are reduced.

Description

Technical Field

[0001] This application relates to the field of electrification, and particularly to an alarm method, device, equipment, medium and product in an electrified scenario. Background Art

[0002] Live working and operations such as maintenance of substation equipment are routine operations in electrified scenarios or power scenarios. However, due to various reasons, various safety accidents may occur in electrified scenarios, such as electric shock to personnel, short circuit of electrical equipment, etc.

[0003] Currently, the experience of staff is used to prevent safety accidents and reduce their occurrence, but the cost of manual operation and maintenance is high, and the latency is high.

[0004] Therefore, there is a problem of great potential safety hazards in electrified scenarios. Summary of the Invention

[0005] This application provides an alarm method, device, equipment, medium and product in an electrified scenario to solve the technical problem of great potential safety hazards in electrified scenarios.

[0006] In a first aspect, this application provides an alarm method in an electrified scenario, including:

[0007] If it is determined that there is an object within a preset range of an electrified device, determine the category information of the object;

[0008] According to the category information of the object, determine the characteristic information of the object; wherein, the characteristic information characterizes the current state of the object;

[0009] Generate an alarm information according to the characteristic information of the object; wherein, the alarm information is used to prompt the object to leave the electrified scenario.

[0010] Optionally, in the method as described above, the category information is the first category; according to the category information of the object, determining the characteristic information of the object includes:

[0011] Determine the distance information between the object and a preset radar device through the preset radar device; wherein, the radar device is installed in the electrified scenario;

[0012] Determine the distance information as the characteristic information of the object.

[0013] Optionally, in the method as described above, determining the distance information between the object and the radar device through the preset radar device includes:

[0014] Determine the time information and frequency information of the radar signal through the radar signal emitted by the radar device; wherein, the time information characterizes the duration from when the radar signal is sent to the object and returns to the radar device, and the frequency information characterizes the frequency change of the radar signal;

[0015] Determine the distance information between the object and the radar device according to the time information and the frequency information.

[0016] Optionally, as in the above method, generate an alarm message according to the characteristic information of the object, including:

[0017] If the distance information is less than a preset distance threshold, generate an alarm message based on a preset information template.

[0018] Optionally, as in the above method, the category information is the second category; determine the characteristic information of the object according to the category information of the object, including:

[0019] Obtain the voiceprint information of the object through a preset voiceprint collector, and obtain the image information of the object through a preset image acquisition device;

[0020] Determine the type information of the object according to the voiceprint information and the image information of the object; wherein, the type information is a sub-category of the second category;

[0021] Determine the type information of the object as the characteristic information of the object.

[0022] Optionally, as in the above method, determine the type information of the object according to the voiceprint information and the image information of the object, including:

[0023] Perform feature extraction processing on the voiceprint information and the image information of the object according to a preset feature extraction network to obtain a fused feature; wherein, the fused feature represents the voiceprint information and the image information of the object;

[0024] Input the fused feature and the preset prompt word information into a preset large language model to obtain the output type information of the object.

[0025] Optionally, as in the above method, the preset feature extraction network includes a first extraction network and a plurality of second extraction networks; perform feature extraction on the voiceprint information of the object and the image information of the animal according to the preset feature extraction network to obtain a fused feature process, including:

[0026] Input the voiceprint information of the object into the first extraction network to obtain a voiceprint feature;

[0027] Input the image information of the object into each second extraction network respectively to obtain a plurality of image features; wherein, the sizes of the convolution kernels in each second extraction network are different, and the second extraction network corresponds to the image feature one by one;

[0028] Perform a first fusion process on the voiceprint feature and each image feature respectively to obtain a plurality of initial features; wherein, the initial feature represents the voiceprint feature and the image feature, and the number of initial features is the same as the number of image features;

[0029] Perform a second fusion process on multiple initial features to obtain fused features.

[0030] Optionally, as in the above method, generate an alarm message according to the feature information of the object, including:

[0031] According to the type information of the object, determine the expulsion plan of the object based on a preset association relationship; wherein, the preset association relationship represents the association relationship between the type information and the expulsion plan, and the expulsion plan represents the method of prompting the object to leave.

[0032] Generate an alarm message according to the expulsion plan of the object.

[0033] Optionally, as in the above method, if it is determined that there is an object within the preset range of the live equipment, determine the category information of the object, including:

[0034] Obtain an infrared thermal image within the preset range through a preset infrared induction device; wherein, the infrared thermal image represents the temperature distribution within the preset range.

[0035] If it is determined that there is a heat source in the infrared thermal image, it is determined that there is an object within the preset range of the live equipment; wherein, the temperature value of the heat source is greater than the preset temperature threshold.

[0036] Determine the category information of the object according to the area of the heat source in the infrared thermal image.

[0037] Optionally, as in the above method, determine the category information of the object according to the area of the heat source in the infrared thermal image, including:

[0038] If the area of the heat source in the infrared thermal image is greater than the preset area threshold, determine that the category information of the object is the first category.

[0039] If the area of the heat source in the infrared thermal image is less than or equal to the preset area threshold, determine that the category information of the object is the second category.

[0040] In a second aspect, the present application provides an alarm device in a live scenario, including:

[0041] A first determination unit, configured to determine the category information of the object if it is determined that there is an object within the preset range of the live equipment.

[0042] A second determination unit, configured to determine the feature information of the object according to the category information of the object; wherein, the feature information represents the current state of the object.

[0043] An alarm unit, configured to generate an alarm message according to the feature information of the object; wherein, the alarm message is used to prompt the object to leave the live scenario.

[0044] In a third aspect, the present application provides an electronic device, including: a memory and a processor;

[0045] The memory stores computer-executable instructions;

[0046] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementation manners of the first aspect.

[0047] In a fourth aspect, the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the above first aspect and / or various possible implementation manners of the first aspect.

[0048] In a fifth aspect, the present application provides a computer program product, including: a computer program, which implements the above first aspect and / or various possible implementation manners of the first aspect when executed by a processor.

