Vehicle camping disaster early warning method, device, equipment, medium, product and vehicle

Abstract

The invention discloses a vehicle camping disaster early warning method, device and equipment, a medium, a product and a vehicle, and relates to the technical field of disaster early warning. The method comprises the following steps: acquiring multi-source heterogeneous data acquired by a vehicle, wherein the multi-source heterogeneous data comprises video image data, voice data, millimeter wave radar data and rainfall sensor data; fusion prediction is carried out based on the multi-source heterogeneous data, a disaster prediction result of the vehicle is determined and obtained, and the disaster prediction result is used for indicating a target disaster category faced by the vehicle and a target disaster grade corresponding to the target disaster category. The target disaster category comprises at least one of wild animal attack disasters, illegal invasion disasters and natural rain and snow disasters. According to the embodiment of the invention, disaster early warning of various field disasters can be timely and accurately realized in an automobile field camping scene.

Description

Technical Field

[0001] The present application belongs to the field of disaster warning technology, and in particular, relates to a vehicle camping disaster warning method, device, equipment, medium, product and vehicle. Background Art

[0002] At present, many cars have the function of camping outdoors. However, when the vehicles are camping and resting outdoors, there are problems such as sudden attacks by wild animals, illegal invasions, and natural disasters. The above disaster problems will cause personal and property safety problems for users. However, at this stage, the disaster warning scheme for car camping scenarios has problems such as single protection function and high false alarm rate, which cannot effectively protect the personal and property safety of users. Based on this, the industry is still in urgent need of a new type of car camping disaster warning scheme to achieve the purpose of timely and accurate warning of various types of outdoor disaster problems and realize active protection of cars. Summary of the invention

[0003] The embodiments of the present application provide a vehicle camping disaster warning method, device, equipment, medium, product and vehicle, which can timely and accurately realize disaster warning of various outdoor disasters in a car outdoor camping scenario, thereby facilitating the effective implementation of active protection of the car, thereby fully improving the user's personal and property safety.

[0004] In a first aspect, an embodiment of the present application provides a vehicle camping disaster warning method, the vehicle camping disaster warning method comprising:

[0005] Acquire multi-source heterogeneous data collected by vehicles, including video image data, voice data, millimeter wave radar data, and rainfall sensor data;

[0006] Based on fusion prediction of multi-source heterogeneous data, the disaster prediction result of the vehicle is determined. The disaster prediction result is used to indicate the target disaster category faced by the vehicle and the target disaster level corresponding to the target disaster category. The target disaster category includes at least one of wild animal attack disasters, illegal invasion disasters, and rain and snow natural disasters.

[0007] In some possible implementations, defining the disaster prediction result includes a first disaster prediction result and / or a second disaster prediction result; and performing fusion prediction based on multi-source heterogeneous data to determine the disaster prediction result of the vehicle includes:

[0008] Based on the video image data, the voice data and the millimeter wave radar data, a first disaster prediction result is predicted, and the first disaster prediction result is associated with whether the vehicle faces a wild animal attack disaster and / or an illegal invasion disaster, and the corresponding disaster level;

[0009] and / or,

[0010] Based on at least one of the video image data and the rainfall sensor data, a second disaster prediction result is predicted, and the second disaster prediction result is associated with whether the vehicle is facing a natural disaster of rain or snow and a corresponding disaster level.

[0011] In some possible implementations, predicting a first disaster prediction result based on the video image data, the voice data, and the millimeter wave radar data includes:

[0012] Predicting the object category to which the target object belongs based on the video image data and the voice data respectively, to obtain a first category prediction result and a second category prediction result;

[0013] Performing data fusion processing based on the first category prediction result, the second category prediction result and the millimeter wave radar data to obtain the predicted category and predicted distance of the target object;

[0014] Based on the prediction type and prediction distance, the first disaster prediction result is determined.

[0015] In some possible implementations, predicting the type of the target object based on the video image data and the voice data respectively to obtain a first type prediction result and a second type prediction result includes:

[0016] Based on the video image data, the type of the target object is predicted by the YOLO model to obtain a first type prediction result;

[0017] And / or, based on the speech data, the type of the target object is predicted by an LSTM long short-term memory network model to obtain a second type prediction result.

[0018] In some possible implementations, data fusion processing is performed based on the first category prediction result, the second category prediction result, and the millimeter wave radar data to obtain the predicted category and predicted distance of the target object, including:

[0019] Data fusion is performed based on the first category prediction results, the second category prediction results and the millimeter wave radar data, and the target object is converted to a bird's-eye view space to obtain the predicted category and predicted distance.

[0020] In some possible implementations, determining the first disaster prediction result based on the prediction type and the prediction distance includes:

[0021] Based on the predicted category, determine the target disaster category faced by the vehicle;

[0022] The target disaster level is determined based on the prediction type, prediction distance and disaster level classification rules corresponding to the target disaster category. The disaster level classification rules include the correspondence between different distance range intervals and disaster levels.

[0023] In some possible implementations, the object types include humans, small wild animals, and large wild animals; based on the predicted types, determining the target disaster category faced by the vehicle includes:

[0024] In a case where the predicted category indicates that the target object belongs to a human being, determining that the target disaster category includes an illegal invasion disaster;

[0025] In a case where the predicted category indicates that the object category to which the target object belongs is a small wild animal or a large wild animal, it is determined that the target disaster category includes a wild animal attack disaster.

[0026] In some possible implementations, predicting and obtaining a second disaster prediction result based on at least one of the video image data and the rainfall sensor data includes:

[0027] Determine the scene classification information of the rain and snow natural disaster scene faced by the vehicle based on the video image data;

[0028] When the scene classification information indicates that the rain and snow natural disaster scene faced by the vehicle is a snowfall scene, it is predicted that the target disaster category includes rain and snow natural disasters, and the target disaster level is severe;

[0029] When the scene classification information indicates that the rain and snow natural disaster scene faced by the vehicle includes a rainfall scene, the target disaster category is predicted to be a rain and snow natural disaster; and the target disaster level is determined based on the rainfall amount indicated by the rainfall sensor data.

