A vehicle temperature detection method, device, apparatus, electronic device, and medium

Abstract

Embodiments of the present application provide a vehicle temperature detection method, device, apparatus, electronic device and medium, and relate to the technical field of intelligent transportation. The specific implementation scheme is as follows: obtaining collected data of a passing vehicle in a road scene, and obtaining a thermal image of the vehicle obtained by thermal imaging; determining a target detection part of the vehicle in the collected data; determining a thermal region corresponding to the target detection part in the thermal image; and detecting whether the temperature of the target detection part of the vehicle is abnormal according to region information of the thermal region. The scheme provided by the embodiments of the present application can detect the temperature of the vehicle.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technology, and in particular to a vehicle temperature detection method, device, apparatus, electronic device, and medium. Background Technology

[0002] During vehicle operation, friction, collisions, and other factors can cause the vehicle temperature to overheat, potentially leading to spontaneous combustion, explosions, or other safety incidents. Therefore, it is necessary to monitor vehicle temperature to ensure safe operation. Summary of the Invention

[0003] The purpose of this application is to provide a vehicle temperature detection method, device, apparatus, electronic device, and medium for detecting the temperature of a vehicle. The specific technical solution is as follows:

[0004] This application provides a vehicle temperature detection method, the method comprising:

[0005] Data is collected from vehicles passing through a road scene, and a thermal image is obtained by thermal imaging of the vehicles.

[0006] Determine the target detection location of the vehicle in the collected data;

[0007] The thermal region corresponding to the target detection location is determined in the thermal map;

[0008] Based on the regional information of the thermal zone, the temperature of the target detection part of the vehicle is detected to be abnormal.

[0009] This application embodiment also provides a vehicle temperature detection device, the device comprising:

[0010] The data collector is used to acquire data on vehicles passing through a road scene and to send the acquired data to the main control board.

[0011] A thermal imaging module is used to obtain a thermal map of the vehicle by performing thermal imaging; and to send the thermal map to the main control board.

[0012] The main control board is used to determine the target detection part of the vehicle in the collected data; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region.

[0013] This application embodiment also provides a vehicle temperature detection device, the device comprising:

[0014] The heat map acquisition module is used to acquire data collected from vehicles passing through a road scene and to obtain a heat map obtained by thermal imaging of the vehicles.

[0015] The detection location determination module is used to determine the target detection location of the vehicle in the collected data;

[0016] A thermal region determination module is used to determine the thermal region corresponding to the target detection location in the thermal map.

[0017] The temperature anomaly detection module is used to detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal zone.

[0018] This application also provides an electronic device, including:

[0019] Memory, used to store computer programs;

[0020] The processor, when executing the program stored in the memory, implements the above-mentioned vehicle temperature detection method.

[0021] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described vehicle temperature detection method.

[0022] This application also provides a computer program product containing instructions that, when run on a computer, causes the computer to execute the above-described vehicle temperature detection method.

[0023] Beneficial effects of the embodiments in this application:

[0024] In the solution provided in this application embodiment, by collecting data of the road scene, it is determined that vehicles pass through the road scene and a heat map of the vehicles is obtained. The heat map contains temperature information, so the temperature of the vehicles can be detected based on the heat map.

[0025] Of course, implementing any product or method of this application does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0027] Figure 1 A schematic flowchart illustrating the first vehicle temperature detection method provided in this application embodiment;

[0028] Figure 2 A schematic flowchart illustrating the second vehicle temperature detection method provided in this application embodiment;

[0029] Figure 3A A schematic flowchart of the vehicle temperature detection method provided in the embodiments of this application;

[0030] Figure 3B A vehicle image captured by a data acquisition device provided in an embodiment of this application;

[0031] Figure 4 A flowchart illustrating a method for obtaining a conversion relationship provided in an embodiment of this application;

[0032] Figure 5 This is a schematic diagram of the structure of the first type of vehicle temperature detection device provided in the embodiments of this application;

[0033] Figure 6 This is a schematic diagram of the structure of the second type of vehicle temperature detection device provided in the embodiments of this application;

[0034] Figure 7 This is a schematic diagram of the structure of a vehicle temperature detection device provided in an embodiment of this application;

[0035] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0036] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0037] To detect the temperature of a vehicle, embodiments of this application provide a vehicle temperature detection method, device, apparatus, electronic device, and storage medium, which are described below.