[0049] For the warning method, device, equipment, medium and product in the live electrical scenario provided by the present application, if it is determined that there is an object within the preset range of the live electrical equipment, the category information of the object is determined, and according to the category information of the object, the characteristic information of the object is determined. Furthermore, according to the characteristic information of the object, a warning message is generated, where the characteristic information characterizes the current state of the object, and the warning message is used to prompt the object to leave the live electrical scenario. For the warning method in the live electrical scenario provided by the present application, by determining the category information of the object existing within the preset range of the live electrical equipment, a warning message can be generated according to the different category information of the object and the characteristic information under different category information, so as to prompt the object to leave the live electrical scenario, reducing the workload of manual maintenance. Moreover, the warning message can be generated according to the different category information of the object and the characteristic information under different category information, making the warning message for prompting the object to leave the live electrical scenario more targeted. The warning method in the live electrical scenario provided by the present application reduces the potential safety hazards in the live electrical scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0051] Figure 1 is a schematic flowchart of a warning method in a live electrical scenario provided by the present application Figure 1 ;

[0052] Figure 2 is a schematic flowchart of a warning method in a live electrical scenario provided by the present application Figure 2 ;

[0053] Figure 3Schematic diagram of a radar device provided by this application;

[0054] Figure 4 Flow schematic of an alarm method in a live working scenario provided by this application Figure 3 ;

[0055] Figure 5 Schematic architecture diagram of a method for determining fusion features provided by this application;

[0056] Figure 6 Schematic structure diagram of an alarm device in a live working scenario provided by this application Figure 1 ;

[0057] Figure 7 Schematic structure diagram of an alarm device in a live working scenario provided by this application Figure 2 ;

[0058] Figure 8 Schematic structure diagram of an electronic device provided by this application.

[0059] Through the above-mentioned drawings, specific embodiments of this application have been shown, and there will be more detailed descriptions hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of this application in any way, but to illustrate the concept of this application to those skilled in the art by referring to specific embodiments. Detailed implementation manners

[0060] Here, exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with this application. On the contrary, they are merely examples of devices and methods consistent with some aspects of this application as detailed in the appended claims.

[0061] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Moreover, the collection, use, and processing of relevant data need to comply with relevant laws, regulations, and standards, and corresponding operation entrances are provided for users to choose to authorize or refuse.

[0062] Live working and operations such as maintenance of substation equipment are routine operations in live working scenarios or power scenarios.

[0063] However, for example, during on-site operations such as power outage defect elimination, line relocation or new line erection of transmission lines, when there are double-circuit or multi-circuit transmission lines on the same tower and the operating personnel only de-energize one of the circuits for operation, there will be a risk of adjacent live operation. When on-site supervision is inadequate or the tower operators make incorrect judgments, the operating personnel may accidentally enter the live line for operation, resulting in electric shock accidents.

[0064] For another example, small animals (especially rats) may enter the substation through hidden structures such as cable trenches and cable trays, causing damage to key equipment such as high-voltage switchgear and cable layers, triggering short circuits, tripping or even more serious power accidents. In addition, the activities of small animals may also interfere with the normal operation of the secondary system, resulting in communication interruption or malfunction.

[0065] It should be noted that the small animals involved in this application can be understood as animals with a volume smaller than that of a human.

[0066] It can be understood that due to various reasons, various safety accidents will occur in live scenarios, such as electric shock to personnel and short circuits of electrical equipment.

[0067] Currently, the experience of the staff is used to prevent safety accidents and reduce their occurrence, but the cost of manual operation and maintenance is high, and the latency is high. Exemplarily, for the situation where operating personnel are adjacent to live operation, it mainly relies on manual supervision and the subjective judgment of operating personnel, and there is a possibility of accidentally entering the live line, resulting in electric shock accidents. For the problem of small animal intrusion, existing protection measures are difficult to completely eliminate the intrusion risk in complex environments, and lack effective identification and warning means, and thus corresponding measures cannot be taken.

[0068] Therefore, there is a big problem of safety hazards in live scenarios.

[0069] For the warning method, device, equipment, medium and product in the live scenario provided by this application, if it is determined that there is an object within the preset range of a live device, the category information of the object is determined, and according to the category information of the object, the characteristic information of the object is determined, and then according to the characteristic information of the object, a warning information is generated, where the characteristic information represents the current state of the object, and the warning information is used to prompt the object to leave the live scenario. The warning method in the live scenario provided by this application can, by determining the category information of the object existing within the preset range of the live device, generate warning information according to the different category information of the object and the characteristic information under different category information, so as to prompt the object to leave the live scenario, reducing the workload of manual maintenance, and the warning information can be generated according to the different category information of the object and the characteristic information under different category information, making the warning information for prompting the object to leave the live scenario more targeted. The warning method in the live scenario provided by this application reduces the safety hazards in the live scenario.

[0070] The technical solution of the present application and how the technical solution of the present application solves the above technical problems will be described in detail below with specific embodiments. The following several specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0071] Figure 1 Flow schematic of an alarm method in a live scenario provided by the present application Figure 1 The execution subject of this method can be a server, a host or other devices, such as Figure 1 As shown, the method may include:

[0072] S101. If it is determined that there is an object within the preset range of the live device, determine the category information of the object.

[0073] Among them, the live device may refer to an electrical device in a live scenario. Exemplarily, the live device may be the execution subject of the present application or other electrical devices. For example, high-voltage switch cabinets, cable layers, etc. in a substation. It can be understood that the live device may be a device in an energized state.

[0074] The preset range of the live device may refer to the range preset by the staff. The preset range may be a specific area around the live device. Exemplarily, taking the execution subject of the present application as the live device and taking the execution subject of the present application as the center of the circle and 10 meters as the radius to form a circular area, which may be the preset range of the live device.

[0075] The object may refer to a person, an animal or other living things. It can be understood that if there is an object within the preset range of the live device, it means that there are living things within the preset range of the live device, then there is a risk of electric shock to people or animals, and even further lead to safety accidents such as short circuits within the preset range of the live device.

[0076] It should be noted that there may be more than one live device in the present application, and the positions of some live devices may overlap, and the positions of the live devices can also be moved by the staff. Similarly, the position of the execution subject of the present application can also be moved by the staff.

[0077] Exemplarily, when the live device is a double-circuit or multi-circuit transmission line on the same tower, the execution subject of the present application is set at the double-circuit or multi-circuit transmission line on the same tower. That is, the live device may be a double-circuit or multi-circuit transmission line on the same tower or the execution subject of the present application. Taking the live device as the center of the circle and 15 meters as the radius to form a circular area, which may be the preset range of the live device.

[0078] If it is determined that there is an object within the preset range of the energized device, the category information of the object can be determined.

[0079] Among them, the category information of the object can refer to the basic attributes of the object. Exemplarily, the basic attributes of the object can be divided into multiple categories. For example, a person can be one category, and a small animal can be one category.

[0080] In a possible implementation manner, determining that there is an object within the preset range of the energized device can be achieved through a sensor device. Exemplarily, the execution entity of the present application is provided with a sensor, and the sensor can be an infrared sensor, a camera, a radar, or other environmental perception devices, etc.