[0030] In some possible implementations, predicting a first disaster prediction result based on the video image data, the voice data, and the millimeter wave radar data includes:

[0031] Predicting a first disaster prediction result based on the video image data, the voice data, and the millimeter wave radar data according to the first execution frequency;

[0032] Based on the video image data and rainfall sensor data, the second disaster prediction results are predicted, including:

[0033] A second disaster prediction result is predicted based on the video image data and the rainfall sensor data according to the second execution frequency; wherein the first execution frequency is higher than the second execution frequency.

[0034] In some possible implementations, after performing fusion prediction based on multi-source heterogeneous data to determine the vehicle disaster prediction result, the vehicle camping disaster warning method further includes:

[0035] Based on the disaster prediction results, target disaster response measures for the vehicle are determined, and components in the vehicle are controlled to execute the target disaster response measures.

[0036] Based on the same inventive concept, in a second aspect, an embodiment of the present application provides a vehicle camping disaster warning device, the vehicle camping disaster warning device comprising:

[0037] A first acquisition module is used to acquire multi-source heterogeneous data collected by the vehicle, where the multi-source heterogeneous data includes video image data, voice data, millimeter wave radar data, and rainfall sensor data;

[0038] A first determination module is used to perform fusion prediction based on multi-source heterogeneous data to determine a disaster prediction result for the vehicle, where the disaster prediction result is used to indicate a target disaster category faced by the vehicle and a target disaster level corresponding to the target disaster category, where the target disaster category includes at least one of a wild animal attack disaster, an illegal invasion disaster, and a rain and snow natural disaster;

[0039] The second determination module is used to determine the target disaster response measures of the vehicle based on the disaster prediction results, and control the components in the vehicle to execute the target disaster response measures.

[0040] In a third aspect, an embodiment of the present application provides a vehicle camping disaster warning device, the vehicle camping disaster warning device comprising:

[0041] a processor and a memory storing computer program instructions;

[0042] When the processor executes the computer program instructions, the vehicle camping disaster warning method provided in any one of the above-mentioned embodiments of the present application is implemented.

[0043] In a fourth aspect, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon, and when the computer program instructions are executed by a processor, a vehicle camping disaster warning method as provided in any one of the above-mentioned embodiments of the present application is implemented.

[0044] In a fifth aspect, an embodiment of the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes a vehicle camping disaster warning method as provided in any one of the above-mentioned embodiments of the present application.

[0045] In a sixth aspect, an embodiment of the present application provides a vehicle, the vehicle comprising at least one of the following: a vehicle camping disaster warning device as provided in the above-mentioned embodiment of the present application; a vehicle camping disaster warning device as provided in the above-mentioned embodiment of the present application; a computer-readable storage medium as provided in the above-mentioned embodiment of the present application; a computer program product as provided in the above-mentioned embodiment of the present application.

[0046] The embodiments of the present application provide a vehicle camping disaster warning method, device, equipment, medium, product and vehicle, which obtains multi-source heterogeneous data collected by the vehicle, and the multi-source heterogeneous data includes video image data, voice data, millimeter wave radar data and rainfall sensor data. Then, based on the multi-source heterogeneous data, a fusion prediction is performed to determine the disaster prediction result of the vehicle, that is, the target disaster category faced by the vehicle and the target disaster level corresponding to the target disaster category, and the target disaster category includes at least one of a wild animal attack disaster, an illegal invasion disaster and a rain and snow natural disaster.

[0047] From the above description, it can be seen that the vehicle camping disaster warning method, device, equipment, medium, product and vehicle of the embodiment of the present application, in view of the problems such as sudden beast attacks, illegal invasions, natural disasters, etc. in the car camping scene, collects multi-source heterogeneous data, and performs fusion and association processing based on the above multi-source heterogeneous data to comprehensively and comprehensively judge the current disaster scene category and danger level, so as to carry out corresponding protection later. In this way, the embodiment of the present application can timely and accurately realize disaster warning for various outdoor disasters in the car outdoor camping scene, thereby facilitating the effective realization of active protection of the car, and thus fully improving the personal and property safety of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0049] Figure 1 It is a flowchart of a vehicle camping disaster warning method provided by an embodiment of the present application;

[0050] Figure 2 This is one of the algorithm structure diagrams of the vehicle camping disaster warning method provided in one embodiment of the present application;

[0051] Figure 3 This is the second algorithm structure diagram of the vehicle camping disaster warning method provided in one embodiment of the present application;

[0052] Figure 4 This is a schematic diagram of the system architecture of a vehicle camping disaster warning method provided by an embodiment of the present application;

[0053] Figure 5 It is a structural schematic diagram of a vehicle camping disaster warning device provided in one embodiment of the present application;

[0054] Figure 6 It is a structural schematic diagram of a vehicle camping disaster warning device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0055] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0056] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article 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, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0057] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0058] As mentioned in the background technology section, the current disaster warning scheme for car camping scenes has problems such as single protection function and high false alarm rate, which cannot effectively protect the personal and property safety of users. Specifically, the current outdoor camping protection system can only detect wild animals, but cannot warn people and natural disasters such as rain and snow. Therefore, there are the above defects, so there is an urgent need for an early warning scheme that can effectively warn and actively protect against disaster problems.

[0059] In view of the above, in order to solve the problems of the prior art, the embodiments of the present application provide a vehicle camping disaster warning method, device, equipment, medium, product and vehicle. It should be noted that the embodiments provided in the present application are not intended to limit the scope of the present application.

[0060] The following first introduces the vehicle camping disaster warning method provided in the embodiment of the present application.

[0061] Figure 1 The schematic diagram of the process of the vehicle camping disaster warning method provided by an embodiment of the present application is shown. The vehicle camping disaster warning method is applied to an electronic device, such as a vehicle or a control component in a vehicle. Figure 1 As shown, the vehicle camping disaster warning method includes the following steps:

[0062] S110, acquiring multi-source heterogeneous data collected by the vehicle, where the multi-source heterogeneous data includes video image data, voice data, millimeter wave radar data, and rainfall sensor data;

[0063] S120, performing fusion prediction based on multi-source heterogeneous data to determine a disaster prediction result for the vehicle. The disaster prediction result is used to indicate a target disaster category faced by the vehicle and a target disaster level corresponding to the target disaster category. The target disaster category includes at least one of a wildlife attack disaster, an illegal invasion disaster, and a rain and snow natural disaster.