[0081] The sensor monitors the preset range of the energized device in real time and transmits the detected signal to the execution entity of the present application, so that the execution entity of the present application determines that there is an object within the preset range of the energized device.

[0082] In an alternative implementation manner, step S101 may include:

[0083] Obtain an infrared thermal image within the preset range through a preset infrared sensing device; wherein, the infrared thermal image represents the temperature distribution within the preset range; if it is determined that there is a heat source in the infrared thermal image, it is determined that there is an object within the preset range of the energized device; wherein, the temperature value of the heat source is greater than a preset temperature threshold; determine the category information of the object according to the area of the heat source in the infrared thermal image.

[0084] Among them, the preset infrared sensing device can refer to a pre-set infrared sensing device. The infrared sensing device can refer to a device based on infrared radiation detection technology, which can be used to monitor the temperature distribution within the preset range of the energized device, and then determine whether there is an object within the preset range of the energized device.

[0085] Specifically, the infrared thermal imaging technology of the infrared sensing device can display the temperature distribution on the surface of the object within the preset range in the form of an image in real time, and this image is the infrared thermal image.

[0086] From the infrared thermal image within the preset range, it can be determined whether there is a heat source, and then according to the area of the heat source, the category information of the object can be determined.

[0087] Specifically, the temperature value of the heat source is greater than a preset temperature threshold. Among them, the preset temperature threshold can refer to the temperature threshold pre-set by the staff. Exemplarily, the temperature threshold can be 34 degrees Celsius. If there is a temperature greater than 34 degrees Celsius in the infrared thermal image, it can be determined that there is a heat source in the infrared thermal image, and it can be understood that there is an object within the preset range of the energized device.

[0088] Furthermore, the category information of the object can be determined by analyzing the area of the heat source.

[0089] It can be understood that the heat source areas presented by different objects in the infrared thermal imaging are different. For example, the area of the heat source corresponding to a human body is larger than that corresponding to a small animal. By a preset area threshold or range, it is possible to distinguish whether the object is a human, a small animal, or other living things, thereby realizing the identification of different categories of objects or the determination of category information.

[0090] The beneficial effect of such a setting is that the infrared thermal imaging within a preset range is obtained through the infrared thermal imaging technology of the infrared sensing device, and then it is possible to determine in real time whether there is an object within the preset range of the live equipment, reducing the cost of manual detection.

[0091] In an alternative embodiment, determining the category information of the object according to the area of the heat source in the infrared thermal imaging may include:

[0092] If the area of the heat source in the infrared thermal imaging is greater than the preset area threshold, then determine that the category information of the object is the first category; if the area of the heat source in the infrared thermal imaging is less than or equal to the preset area threshold, then determine that the category information of the object is the second category.

[0093] Among them, the preset area threshold can be the area threshold preset by the staff. It can be understood that the setting of the area threshold can be adjusted according to the specific scenario.

[0094] Exemplarily, if an infrared sensing device that can image 640×480 pixels is used, the area threshold can be set to 750 square pixels.

[0095] If the area of the heat source in the infrared thermal imaging is greater than 750 square pixels, such as 900 square pixels, then determine that the category information of the object is the first category; if the area of the heat source in the infrared thermal imaging is less than or equal to 750 square pixels, such as 500 square pixels, then determine that the category information of the object is the second category.

[0096] It can be understood that the first category can refer to one category in the category information of the object. Exemplarily, the first category may include objects with a larger volume such as humans.

[0097] The second category can also refer to one category in the category information of the object. Exemplarily, the second category may include objects with a smaller volume such as small animals.

[0098] It should be noted that the category information of the object can also be determined according to the area change rate of the heat source in the infrared thermal imaging.

[0099] The area change rate can characterize the area change of the heat source in the infrared thermal imaging at different time nodes of intervals.

[0100] Exemplarily, if the area change rate of the heat source in the infrared thermal imaging is less than the preset area change rate threshold, the category information of the object is determined to be the first category; if the area change rate of the heat source in the infrared thermal imaging is greater than or equal to the preset area change rate threshold, the category information of the object is determined to be the second category.

[0101] Among them, for example, the preset area change rate threshold is set to 200 square pixels per second. If the area change rate of the heat source in the infrared thermal imaging is less than 200 square pixels per second, such as 120 square pixels per second, the category information of the object is determined to be the first category; if the area change rate of the heat source in the infrared thermal imaging is greater than or equal to 200 square pixels per second, such as 250 square pixels per second, the category information of the object is determined to be the second category.

[0102] In a possible implementation manner, the category information of the object can be comprehensively determined according to data such as the preset area threshold and the preset area change rate threshold, and there is no limitation on how to determine the category information of the object here.

[0103] The beneficial effect of such a setting is that by determining the category information of the object according to the area of the heat source in the infrared thermal imaging, manual participation is not required, the labor cost is reduced, and the category information of the object can be determined more accurately to perform subsequent operations related to alarming.

[0104] S102. Determine the characteristic information of the object according to the category information of the object; where the characteristic information characterizes the current state of the object.

[0105] Among them, the characteristic information of the object can refer to the state information of the object at the current moment. By determining the characteristic information of the object, it can be more accurately judged whether the object is in a dangerous state or whether an alarm is required, etc.

[0106] Specifically, for objects of the first category, the objects of the first category can be people, and the characteristic information of people can be the state information of people (such as the moving state or stationary state of people), the position information of people (such as the distance between people and live equipment), etc.

[0107] For objects of the second category, the objects of the second category can be small animals, and the characteristic information of small animals can be the state information of small animals (such as the moving speed or moving direction of small animals), the species information of small animals (such as rats, birds), etc.

[0108] S103. Generate an alarm message according to the characteristic information of the object; where the alarm message is used to prompt the object to leave the live scene.

[0109] It can be understood that after determining the characteristic information of an object, it is possible to determine whether the object is in a dangerous state based on the characteristic information of the object. If the object is in a dangerous state, an alarm message is generated to prompt the object to leave the live scene.

[0110] Among them, the form of the alarm message can be various, including but not limited to sound alarms, light flashes, SMS notifications, or sending the alarm message to relevant personnel through other communication methods.

[0111] Exemplarily, if it is determined that the position information of the object of the first category is that the distance between the object of the first category and the live equipment is less than a preset distance threshold, such as 10 meters, an alarm message is generated to prompt the object of the first category to stay away from the live equipment. Specifically, by issuing a voice prompt and sending an SMS notification to the object of the first category, the object of the first category is prompted to stay away from the live equipment.

[0112] If it is determined that the category information of the object of the second category is a mouse, an alarm message is generated to prompt the mouse to stay away from the live equipment. Specifically, by sounding a whistle to drive away the mouse, or by sending an SMS notification to notify the staff to drive away the mouse.

[0113] An alarm method in a live scene provided by the present application can generate an alarm message based on the category information of the object existing within the preset range of the live equipment, and according to the different category information of the object and the characteristic information under different category information, to prompt the object to leave the live scene, reducing the workload of manual maintenance. Moreover, the alarm message can be generated according to the different category information of the object and the characteristic information under different category information, making the alarm message for prompting the object to leave the live scene more targeted. The alarm method in the live scene provided by the present application reduces potential safety hazards in the live scene.

[0114] Figure 2 Schematic flow of an alarm method in a live scene provided by the present application Figure 2 The execution subject of this method can be a server, a host, or other devices, such as Figure 2 shown, the method may include:

[0115] S201. If it is determined that there is an object within the preset range of the live equipment, determine the category information of the object.

[0116] Exemplarily, this step can refer to the above step S101 and will not be elaborated here.

[0117] S202. If the category information is determined to be the first category, determine the distance information between the object and the radar device through a preset radar device; wherein, the radar device is installed in the live scene.

[0118] It can be understood that after determining that the category information is the first category, the distance information between the object of the first category and the radar device can be determined through a preset radar device.

[0119] Among them, the preset radar device can refer to the radar device pre-set by the staff, and the radar device is installed in the live working scenario.

[0120] Exemplarily, the line operator performs operations according to the operation task. The operation task includes the installation position of the radar device and the operation site. The operation site is in the live working scenario, and the installation position of the radar device is the center of the operation site. The operator installs the radar device according to the installation position of the radar device in the task.

[0121] It can be understood that after installing the radar device at the center of the operation site, the distance information between the object of the first category and the radar device can be determined. Furthermore, based on the distance information between the object of the first category and the radar device, it can be determined whether the object of the first category is in a dangerous state.

[0122] Specifically, the radar device can include, but is not limited to, millimeter-wave radar and lidar.

[0123] In an alternative embodiment, determining the distance information between the object and the radar device through the preset radar device may include:

[0124] Determining the time information and frequency information of the radar signal through the radar signal emitted by the radar device; wherein, the time information represents the duration from when the radar signal is sent to the object and returns to the radar device, and the frequency information represents the frequency change of the radar signal; based on the time information and frequency information, determining the distance information between the object and the radar device.

[0125] It can be understood that the radar device can measure the distance between the object and the radar device by emitting a radar signal to the object and receiving the echo signal reflected from the object.

[0126] Among them, the radar signal can include, but is not limited to, electromagnetic wave signal, pulse signal, Frequency Modulated Continuous Wave (FMCW) signal.

[0127] Specifically, the time information of the radar signal can represent the duration from when the radar signal is sent to the object and returns to the radar device, and the frequency information of the radar signal can represent the frequency change of the radar signal. Based on the time information and frequency information, the distance information between the object of the first category and the radar device can be determined.

[0128] In a possible implementation, the radar device may emit radar signals to the object multiple times, or the number of radar signals emitted by the radar device to the object each time may be multiple.

[0129] Exemplarily, if the radar signal is a frequency-modulated continuous wave signal, the radar device emits the frequency-modulated continuous wave signal to the object and receives the electromagnetic wave signal returned from the object, then the time information of the radar signal can be measured by a preset timer, and the frequency information of the radar signal can be measured by a preset frequency sensor.

[0130] In a possible implementation, to determine the distance information between the object and the radar device, it can be jointly solved by the following formula:

[0131] ;

[0132] ;

[0133] ;

[0134] Wherein, represents the time information of the radar signal, represents the frequency information of the radar signal, represents the maximum frequency offset of the frequency-modulated continuous wave signal in the frequency information, represents the modulation period of the frequency-modulated continuous wave signal, represents the speed of light, represents the distance information corresponding to the time information, represents the distance information corresponding to the frequency information, represents the distance information between the object and the radar device.

[0135] In a possible implementation, if the radar signal is a pulse signal, the distance information between the object and the radar device can be directly determined according to the time information of the radar signal. Specifically, the distance information between the object and the radar device satisfies:

[0136] ;

[0137] Wherein, represents the distance information between the object and the radar device, represents the time information of the radar signal, represents the speed of light.

[0138] In a possible implementation, if the radar signal is a frequency-modulated continuous wave signal, the distance information between the object and the radar device can be directly determined according to the frequency information of the radar signal. Specifically, the distance information between the object and the radar device satisfies:

[0139] ;

[0140] wherein, represents the distance information between the object and the radar device, represents the frequency information of the radar signal, represents the frequency modulation slope of the radar signal, represents the speed of light.

[0141] It should be noted that to determine the distance information between the object and the radar device, there can be other radar-based ranging methods, which will not be elaborated here.

[0142] To better describe the radar device, Figure 3 FIG. is a schematic diagram of a radar device provided by the present application. As Figure 3 shown, the radar device is a lidar device. Specifically, the lidar device includes a lidar matrix. In addition, the radar device further includes a detector host and a telescopic folding bracket.

[0143] Wherein, the lidar matrix can be composed of multiple signal transmitting units and signal receiving units. Exemplarily, the lidar matrix can be composed of 5 signal transmitting units and 5 signal receiving units, and the signal can refer to a laser signal.

[0144] Exemplarily, the lidar matrix can send an FMCW signal to the object and receive the FMCW signal reflected from the object, so as to provide data support for determining the distance information between the object and the radar device.

[0145] It can be understood that by setting the lidar matrix, the accuracy of determining the distance information between the object and the radar device through the radar device can be improved.

[0146] The detector host can refer to a device for data recording or data processing. In a possible implementation manner, the detector host can be a part of the execution subject of the present application. Specifically, the detector host can determine the time information and frequency information of the radar signal.

[0147] In a possible implementation manner, in the detector host of the execution subject of the present application, an alarm unit can also be included. The alarm unit can issue voice prompts, etc., to prompt objects of the first category to stay away from live equipment.

[0148] The telescopic folding bracket can be used to support the lidar matrix and the detector host, so that the radar device can be installed at various positions in the live scenario.

[0149] At the same time, the bracket has the function of telescoping and folding, which provides convenience for operators to install the radar device.

[0150] S203. Determine the distance information as the feature information of the object.

[0151] It can be understood that after obtaining the distance information between the object and the radar device, this distance information can be used as the feature information of the object.