[0064] The embodiment of the present application provides a vehicle camping disaster warning method, which obtains multi-source heterogeneous data collected by the vehicle, and the multi-source heterogeneous data includes video image data, voice data, millimeter wave radar data and rainfall sensor data. Then, based on the multi-source heterogeneous data, fusion prediction is performed to determine the disaster prediction result of the vehicle, that is, the target disaster category faced by the vehicle and the target disaster level corresponding to the target disaster category. The target disaster category includes at least one of a wild animal attack disaster, an illegal invasion disaster and a rain and snow natural disaster.

[0065] From the above description, it can be seen that a vehicle camping disaster warning method of an embodiment of the present application, in view of the problems such as sudden beast attacks, illegal invasions, natural disasters, etc. in the car camping scene, collects multi-source heterogeneous data, and performs fusion and association processing based on the above multi-source heterogeneous data to comprehensively and comprehensively judge the current disaster scene category and danger level, so as to carry out corresponding protection later. In this way, the embodiment of the present application can timely and accurately realize disaster warnings for various outdoor disasters in the car outdoor camping scene, thereby facilitating the effective realization of active protection of the car, thereby fully improving the personal and property safety of users.

[0066] The specific implementation of the above steps 110 to 120 is described in detail below.

[0067] In S110, multi-source heterogeneous data collected by the vehicle is acquired, where the multi-source heterogeneous data includes video image data, voice data, millimeter wave radar data, and rainfall sensor data.

[0068] In an actual car camping scenario, the above video image data can be acquired through a camera on the vehicle to collect visual information of the vehicle's surroundings. When people, wild animals, or rainy or snowy weather appear, the camera can accurately capture the corresponding video image data.

[0069] The above-mentioned voice data may come from sensors such as microphones inside the vehicle, thereby realizing voice detection and collection in the vehicle's surrounding environment.

[0070] The above-mentioned millimeter wave radar sensor can be a millimeter wave radar sensor installed in the vehicle. The millimeter wave radar data can be used to detect information such as the distance and relative speed between the vehicle and surrounding objects.

[0071] The above-mentioned rain sensor data may be based on a rain sensor provided in the vehicle. The rain sensor may be integrated on the front windshield of the vehicle or other positions of the vehicle, so as to better realize the detection of rainfall in the surrounding environment of the vehicle.

[0072] In S120, during the specific implementation, a fusion prediction is performed based on multi-source heterogeneous data to determine a disaster prediction result for the vehicle. The disaster prediction result is used to indicate the target disaster category faced by the vehicle and the target disaster level corresponding to the target disaster category. The target disaster category includes at least one of a wildlife attack disaster, an illegal invasion disaster, and a rain and snow natural disaster.

[0073] In this embodiment, the video image data, voice data, millimeter wave radar data and rain sensor data are used to provide comprehensive environmental perception capabilities for the vehicle. Therefore, after obtaining the video image data, voice data, millimeter wave radar data and rain sensor data, data fusion processing is performed based on these multi-source heterogeneous data, so that the target disaster category currently faced by the vehicle and the target disaster level corresponding to the target disaster category can be comprehensively and effectively determined.

[0074] Here, combined with the actual outdoor camping scene, the disaster types that vehicles may encounter are divided into three categories, namely: wild animal attack disasters, illegal invasion disasters, and rain and snow natural disasters. Different data features of different data can correspond to different target disaster categories to which the vehicle belongs. The judgment method of the disaster level under different target disaster categories can be set separately. The disaster level can include minor or severe, etc., which is conducive to achieving more accurate disaster warning.

[0075] Among them, when realizing the above-mentioned data fusion processing and disaster result prediction, it can be implemented by using pre-selected and trained deep learning models, machine learning models, or fusion algorithms based on attention mechanisms, etc., and no strict restrictions are made here.

[0076] It should be added that in most scenarios the vehicle may also be in a normal camping environment. Therefore, when the vehicle is not facing a disaster in the wild, the disaster prediction result may not be output, or a disaster result indicating that the target disaster category is other may be output to indicate that there is no disaster.

[0077] In some possible implementations, in order to further improve the active protection of the vehicle against wild camping disasters, after the above-mentioned fusion prediction based on multi-source heterogeneous data is performed to determine the disaster prediction result of the vehicle, the vehicle camping disaster warning method further includes:

[0078] Based on the disaster prediction results, target disaster response measures for the vehicle are determined, and components in the vehicle are controlled to execute the target disaster response measures.

[0079] In this embodiment, different disaster response measures can be set in advance for different disaster levels under the three disaster categories of wild animal attack disasters, illegal invasion disasters, and rain and snow natural disasters, and the data can be stored to facilitate the subsequent effective protection of the car.

[0080] In this way, after obtaining the above disaster prediction results, the target disaster response measures corresponding to the current vehicle are specifically queried according to the target disaster category faced by the vehicle indicated by the disaster prediction results and the target disaster level corresponding to the target disaster category. The components in the vehicle are controlled to execute the target disaster response measures, thereby effectively improving the personal and property safety of users in the car.

[0081] It should be added that when controlling the components in the vehicle to execute the target disaster response measures for active protection of the vehicle, the components in the vehicle, such as the alarm (horn), can control the horn to emit a blast to remind the owner, or to scare away illegal intruders, wild animals, etc. The vehicle components can also be, for example, vehicle door and lathe door locks, so that during active protection, the vehicle's protection level can be improved by controlling the vehicle's doors and windows to be locked.