[0152] Specifically, the distance information can be used to determine whether an object of the first category is in a dangerous state.

[0153] S204. If the distance information is less than a preset distance threshold, generate an alarm message based on a preset information template.

[0154] Among them, the preset distance threshold can refer to the distance threshold preset by the staff. By comparing the distance information with the preset distance threshold, it can be determined whether an object of the first category is in a dangerous state, and then an alarm message is generated based on the preset information template.

[0155] Exemplarily, the preset distance threshold can be 3 meters. If the distance information is less than 3 meters, for example, 2 meters, it means that the object of the first category is close to the radar device, and the radar device is close to the live equipment. It can be determined that the operator has approached the live equipment and the operator faces the safety hazard of electric shock. Then, an alarm message is generated based on the preset information template.

[0156] The preset information template can be set by the staff in advance. Exemplarily, the information template can include information such as an alarm title, an alarm level, alarm content, recommended measures, a timestamp, etc. The specific content of the information template is the alarm message.

[0157] Among them, the alarm title can refer to the prompt of the alarm, such as "Personnel have entered the dangerous area".

[0158] The alarm level can refer to the level divided according to the degree of danger, such as low risk, medium risk, and high risk. Exemplarily, when the preset distance threshold is 3 meters, the distance information corresponding to low risk can be 2.9 meters, the distance information corresponding to medium risk can be 2 meters, and the distance information corresponding to high risk can be 1 meter.

[0159] It can be understood that low risk means that the distance information is slightly lower than the preset distance threshold, medium risk means that the distance information is significantly lower than the preset distance threshold, and high risk means that the distance information is extremely lower than the preset distance threshold.

[0160] The alarm content can include the category information of the object, the distance information between the object and the radar device, etc.

[0161] The recommended measures can refer to the recommended measures corresponding to the alarm level. For example, the recommended measure corresponding to low risk is "Please leave the dangerous area immediately".

[0162] Timestamps can be used to record the time when an alarm occurs, facilitating subsequent analysis and traceability of alarm information.

[0163] In a possible implementation, the execution entity of this application is provided with a display, which is used to visually display the distance information and warning information between the object and the radar device to the operator or relevant personnel.

[0164] In an alternative implementation, in addition to determining whether to generate alarm information based on the distance information, it is also possible to determine whether to generate alarm information based on other information, such as the environmental temperature and humidity within a preset range.

[0165] Exemplarily, it is also possible to comprehensively determine whether to generate alarm information by combining the distance information with other information such as the environmental temperature and humidity within a preset range. For example, if the distance information is greater than the preset distance threshold but the environmental humidity within the preset range is greater than the preset humidity threshold, an alarm information is generated to avoid electric shock of the object caused by high environmental humidity.

[0166] There is no limitation on the judgment conditions for determining whether to generate alarm information here.

[0167] An alarm method in a live working scenario provided by this application, after determining that the category information of the object is the first category, determines the distance information between the object and the radar device through a preset radar device, and then determines whether to generate alarm information based on the distance information. The alarm information is used to prompt the object of the first category to stay away from the live working scenario, without the need for manual maintenance, reducing labor costs. The alarm method in the live working scenario provided by this application reduces potential safety hazards in the live working scenario.

[0168] Figure 4 It is a flowchart of an alarm method in a live working scenario provided by this application Figure 3 , the execution entity of the method can be a server, a host or other devices, such as Figure 4 shown, the method may include:

[0169] S401. If it is determined that there is an object within the preset range of the live equipment, determine the category information of the object.

[0170] Exemplarily, this step can refer to the above step S101 and will not be elaborated here.

[0171] S402. Determine that the category information is the second category, obtain the voiceprint information of the object through a preset voiceprint collector, and obtain the image information of the object through a preset image acquisition device.

[0172] It can be understood that after determining that the category information is the second category, the voiceprint information of the object can be obtained through a preset voiceprint collector, and the image information of the object can be obtained through a preset image acquisition device.

[0173] Among them, the preset voiceprint collector is used to obtain the voiceprint information of the objects of the second category. Exemplarily, the voiceprint collector may include, but is not limited to, a microphone. The voiceprint information may refer to the sounds made by the objects of the second category. It can be understood that the frequencies and intensities of the sounds made by different objects are different.

[0174] The preset image acquisition device can be used to obtain the image information of the objects of the second category. Exemplarily, the image acquisition device may include, but is not limited to, a camera. The image information may refer to the appearance of the objects of the second category. It can be understood that the shapes, sizes, and colors of different objects are different.

[0175] It can be understood that obtaining the voiceprint information and image information of the object can better classify the objects of the second category, making the generated alarm information more targeted.

[0176] S403. Determine the type information of the object according to the voiceprint information and image information of the object; wherein, the type information is a subcategory of the second category.

[0177] Among them, the type information of the object may refer to the subcategory of the objects of the second category.

[0178] Exemplarily, the objects of the second category may be small animals, and the subcategories of the second category, that is, the subcategories of small animals, may be rodents, birds, etc.

[0179] It can be understood that according to the voiceprint information and image information of the small animals, the type information of the small animals can be determined.

[0180] In an optional implementation manner, determining the type information of the object according to the voiceprint information and image information of the object may include:

[0181] Performing feature extraction processing on the voiceprint information and image information of the object according to a preset feature extraction network to obtain a fused feature; wherein, the fused feature characterizes the voiceprint information and image information of the object; inputting the fused feature and the preset prompt word information into a preset large language model to obtain the output type information of the object.

[0182] Among them, the preset feature extraction network can be used to perform extraction processing on the voiceprint information and image information of the object to obtain a fused feature that can characterize the voiceprint information and image information of the object.

[0183] Exemplarily, the preset feature extraction network may include, but is not limited to, Convolutional Neural Networks (CNN) and Recurrent Neural Network (RNN).

[0184] The fused feature may refer to the feature obtained by fusing the feature corresponding to the voiceprint information and the feature corresponding to the image information.

[0185] Exemplarily, the fusion process may refer to concatenating the feature vector corresponding to the voiceprint information and the feature vector corresponding to the image information to obtain the fused feature. The fused feature may also refer to respectively assigning corresponding weights to the feature vector corresponding to the voiceprint information and the feature vector corresponding to the image information, and then concatenating the weighted feature vector corresponding to the voiceprint information and the weighted feature vector corresponding to the image information to obtain the fused feature.

[0186] The preset Large Language Model (LLM) may be a pre-trained model, which can be used to process the input of multimodal data and output corresponding results.

[0187] It can be understood that by inputting the fused feature and the preset prompt information into the preset large language model, the category information of the object output by the large language model can be obtained.

[0188] Among them, the preset prompt information may refer to the text prompt designed in advance to guide the large language model to generate a specific output.

[0189] Exemplarily, the prompt information may be "What are the matching degrees of small animals matching the fused feature of this voiceprint image, represent the matching degrees by proportions, and select the category with the highest proportion as the category of small animals". Furthermore, the category information of the object output by the large language model may be rodents.

[0190] Specifically, the large language model may be pre-trained in advance with the fused feature and the animal category corresponding to the fused feature. Exemplarily, obtain a general large language model from an open-source website as the initial model, input the data set (fused feature) into the initial model, and use the preset prompt to guide the initial model to output the result. If the matching degree between the category information of the object output by the initial model and the actual category information corresponding to the fused feature meets the preset condition, it can be determined that the initial model training is completed, and the trained large language model is obtained.

[0191] The beneficial effect of such a setting is that by inputting the fused feature into the large language model to further generate the category information of the object, the accuracy and reliability of determining the category information of the object can be improved.

[0192] In an alternative embodiment, the preset feature extraction network includes a first extraction network and a plurality of second extraction networks. According to the preset feature extraction network, feature extraction processing is performed on the voiceprint information of the object and the image information of the animal to obtain a fusion feature, which may include:

[0193] Input the voiceprint information of the object into the first extraction network to obtain a voiceprint feature; input the image information of the object into each second extraction network respectively to obtain a plurality of image features; wherein, the sizes of the convolution kernels in each second extraction network are different, and the second extraction network corresponds to the image feature one by one; perform a first fusion process on the voiceprint feature and each image feature respectively to obtain a plurality of initial features; wherein, the initial feature represents the voiceprint feature and the image feature, and the number of initial features is the same as the number of image features; perform a second fusion process on the plurality of initial features to obtain a fusion feature.

[0194] Among them, the first extraction network can be used to extract the voiceprint feature according to the voiceprint information of the object, and the second extraction network can be used to extract the image feature according to the image information of the object.

[0195] It can be understood that the number of the first extraction networks can be one, and the number of the second extraction networks can be multiple, and the image features extracted by each second extraction network are different.

[0196] Specifically, the first extraction network may include, but is not limited to, a long short-term memory network (Long Short-Term Memory, LSTM). By inputting the voiceprint information of the object of the second category into the long short-term memory network, the voiceprint feature output by the long short-term memory network can be obtained.

[0197] Exemplarily, the voiceprint feature may refer to Mel Frequency Cepstrum Coefficient (MFCC). Specifically, the determination of the Mel Frequency Cepstrum Coefficient may include:

[0198] Perform preprocessing on the voiceprint information of the object (such as noise reduction, pre-emphasis, framing, windowing, etc.); perform fast Fourier transform processing on the preprocessed voiceprint information, that is, convert the preprocessed voiceprint information from a time-domain signal to a frequency-domain signal; map the frequency-domain signal to the Mel scale similar to human auditory perception through a Mel filter, and then extract the Mel Frequency Cepstrum Coefficient through a Discrete Cosine Transform (DCT).

[0199] The second extraction network may be multiple, and the multiple second extraction networks are respectively used to process the image information to extract a plurality of image features. It can be understood that the second extraction network corresponds to the image feature one by one.

[0200] Specifically, the second extraction network may include, but is not limited to, a Convolutional Neural Network (CNN). The convolutional neural network includes convolutional kernels, which can be used to extract image features.

[0201] It can be understood that if the sizes of the convolutional kernels in each second extraction network are different, the scales of the image features output by each convolutional kernel are also different.

[0202] Exemplarily, the size of the convolutional kernel can be 3×3, 5×5, 7×7, etc.

[0203] In a possible implementation, the second extraction network includes a plurality of convolutional modules and an FPN module.

[0204] Among them, the convolutional module may include a convolutional layer, a BN (Batch Normalization) layer, an activation layer, etc. The convolutional layer can be used to extract local features of the image. The BN layer is used to normalize the input data, and the activation layer can be used to perform a non-linear transformation on the input data. It can be understood that the convolutional module is used to extract local features of the image information, and the convolutional module improves the accuracy of image feature extraction.

[0205] The FPN (Feature Pyramid Network) module is used to fuse the output results of multiple convolutional modules to obtain a feature representation, that is, the image features corresponding to the image information.

[0206] It can be understood that the image information of the object passes through the processing of multiple convolutional modules in sequence, and then through the processing of the FPN module. The output result of the FPN module is the image features corresponding to the image information of the object.

[0207] To better describe how to obtain the fusion features, Figure 5 The following is a schematic diagram of the architecture of a method for determining fusion features provided by this application. As Figure 5 shown, the preset feature extraction network includes a first extraction network, n second extraction networks, and n cross-attention modules (English: Cross Attention), where n can be a positive integer greater than or equal to 1. The cross-attention module is used to implement the first fusion process, and the cross-attention module corresponds to the second extraction network one by one.

[0208] Specifically, the first extraction network performs feature extraction processing on the voiceprint information and outputs voiceprint features; multiple second extraction networks (Second Extraction Network 1, Second Extraction Network 2, Second Extraction Network 3, ……, Second Extraction Network n) respectively perform feature extraction processing on the image information to obtain multiple image features (Image Feature 1, Image Feature 2, Image Feature 3, ……, Image Feature n);

[0209] The voiceprint features and each image feature are respectively input into each cross-attention module (Cross Attention 1, Cross Attention 2, Cross Attention 3, ……, Cross Attention n) to obtain multiple initial features (Initial Feature 1, Initial Feature 2, Initial Feature 3, ……, Initial Feature n);

[0210] The multiple initial features are subjected to second fusion processing to obtain a fused feature.

[0211] Among them, the first fusion processing may refer to fusing the voiceprint feature with each image feature to obtain multiple initial features, and the first fusion processing is implemented through a cross-attention module.

[0212] Exemplarily, the cross-attention module calculates the correlation between the voiceprint feature and the image feature through an attention mechanism and generates a fused initial feature.

[0213] The second fusion processing may refer to further fusing the multiple initial features to generate a final and unique fused feature.

[0214] Exemplarily, the second fusion processing may include: aligning and merging the multiple initial features, that is, mapping the multiple initial features to the same scale through a 1×1 convolution kernel, assigning weights to the multiple initial features on the same scale, that is, respectively assigning corresponding weight values to each initial feature, and then splicing the initial features with assigned weight values to obtain a fused feature.