[0082] Optionally, in some feasible embodiments of the present application, considering that in the fusion processing of multi-source heterogeneous data, there are differences in the data required for different disaster categories and the data processing methods, the disaster prediction result is defined to include the first disaster prediction result and / or the second disaster prediction result; based on the fusion prediction of the multi-source heterogeneous data, the disaster prediction result of the vehicle is determined to be obtained, including:

[0083] Based on the video image data, the voice data and the millimeter wave radar data, a first disaster prediction result is predicted, and the first disaster prediction result is associated with whether the vehicle faces a wild animal attack disaster and / or an illegal invasion disaster, and the corresponding disaster level;

[0084] and / or,

[0085] Based on at least one of the video image data and the rainfall sensor data, a second disaster prediction result is predicted, and the second disaster prediction result is associated with whether the vehicle is facing a natural disaster of rain or snow and a corresponding disaster level.

[0086] In this embodiment, it is determined whether the vehicle is facing a wild animal attack disaster and / or an illegal invasion disaster based on video image data, voice data and millimeter wave radar data respectively. If a wild animal attack disaster and / or an illegal invasion disaster exists, the specific disaster category and the corresponding disaster level are determined.

[0087] And, based on at least one of the video image data and the rain sensor data, it is judged whether the vehicle is facing a natural disaster of rain or snow, and if there is a natural disaster of rain or snow, its disaster response is further judged.

[0088] In this way, different data are screened for different disaster categories for data processing, so that the corresponding disaster prediction results can be obtained more accurately and effectively, which is conducive to more fully improving the safety and protection of car camping outdoors.

[0089] In order to more accurately process data and make disaster judgments for wildlife attack disasters and / or illegal invasion disasters, please refer to the following Figure 2 . Figure 2 This is one of the algorithm structure diagrams of the vehicle camping disaster warning method provided in one embodiment of the present application. Optionally, in some feasible embodiments of the present application, the first disaster prediction result is predicted based on the video image data, voice data and millimeter wave radar data, including:

[0090] Predicting the object category to which the target object belongs based on the video image data and the voice data respectively, to obtain a first category prediction result and a second category prediction result;

[0091] Performing data fusion processing based on the first category prediction result, the second category prediction result and the millimeter wave radar data to obtain the predicted category and predicted distance of the target object;

[0092] Based on the prediction type and prediction distance, the first disaster prediction result is determined.

[0093] In specific implementation, the object category to which the target object belongs can first be predicted based on the video image data and voice data collected by the vehicle, thereby obtaining a first category prediction result corresponding to the video image data and a second category prediction result corresponding to the voice data.

[0094] Then, the first type prediction result, the second type prediction result and the millimeter wave radar data are fused and processed to finally obtain the predicted type and predicted distance of the target object. The predicted type and predicted distance can be used to determine the target disaster category of the vehicle, including wild animal attack disaster and / or illegal invasion disaster, and the corresponding disaster level.

[0095] It should be added that after data fusion processing of the first type of prediction results, the second type of prediction results and the millimeter wave radar data, it is of course possible to conclude that the vehicle is currently not in a situation of wild animal attack disaster and / or illegal invasion disaster. The focus of the embodiments of the present application is on how to conduct disaster warning when a disaster occurs.

[0096] It should be noted that the target object may be an object appearing in the video image, and the type of the object may be, for example, a person, an animal, a plant, etc., and is not strictly limited herein.

[0097] Continue as Figure 2 As shown, more specifically, considering the actual prediction effect of the video image data and the voice data, the above prediction of the type of the target object based on the video image data and the voice data is performed to obtain the first type prediction result and the second type prediction result, including:

[0098] Based on the video image data, the type of the target object is predicted by the YOLO model to obtain a first type prediction result;

[0099] And / or, based on the speech data, the type of the target object is predicted by an LSTM long short-term memory network model to obtain a second type prediction result.

[0100] The YOLO (You Only Look Once) model is a popular target detection algorithm with its fast speed and high accuracy. Figure 2 The YOLO model can significantly improve the frame rate and reduce the model size while maintaining a high accuracy, thereby significantly improving the speed of loading and running the model in the vehicle, thereby helping to effectively control the vehicle's energy consumption.

[0101] In speech data processing, the LSTM (Long Short-Term Memory) model can capture contextual information at distant moments in the speech signal through its long-term memory capability and ensure a global understanding of speech features, thereby significantly improving the accuracy of speech recognition.

[0102] Therefore, in this embodiment, the YOLO model and the LSTM model are used respectively to realize the video image data processing and the voice data processing, so as to fully guarantee the accuracy of the above-mentioned first type of prediction results and the above-mentioned second type of prediction results. At the same time, due to the lightweight characteristics of the selected model, it is also beneficial to ensure that the implementation of the disaster warning plan can control the vehicle energy loss within a certain range.

[0103] Optionally, in some feasible embodiments of the present application, please continue to refer to Figure 2 The above data fusion processing based on the first type prediction result, the second type prediction result and the millimeter wave radar data to obtain the predicted type and predicted distance of the target object includes:

[0104] Data fusion is performed based on the first category prediction results, the second category prediction results and the millimeter wave radar data, and the target object is converted to a bird's-eye view space to obtain the predicted category and predicted distance.

[0105] In specific implementation, data fusion based on the first type of prediction results, the second type of prediction results and the millimeter wave radar data can be, for example, based on Kalman filtering to achieve data fusion and prediction. In addition, in this process, the target object mapping is converted to the bird's eye view (BEV) space, which can make the objects and spatial relationships in the scene more clearly visible, thereby providing a more global environmental perception capability. This can provide better prediction effects and prediction accuracy for the final determination of the prediction type and prediction distance.

[0106] More specifically, combined with Figure 2 As shown, the types of the target objects may include people, small wild animals, and large wild animals; based on the predicted types, the target disaster categories faced by the vehicle are determined, including:

[0107] In a case where the predicted category indicates that the target object belongs to a human being, determining that the target disaster category includes an illegal invasion disaster;

[0108] In a case where the predicted category indicates that the object category to which the target object belongs is a small wild animal or a large wild animal, it is determined that the target disaster category includes a wild animal attack disaster.

[0109] It should be added that, in some other embodiments, the above object type can also be Figure 2 Others shown, such as rolling stones, floating grass fluff, etc., are target objects of this type and can be considered irrelevant to disaster warning, and will not be paid too much attention to here.