[0215] Through the first extraction network and multiple second extraction networks, feature extraction processing is performed on the voiceprint information of the object and the image information of the animal to obtain a fused feature. The beneficial effect of such a setting is that it fuses the voiceprint information and image information of the object of the second category, making the determination of the category information of the object more accurate.

[0216] S404. Determine the category information of the object as the feature information of the object.

[0217] It can be understood that after determining the category information of the object, the category information can be determined as the feature information of the object of the second category.

[0218] Exemplarily, if the type information of the object is rodent, then the characteristic information of the object is rodent.

[0219] S405. Determine the expulsion plan for the object based on the preset association relationship according to the type information of the object; wherein, the preset association relationship represents the association relationship between the type information and the expulsion plan, and the expulsion plan represents the method of prompting the object to leave.

[0220] Among them, the expulsion plan for the object can refer to the specific method or strategy for prompting the object to leave the preset range of the energized equipment. There is an association relationship between the type information of the object and the expulsion plan for the object.

[0221] It can be understood that if the type information of the object is rodent, the expulsion plan for the object can be the specific method or strategy for prompting the rodent to leave the preset range of the energized equipment; if the type information of the object is bird, the expulsion plan for the object can be the specific method or strategy for prompting the bird to leave the preset range of the energized equipment.

[0222] S406. Generate an alarm message according to the expulsion plan for the object.

[0223] Exemplarily, if the type information of the object is rodent, the alarm message can be a sound whistle to expel the rodent; the alarm message can also be notified to the staff by text message to expel the rodent, for example, using a mousetrap to keep the rodent away from the preset range of the energized equipment.

[0224] A warning method in a live scenario provided by the present application, after determining that the category information of the object is the second category, determines the type information of the object according to the voiceprint information and image information of the object, and then determines the expulsion plan for the object and generates an alarm message, which can accurately identify and process potentially dangerous objects in the live scenario, and then generate an alarm message to prompt the object to leave the preset range of the live scenario. The warning method in the live scenario provided by the present application reduces the potential safety hazards in the live scenario.

[0225] Figure 6 The following is a schematic structure of an alarm device in a live scenario provided by the present application Figure 1 , as Figure 6 shown, the alarm device 60 in the live scenario includes: a first determination unit 601, a second determination unit 602, and an alarm unit 603.

[0226] The first determination unit 601 is configured to determine the category information of the object if it is determined that there is an object within the preset range of the energized equipment;

[0227] The second determination unit 602 is configured to determine the characteristic information of the object according to the category information of the object; wherein, the characteristic information represents the current state of the object;

[0228] An alarm unit 603, configured to generate an alarm message according to the feature information of an object; wherein, the alarm message is used to prompt the object to leave the live scenario.

[0229] Figure 7 The structural schematic of an alarm device in a live scenario provided by the present application Figure 2 , such as Figure 7 As shown, the alarm device 70 in the live scenario includes: a first determination unit 701, a second determination unit 702, and an alarm unit 703. Among them, the first determination unit 701 further includes a first processing module 7011, a second processing module 7012, and a third processing module 7013.

[0230] The first processing module 7011 is configured to obtain an infrared thermal image within a preset range through a preset infrared sensing device; wherein, the infrared thermal image represents the temperature distribution within the preset range.

[0231] The second processing module 7012 is configured to determine that there is an object within the preset range of the live device if it is determined that there is a heat source in the infrared thermal image; wherein, the temperature value of the heat source is greater than a preset temperature threshold.

[0232] The third processing module 7013 is configured to determine the category information of the object according to the area of the heat source in the infrared thermal image.

[0233] In an optional example, the third processing module 7013 is further specifically configured to determine that the category information of the object is the first category if the area of the heat source in the infrared thermal image is greater than a preset area threshold;

[0234] If the area of the heat source in the infrared thermal image is less than or equal to the preset area threshold, it is determined that the category information of the object is the second category.

[0235] In an optional example, when the category information is the first category, the second determination unit 702 is configured to determine the distance information between the object and the radar device through a preset radar device; wherein, the radar device is installed in the live scenario;

[0236] Determine the distance information as the feature information of the object.

[0237] In an optional example, the second determination unit 702 further includes a fourth processing module, and the fourth processing module is configured to determine the time information and frequency information of the radar signal through the radar signal emitted by the radar device; wherein, the time information represents the duration from the radar signal being sent to the object and returning to the radar device, and the frequency information represents the frequency change of the radar signal;

[0238] Determine the distance information between the object and the radar device according to the time information and the frequency information.

[0239] In an optional example, the alarm unit 703 is configured to generate an alarm message based on a preset information template if the distance information is less than a preset distance threshold.

[0240] In an optional example, the category information is the second category. The second determination unit 702 is further configured to obtain the voiceprint information of the object through a preset voiceprint collector and obtain the image information of the object through a preset image acquisition device.

[0241] Determine the type information of the object according to the voiceprint information and image information of the object; wherein, the type information is a subcategory of the second category.

[0242] Determine the type information of the object as the feature information of the object.

[0243] In an optional example, the second determination unit 702 further includes a fifth processing module. The fifth processing module is configured to perform feature extraction processing on the voiceprint information and image information of the object according to a preset feature extraction network to obtain a fusion feature; wherein, the fusion feature represents the voiceprint information and image information of the object.

[0244] Input the fusion feature and the preset prompt word information into a preset large language model to obtain the output type information of the object.

[0245] In an optional example, the preset feature extraction network includes a first extraction network and multiple second extraction networks. The fifth processing module is further specifically configured to input the voiceprint information of the object into the first extraction network to obtain a voiceprint feature.

[0246] Input the image information of the object into each second extraction network respectively to obtain multiple image features; wherein, the sizes of the convolution kernels in each second extraction network are different, and the second extraction network corresponds to the image feature one by one.

[0247] Perform a first fusion process on the voiceprint feature and each image feature respectively to obtain multiple initial features; wherein, the initial feature represents the voiceprint feature and the image feature, and the number of initial features is the same as the number of image features.

[0248] Perform a second fusion process on the multiple initial features to obtain a fusion feature.

[0249] In an optional example, the alarm unit 703 is further configured to determine an expulsion plan for the object according to the type information of the object based on a preset association relationship; wherein, the preset association relationship represents the association relationship between the type information and the expulsion plan, and the expulsion plan represents a method of prompting the object to leave.

[0250] Generate an alarm message according to the expulsion plan of the object.

[0251] Figure 8 A schematic structural diagram of the electronic device provided by this application is as follows Figure 8 As shown, the electronic device 80 provided in this embodiment includes: at least one processor 801 and a memory 802. Optionally, the device 80 further includes a communication component 803. Among them, the processor 801, the memory 802, and the communication component 803 are connected through a bus 804.

[0252] In a specific implementation process, at least one processor 801 executes the computer-executable instructions stored in the memory 802, so that at least one processor 801 executes the above method.

[0253] For the specific implementation process of the processor 801, reference can be made to the above method embodiment, and its implementation principle and technical effects are similar, so they will not be elaborated here in this embodiment.

[0254] In the above embodiment, it should be understood that the processor may be a central processing unit (English: Central Processing Unit, abbreviated: CPU), or other general-purpose processors, digital signal processors (English: Digital Signal Processor, abbreviated: DSP), application-specific integrated circuits (English: Application Specific Integrated Circuit, abbreviated: ASIC), etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly embodied as being executed by a hardware processor, or executed by a combination of hardware and software modules in the processor.

[0255] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.

[0256] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.

[0257] This application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0258] The present application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-mentioned method.

[0259] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0260] Furthermore, it should be noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0261] It should be understood that the above-mentioned device embodiments are illustrative only, and the devices of the present application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.

[0262] In addition, without special description, in each embodiment of the present application, each functional unit / module can be integrated in one unit / module, or each unit / module can exist physically alone, or two or more units / modules can be integrated together. The above integrated unit / module can be implemented in the form of hardware or in the form of a software program module.

[0263] When the integrated unit / module is implemented in the form of hardware, the hardware can be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc. Unless otherwise specified, the processor can be any suitable hardware processor, such as a CPU, GPU, FPGA, DSP, and ASIC, etc. Unless otherwise specified, the storage unit can be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc.

[0264] When the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes: USB flash drives, read-only memory (ROM), random access memory (RAM), mobile hard disks, magnetic disks, or optical discs and other various media that can store program codes.

[0265] In the above embodiments, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0266] Other embodiments of the present application will be readily apparent to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application, which follow the general principles of the present application and include known common general knowledge or conventional technical means in the technical field not disclosed in the present application. The specification and examples are only regarded as exemplary, and the true scope and spirit of the present application are pointed out by the following claims.

[0267] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. An alarm method in a live scene, characterized in that: include: If it is determined that there is an object within the preset range of the powered device, determining the category information of the object; Determining feature information of the object according to the category information of the object; wherein the feature information represents the current state of the object; Generate warning information based on the characteristic information of the object; wherein the warning information is used to prompt the object to leave the charged scene.

2. The method according to claim 1, characterized in that The category information is a first category; determining feature information of the object according to the category information of the object includes: Determining the distance information between the object and the radar device by using a preset radar device; wherein the radar device is installed in an electrically charged scene; The distance information is determined as feature information of the object.

3. The method according to claim 2, characterized in that Determining the distance information between the object and the radar device by using a preset radar device includes: Determine the time information and frequency information of the radar signal from the radar device; wherein the time information represents the time length of time for the radar signal to be sent to the object and returned to the radar device, and the frequency information represents the frequency change of the radar signal; The distance information between the object and the radar device is determined according to the time information and the frequency information.

4. The method according to claim 3, characterized in that Generate warning information according to the characteristic information of the object, including: If the distance information is less than a preset distance threshold, the warning information is generated based on a preset information template.

5. The method according to claim 1, characterized in that The category information is the second category; determining the feature information of the object according to the category information of the object includes: Acquiring the voiceprint information of the object through a preset voiceprint collector, and acquiring the image information of the object through a preset image acquisition device; Determining the category information of the object according to the voiceprint information and the image information of the object; wherein the category information is a subordinate category of the second category; The type information of the object is determined as the feature information of the object.

6. The method according to claim 5, characterized in that Determining the type information of the object according to the voiceprint information and the image information of the object includes: According to a preset feature extraction network, feature extraction processing is performed on the voiceprint information and image information of the object to obtain a fusion feature; wherein the fusion feature represents the voiceprint information and image information of the object; The fused features and the preset prompt word information are input into a preset large language model to obtain the output type information of the object.

7. The method according to claim 6, characterized in that The preset feature extraction network includes a first extraction network and a plurality of second extraction networks; according to the preset feature extraction network, feature extraction processing is performed on the voiceprint information of the object and the image information of the animal to obtain fusion features, including: Inputting the voiceprint information of the object into the first extraction network to obtain a voiceprint feature; Inputting the image information of the object into each second extraction network respectively to obtain a plurality of image features; wherein the size of the convolution kernel in each second extraction network is different, and the second extraction network corresponds to the image features one by one; The voiceprint features are respectively subjected to a first fusion process with each image feature to obtain a plurality of initial features; wherein the initial features represent the voiceprint features and the image features, and the number of the initial features is consistent with the number of the image features; The multiple initial features are subjected to a second fusion process to obtain the fused feature.

8. The method according to claim 5, characterized in that Generate warning information according to the characteristic information of the object, including: Determine, according to the type information of the object and based on a preset association relationship, an eviction scheme for the object; wherein the preset association relationship represents an association relationship between the type information and the eviction scheme, and the eviction scheme represents a method of prompting the object to leave; The warning information is generated according to the eviction plan of the object.

9. The method according to any one of claims 1 to 8, characterized in that If it is determined that there is an object within the preset range of the powered device, then determining the category information of the object includes: Obtaining infrared thermal imaging within the preset range through a preset infrared sensing device; wherein the infrared thermal imaging represents the temperature distribution within the preset range; If it is determined that there is a heat source in the infrared thermal imaging, it is determined that there is an object within the preset range of the charged device; wherein the temperature value of the heat source is greater than a preset temperature threshold; The category information of the object is determined according to the area of ​​the heat source of the infrared thermal imaging.

10. The method according to claim 9, characterized in that Determining the category information of the object according to the area of ​​the heat source of the infrared thermal imaging includes: If the area of ​​the heat source of the infrared thermal imaging is greater than a preset area threshold, determining that the category information of the object is the first category; If the area of ​​the heat source of the infrared thermal imaging is less than or equal to a preset area threshold, it is determined that the category information of the object is the second category.

11. An alarm device for live scenes, characterized in that: include: A first determining unit, configured to determine category information of an object if it is determined that there is an object within a preset range of the powered device; A second determining unit is used to determine feature information of the object according to the category information of the object; wherein the feature information represents a current state of the object; An alarm unit is used to generate alarm information according to the characteristic information of the object; wherein the alarm information is used to prompt the object to leave the charged scene.

12. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 10 when executed by a processor.

14. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 10 when being executed by a processor.

Images