[0110] Optionally, in some feasible embodiments of the present application, determining the first disaster prediction result based on the prediction type and the prediction distance includes:

[0111] Based on the predicted category, determine the target disaster category faced by the vehicle;

[0112] The target disaster level is determined based on the prediction type, prediction distance and disaster level classification rules corresponding to the target disaster category. The disaster level classification rules include the correspondence between different distance range intervals and disaster levels.

[0113] In this embodiment, by adopting different disaster level classification rules for different disaster categories, the corresponding disaster level classification rules can be effectively obtained. In this way, based on the predicted type, predicted distance and the corresponding disaster level classification rules, the corresponding target disaster level under different disaster categories can be determined more reasonably.

[0114] For example, Figure 2 As shown in the figure, when determining the target disaster level, if the target disaster category is a wild animal attack disaster, if the predicted type of the target object is a large wild animal, the predicted distance is 30 meters, and the corresponding disaster level classification rule is: if the distance is less than 50 meters, the disaster level can be judged as severe. In this case, the corresponding target disaster level can be determined to be severe.

[0115] In another example, when the target disaster category is a wild animal attack disaster, if the predicted type of the target object is a small wild animal, the predicted distance is 30 meters, and the corresponding disaster level classification rule is: if the distance is less than 50 meters, the disaster level can be judged as minor. In this case, the corresponding target disaster level can be determined to be minor.

[0116] In another example, when the target disaster category is illegal intrusion disaster, if the predicted type of the target object is human, the predicted distance is 20 meters, and the corresponding disaster level classification rule is: if the distance is less than 50 meters but not less than 30 meters, the disaster level can be judged as minor, and if the distance is less than 30 meters, the disaster level can be judged as minor. In this case, it can be determined that the target disaster level corresponding to the current illegal intrusion disaster is severe, and corresponding response measures can be taken in a targeted manner to provide timely protection.

[0117] Such targeted classification and grading methods will help to achieve more accurate disaster warnings and reduce false alarm rates.

[0118] This embodiment adopts different disaster level classification rules for different disaster categories, which is also conducive to taking different targeted disaster response measures in subsequent classification, thereby achieving the purpose of handling disasters of different degrees in a classified strategy. This embodiment not only reduces the probability of false alarms, but also handles disasters more targetedly, thereby fully improving the user's camping safety and camping experience.

[0119] Optionally, in some feasible embodiments of the present application, after introducing the above-mentioned data processing method for wildlife attack disasters and / or illegal invasion disaster categories, the data processing and disaster judgment method for rain and snow natural disasters will be introduced in detail.

[0120] Specific combination Figure 3 , Figure 3 This is the second algorithm structure diagram of the vehicle camping disaster warning method provided by an embodiment of the present application. For heavy rain and heavy snow disasters, it is necessary to judge the disaster level and disaster category through video image data and rainfall sensor data. Figure 3 As shown, the second disaster prediction result is predicted based on at least one of the video image data and the rainfall sensor data, including:

[0121] Determine the scene classification information of the rain and snow natural disaster scene faced by the vehicle based on the video image data;

[0122] When the scene classification information indicates that the rain and snow natural disaster scene faced by the vehicle is a snowfall scene, it is predicted that the target disaster category includes rain and snow natural disasters, and the target disaster level is severe;

[0123] When the scene classification information indicates that the rain and snow natural disaster scene faced by the vehicle includes a rainfall scene, the target disaster category is predicted to be a rain and snow natural disaster; and the target disaster level is determined based on the rainfall amount indicated by the rainfall sensor data.

[0124] In the specific implementation, firstly, the scene classification information of the rain and snow natural disaster scene faced by the vehicle is determined based on the video image data, for example, whether the current vehicle environment is in a snowfall scene or a rainfall scene. Among them, when performing the specific scene classification, considering that ResNet (Residual Network) has good generalization ability and model performance, Figure 3 As shown, the correctness of scene classification can be ensured by selecting the ResNet network.

[0125] After determining the scene classification information of the rain and snow natural disaster scene faced by the vehicle based on the video image data, if the scene classification information indicates that the rain and snow natural disaster scene faced by the vehicle is a snowfall scene, it can be directly predicted that the target disaster category includes rain and snow natural disasters, and the corresponding target disaster level can be determined to be severe.

[0126] When the scene classification information indicates that the rain and snow natural disaster scene faced by the vehicle includes a rainfall scene, the target disaster category is predicted to be a rain and snow natural disaster. Then, the target disaster level is determined based on the rainfall amount indicated by the rain sensor data. For example, if the rainfall sensor data indicates that the hourly rainfall exceeds a preset threshold, it can be determined that the disaster level corresponding to the current rain and snow natural disaster is severe; if it does not exceed the preset threshold, it can be determined that the disaster level corresponding to the current rain and snow natural disaster is minor. If the subsequent hourly rainfall increases and exceeds the preset threshold, the disaster level corresponding to the vehicle's rain and snow natural disaster can be changed from minor to severe.

[0127] Optionally, in some feasible embodiments of the present application, based on the video image data, the voice data and the millimeter wave radar data, a first disaster prediction result is predicted, including:

[0128] Predicting a first disaster prediction result based on the video image data, the voice data, and the millimeter wave radar data according to the first execution frequency;

[0129] Based on the video image data and rainfall sensor data, the second disaster prediction results are predicted, including:

[0130] A second disaster prediction result is predicted based on the video image data and the rainfall sensor data according to the second execution frequency; wherein the first execution frequency is higher than the second execution frequency.

[0131] In this embodiment, taking the nighttime car camping scene as an example, due to the high randomness and timeliness of the appearance of people or wild animals at night, a higher task execution frequency (for example, once every minute / ten seconds) is used to perform more frequent fusion processing and disaster prediction on the above video image data, voice data and millimeter wave radar data. In this way, in the face of illegal intrusion disasters and / or wild animal attack disasters, the vehicle can quickly issue disaster warnings and corresponding protection.

[0132] As for natural disasters such as rain and snow, it often takes a long time for a disaster to occur. Therefore, this application adopts a relatively lower task execution frequency (for example, once every ten minutes) without affecting the timeliness and accuracy of rain and snow natural disaster warnings. This can greatly reduce the energy consumption loss caused by frequent data processing and prediction tasks, thereby helping to save vehicle energy.

[0133] In order to facilitate understanding of the vehicle camping disaster warning method provided in the above embodiment, the above method is described below using a specific scenario embodiment. Figure 4 It is a schematic diagram of the system architecture of a vehicle camping disaster warning method provided in one embodiment of the present application.

[0134] like Figure 4As shown, when the vehicle is camping and resting in the wild, there are problems such as sudden attacks by wild animals, illegal invasions, and natural disasters. Figure 4 The disaster warning system architecture can sense danger in advance and take corresponding measures to effectively improve the user's camping experience and safety.

[0135] In specific implementation, this disaster warning system fuses and associates multi-source heterogeneous data such as voice data, video image data, millimeter wave radar data, and rainfall sensor data extracted by multiple types of sensors in the vehicle. In this way, the current disaster scene category and danger level can be determined based on the fused multi-source data. Then, the system takes corresponding disaster response measures based on the disaster scene category and danger level.

[0136] Disaster response measures can include, for example, Figure 4 As shown: When the disaster category is a wild animal attack disaster and the disaster level is severe, the horn can be controlled to sound, the vehicle high beam can be turned on, the windows can be raised, and the doors and windows can be locked to fully protect the car. When the disaster category is a wild animal attack disaster and the disaster level is minor, the windows can be controlled to rise and the vehicle's colorful lights can be flashed intermittently to drive away wild animals around the vehicle.

[0137] If the disaster type is illegal intrusion disaster and the disaster level is severe, the horn will be controlled to sound, the vehicle high beam will be turned on, the windows will be raised, and the doors and windows will be locked to fully protect the car and ensure personal safety. If the disaster type is illegal intrusion disaster and the disaster level is minor, the windows will be controlled to be raised, the doors and windows will be locked, and the vehicle's colorful lights will flash intermittently to warn.

[0138] If the disaster category is a rain or snow disaster and the disaster level is severe, turn on the wipers, turn on the air conditioner to a suitable temperature, sound the horn for warning, and raise the windows to fully protect the car. If the disaster category is a rain or snow disaster and the disaster level is minor, turn on the wipers and raise the windows.

[0139] In this scenario embodiment, the system can effectively warn and respond to illegal intrusion disasters, natural rain and snow disasters, and wild animal attacks that occur during camping; it uses multi-source heterogeneous sensor data fusion to fuse and process surrounding environmental targets, and can robustly perceive and semantically understand surrounding environmental information. It uses a hierarchical strategy to handle disasters of different degrees, reducing the probability of false alarms and handling disasters in a more targeted manner. It greatly improves user camping safety and camping experience.

[0140] Based on the vehicle camping disaster warning method provided in the above embodiment, out of the same inventive concept, the present application also provides a vehicle camping disaster warning device corresponding to the above vehicle camping disaster warning method. Figure 5 The vehicle camping disaster warning device is introduced in detail.

[0141] Figure 5 A schematic structural diagram of a vehicle camping disaster warning device provided in one embodiment of the present application is shown. Figure 5 The vehicle camping disaster warning device 500 shown includes:

[0142] A first acquisition module 510 is used to acquire multi-source heterogeneous data collected by the vehicle, where the multi-source heterogeneous data includes video image data, voice data, millimeter wave radar data, and rainfall sensor data;

[0143] The first determination module 520 is used to perform fusion prediction based on multi-source heterogeneous data to determine the disaster prediction result of the vehicle. The disaster prediction result is used to indicate the target disaster category faced by the vehicle and the target disaster level corresponding to the target disaster category. The target disaster category includes at least one of a wildlife attack disaster, an illegal invasion disaster, and a rain and snow natural disaster.

[0144] The embodiment of the present application provides a vehicle camping disaster warning device, which acquires multi-source heterogeneous data collected by the vehicle by setting corresponding functional modules. The multi-source heterogeneous data includes video image data, voice data, millimeter wave radar data, and rainfall sensor data. Then, based on the multi-source heterogeneous data, a fusion prediction is performed to determine the disaster prediction result of the vehicle, that is, the target disaster category faced by the vehicle and the target disaster level corresponding to the target disaster category. The target disaster category includes at least one of a wild animal attack disaster, an illegal invasion disaster, and a rain and snow natural disaster. Finally, based on the disaster prediction result, the target disaster response measures of the vehicle are determined, and the components in the vehicle are controlled to execute the target disaster response measures.

[0145] From the above description, it can be seen that the vehicle camping disaster warning device of the embodiment of the present application, in view of the problems such as sudden beast attacks, illegal invasions, natural disasters, etc. in the car camping scene, collects multi-source heterogeneous data, and performs fusion and association processing based on the above multi-source heterogeneous data to comprehensively and comprehensively judge the current disaster scene category and danger level in order to carry out corresponding protection. In this way, the embodiment of the present application can timely and accurately realize disaster warnings for various outdoor disasters in the car outdoor camping scene, thereby effectively realizing active protection of the car, and thus fully improving the personal and property safety of users.

[0146] Optionally, in some feasible embodiments of the present application, the disaster prediction result is defined to include a first disaster prediction result and / or a second disaster prediction result; the first determination module 520 performs fusion prediction based on multi-source heterogeneous data to determine the disaster prediction result of the vehicle, including:

[0147] The first prediction submodule can be used to predict a first disaster prediction result based on the video image data, the voice data and the millimeter wave radar data, and the first disaster prediction result is associated with whether the vehicle faces a wild animal attack disaster and / or an illegal invasion disaster, and the corresponding disaster level;

[0148] and / or,

[0149] The second prediction submodule can be used to predict a second disaster prediction result based on at least one of the video image data and the rainfall sensor data. The second disaster prediction result is associated with whether the vehicle is facing a natural disaster of rain or snow and the corresponding disaster level.

[0150] Optionally, in some feasible embodiments of the present application, the first prediction submodule predicts a first disaster prediction result based on the video image data, the voice data and the millimeter wave radar data, including:

[0151] The first obtaining unit may be used to predict the object category to which the target object belongs based on the video image data and the voice data, respectively, to obtain a first category prediction result and a second category prediction result;

[0152] The second obtaining unit may be used to perform data fusion processing based on the first type prediction result, the second type prediction result and the millimeter wave radar data to obtain the predicted type and predicted distance of the target object;

[0153] The first determination unit may be configured to determine a first disaster prediction result based on the prediction type and the prediction distance.

[0154] Optionally, in some feasible embodiments of the present application, the first obtaining unit predicts the type of the target object based on the video image data and the voice data respectively to obtain a first type prediction result and a second type prediction result, including:

[0155] The first prediction subunit may be used to predict the type of the target object through the YOLO model based on the video image data to obtain a first type prediction result;

[0156] And / or, the second prediction submodule can be used to predict the type of the target object based on the voice data through the LSTM long short-term memory network model to obtain a second type prediction result.

[0157] Optionally, in some feasible embodiments of the present application, the second obtaining unit performs data fusion processing based on the first category prediction result, the second category prediction result and the millimeter wave radar data to obtain the predicted category and predicted distance of the target object, including:

[0158] Data fusion is performed based on the first category prediction results, the second category prediction results and the millimeter wave radar data, and the target object is converted to a bird's-eye view space to obtain the predicted category and predicted distance.

[0159] Optionally, in some feasible embodiments of the present application, the first determining unit determines the first disaster prediction result based on the prediction type and the prediction distance, including:

[0160] A first determination subunit may be used to determine a target disaster category faced by the vehicle based on the predicted category;

[0161] The second determination subunit can be used to determine the target disaster level based on the predicted type, predicted distance and disaster level classification rules corresponding to the target disaster category. The disaster level classification rules include the correspondence between different distance range intervals and disaster levels.

[0162] Optionally, in some feasible embodiments of the present application, the object types include people, small wild animals and large wild animals; the first determination subunit determines the target disaster category faced by the vehicle based on the predicted type, including:

[0163] In a case where the predicted category indicates that the target object belongs to a human being, determining that the target disaster category includes an illegal invasion disaster;

[0164] In a case where the predicted category indicates that the object category to which the target object belongs is a small wild animal or a large wild animal, it is determined that the target disaster category includes a wild animal attack disaster.

[0165] Optionally, in some feasible embodiments of the present application, the second prediction submodule predicts a second disaster prediction result based on at least one of the video image data and the rainfall sensor data, including:

[0166] The second determination unit may be used to determine scene classification information of a rain and snow natural disaster scene faced by the vehicle based on the video image data;

[0167] The first prediction unit may be configured to predict that the target disaster category includes a rain and snow natural disaster and the target disaster level is severe when the scene classification information indicates that the rain and snow natural disaster scene faced by the vehicle is a snowfall scene;

[0168] The second prediction unit can be used to predict that the target disaster category is a rain and snow natural disaster when the scene classification information indicates that the rain and snow natural disaster scene faced by the vehicle includes a rainfall scene; and determine the target disaster level based on the rainfall amount indicated by the rainfall sensor data.

[0169] Optionally, in some feasible embodiments of the present application, the first prediction submodule predicts a first disaster prediction result based on the video image data, the voice data and the millimeter wave radar data, including:

[0170] Predicting a first disaster prediction result based on the video image data, the voice data, and the millimeter wave radar data according to the first execution frequency;

[0171] The second prediction submodule predicts a second disaster prediction result based on the video image data and the rainfall sensor data, including:

[0172] A second disaster prediction result is predicted based on the video image data and the rainfall sensor data according to the second execution frequency; wherein the first execution frequency is higher than the second execution frequency.

[0173] In some possible implementations, after performing fusion prediction based on multi-source heterogeneous data and determining a disaster prediction result for the vehicle, the vehicle camping disaster warning device further includes:

[0174] The second determination module can be used to determine the target disaster response measures of the vehicle based on the disaster prediction results, and control the components in the vehicle to execute the target disaster response measures.

[0175] Based on the vehicle camping disaster warning method provided in the above embodiment, out of the same inventive concept, the present application also provides a disaster warning device corresponding to the above vehicle camping disaster warning method. Figure 6 A detailed introduction to vehicle camping disaster warning equipment.

[0176] See below Figure 6 , Figure 6 It is a structural schematic diagram of a vehicle camping disaster warning device provided in one embodiment of the present application.

[0177] The vehicle camping disaster warning device may include a processor 601 and a memory 602 storing computer program instructions.

[0178] Specifically, the processor 601 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0179] The memory 602 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 602 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 602 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 602 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 602 is a non-volatile solid-state memory.

[0180] The memory may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical or other physical / tangible memory storage devices. Thus, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0181] The processor 601 implements any one of the vehicle camping disaster warning methods in the above embodiments by reading and executing computer program instructions stored in the memory 602 .

[0182] In one example, the data vehicle camping disaster warning device may also include a communication interface 603 and a bus 610. Figure 6 As shown, the processor 601, the memory 602, and the communication interface 603 are connected via a bus 610 and communicate with each other.

[0183] The communication interface 603 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0184] Bus 610 includes hardware, software or both, and the components of vehicle camping disaster warning equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-side bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 610 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the present application considers any suitable bus or interconnection.

[0185] The vehicle camping disaster warning device executes the vehicle camping disaster warning method in the embodiment of the present application, thereby realizing the vehicle camping disaster warning method described in the embodiment of the present application.

[0186] In addition, in combination with the vehicle camping disaster warning method in the above embodiment, the embodiment of the present application can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by the processor, any one of the vehicle camping disaster warning methods in the above embodiment is implemented.

[0187] Based on the vehicle camping disaster warning method in the above-mentioned embodiment, an embodiment of the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the vehicle camping disaster warning method provided in any one of the above-mentioned embodiments of the present application.

[0188] In addition, an embodiment of the present application further provides a vehicle, the vehicle comprising at least one of the following: a vehicle camping disaster warning device as provided in the above-mentioned embodiment of the present application; a vehicle camping disaster warning device as provided in the above-mentioned embodiment of the present application; a computer-readable storage medium as provided in the above-mentioned embodiment of the present application; a computer program product as provided in the above-mentioned embodiment of the present application.

[0189] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0190] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0191] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0192] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0193] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A vehicle camping disaster warning method, characterized in that: The method comprises: Acquire multi-source heterogeneous data collected by the vehicle, wherein the multi-source heterogeneous data includes video image data, voice data, millimeter wave radar data, and rainfall sensor data; Based on the multi-source heterogeneous data, a fusion prediction is performed to determine a disaster prediction result for the vehicle, and the disaster prediction result is used to indicate a target disaster category faced by the vehicle and a target disaster level corresponding to the target disaster category, and the target disaster category includes at least one of a wildlife attack disaster, an illegal invasion disaster, and a rain and snow natural disaster.

2. The method according to claim 1, characterized in that The disaster prediction result is defined to include a first disaster prediction result and / or a second disaster prediction result; and the fusion prediction based on the multi-source heterogeneous data is performed to determine the disaster prediction result of the vehicle, including: Based on the video image data, the voice data and the millimeter wave radar data, a first disaster prediction result is predicted, wherein the first disaster prediction result is associated with whether the vehicle faces a wild animal attack disaster and / or an illegal invasion disaster, and a corresponding disaster level; and / or, Based on at least one of the video image data and the rainfall sensor data, a second disaster prediction result is predicted, and the second disaster prediction result is associated with whether the vehicle is facing a natural disaster of rain or snow and a corresponding disaster level.

3. The method according to claim 2, characterized in that The predicting and obtaining a first disaster prediction result based on the video image data, the voice data and the millimeter wave radar data includes: Predicting the object category to which the target object belongs based on the video image data and the voice data respectively, to obtain a first category prediction result and a second category prediction result; Performing data fusion processing based on the first category prediction result, the second category prediction result and the millimeter wave radar data to obtain the predicted category and predicted distance of the target object; Based on the prediction type and the prediction distance, the first disaster prediction result is determined.

4. The method according to claim 3, characterized in that The predicting the type of the target object based on the video image data and the voice data respectively to obtain a first type prediction result and a second type prediction result includes: Based on the video image data, predicting the type of the target object by using a YOLO model to obtain the first type prediction result; And / or, based on the voice data, predict the type of the target object through an LSTM long short-term memory network model to obtain the second type prediction result.

5. The method according to claim 3, characterized in that: The performing data fusion processing based on the first type prediction result, the second type prediction result and the millimeter wave radar data to obtain the predicted type and predicted distance of the target object includes: Data fusion is performed based on the first category prediction result, the second category prediction result and the millimeter wave radar data, and the target object is converted into a bird's-eye view space to obtain the predicted category and the predicted distance.

6. The method according to claim 3, characterized in that The determining the first disaster prediction result based on the prediction type and the prediction distance includes: Based on the predicted type, determining the target disaster category faced by the vehicle; The target disaster level is determined based on the prediction type, the prediction distance, and a disaster level classification rule corresponding to the target disaster category, wherein the disaster level classification rule includes a correspondence between different distance range intervals and disaster levels.

7. The method according to claim 6, characterized in that The object types include humans, small wild animals, and large wild animals; and determining the target disaster category faced by the vehicle based on the predicted type includes: In a case where the predicted category indicates that the target object belongs to a human being, determining that the target disaster category includes an illegal invasion disaster; In a case where the predicted category indicates that the target object belongs to a small wild animal or a large wild animal, it is determined that the target disaster category includes a wild animal attack disaster.

8. The method according to claim 2, characterized in that: The predicting and obtaining a second disaster prediction result based on at least one of the video image data and the rainfall sensor data includes: Determining scene classification information of the rain and snow natural disaster scene faced by the vehicle according to the video image data; When the scene classification information indicates that the rain and snow natural disaster scene faced by the vehicle is a snowfall scene, it is predicted that the target disaster category includes the rain and snow natural disaster, and the target disaster level is severe; When the scene classification information indicates that the rain and snow natural disaster scene faced by the vehicle includes a rainfall scene, the target disaster category is predicted to be the rain and snow natural disaster; and the target disaster level is determined based on the rainfall amount indicated by the rainfall sensor data.

9. The method according to any one of claims 2 to 8, characterized in that The predicting and obtaining a first disaster prediction result based on the video image data, the voice data and the millimeter wave radar data includes: Predicting the first disaster prediction result based on the video image data, the voice data, and the millimeter wave radar data according to a first execution frequency; The predicting and obtaining a second disaster prediction result based on the video image data and the rainfall sensor data includes: The second disaster prediction result is predicted based on the video image data and the rainfall sensor data according to a second execution frequency; wherein the first execution frequency is higher than the second execution frequency.

10. The method according to any one of claims 1 to 8, characterized in that: After performing fusion prediction based on the multi-source heterogeneous data to determine the disaster prediction result of the vehicle, the method further includes: Based on the disaster prediction result, target disaster response measures for the vehicle are determined, and components in the vehicle are controlled to execute the target disaster response measures.

11. A vehicle camping disaster warning device, characterized in that: The device comprises: A first acquisition module is used to acquire multi-source heterogeneous data collected by the vehicle, wherein the multi-source heterogeneous data includes video image data, voice data, millimeter wave radar data, and rainfall sensor data; The first determination module is used to perform fusion prediction based on the multi-source heterogeneous data to determine the disaster prediction result of the vehicle, and the disaster prediction result is used to indicate the target disaster category faced by the vehicle and the target disaster level corresponding to the target disaster category. The target disaster category includes at least one of a wildlife attack disaster, an illegal invasion disaster, and a rain and snow natural disaster.

12. A vehicle camping disaster warning device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, the vehicle camping disaster warning method as described in any one of claims 1-10 is implemented.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the vehicle camping disaster warning method according to any one of claims 1 to 10 is implemented.

14. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the vehicle camping disaster warning method as described in any one of claims 1-10.

15. A vehicle, characterized in that: The vehicle comprises at least one of the following: a vehicle camping disaster warning device as claimed in claim 11; a vehicle camping disaster warning device as claimed in claim 12; a computer-readable storage medium as claimed in claim 13; and a computer program product as claimed in claim 14.

Images