[0038] In one embodiment of this application, see Figure 1 A flowchart of a vehicle temperature detection method is provided, which includes the following steps S101-S104.

[0039] Step S101: Obtain the collected data of vehicles passing through the road scene, and obtain the heat map obtained by thermal imaging of the vehicles.

[0040] The road scene refers to the location of the road on which vehicles travel, such as highway tunnels, bridge areas, long downhill slopes, interchanges, ramp entrances, or large mainline toll stations. The data acquisition equipment used to obtain the collected data and the thermal imaging equipment used to perform thermal imaging can be installed at any location on the aforementioned road, for example, in the middle section or at the entrance / exit of the road included in the scene. This application embodiment does not limit this.

[0041] The aforementioned data acquisition device collects data on roads in a road scene, and then identifies data containing vehicles from the collected data, which is the collected data.

[0042] The aforementioned thermal imaging device and the aforementioned acquisition device perform thermal imaging on a fixed location on the same road in a road scene. To achieve this thermal imaging method, the aforementioned thermal imaging device and the aforementioned acquisition device can be installed on the same side or in the same road, and the relative positions of the thermal imaging device and the acquisition device can be fixed and their directions can be aligned. In this way, the field of view information of the thermal imaging device and the acquisition device, such as the vertical pitch angle, horizontal angle, and field of view direction, remain unchanged, and they always take the fixed location on the same road as the target.

[0043] Data acquisition devices can include visible light acquisition devices, radar, etc. Visible light acquisition devices can specifically include visible light sensors, cameras, etc.; radar can include lidar, ultrasonic radar, etc.

[0044] In the presence of multiple data acquisition devices, the data obtained from these devices can be combined to determine if the acquired data includes information representing vehicles. For example, the data acquisition devices may include visible light acquisition devices, radar, etc.

[0045] In one embodiment of this application, the collected data can be obtained in the following manner:

[0046] The system acquires scene data for road scenarios; performs vehicle detection on the scene data to obtain detection results; if the detection results indicate that a vehicle has passed through the road scenario, the scene data is then identified as data collected for the vehicle that passed through the road scenario. The scene data may include any of the following: radar data or visible light images.

[0047] Different types of acquisition devices can acquire different types of scene data. For example, in addition to the radar data and visible images mentioned above, if the acquisition device also includes an infrared image acquisition device, the scene data can also include infrared images, etc. The embodiments of this application do not limit this.

[0048] Radar data consists of point cloud data collected by radar, while visible light images are acquired by visible light acquisition equipment.

[0049] The aforementioned visible light acquisition device can be a visible light sensor, etc.

[0050] In one implementation, the radar and visible light sensors can be integrated into the same device.

[0051] Specifically, radar and visible light sensors can be fixedly installed inside the housing of the vehicle temperature detection equipment. This vehicle temperature detection equipment may also include a thermal imaging module and a main control board. A detailed introduction to the aforementioned vehicle temperature detection equipment will follow later. Figure 4 , Figure 5 The corresponding implementation examples will not be detailed here.

[0052] Different methods are used for vehicle detection depending on the acquisition device. For example, for visible light acquisition devices, existing vehicle detection methods based on prior knowledge or based on stereo vision can be used.

[0053] A vehicle detection method based on radar data can be as follows: cluster the point clouds contained in the radar data to determine the point cloud data belonging to the same obstacle, and then determine whether the obstacle is a vehicle based on the shape and structural features of the point cloud formed by the point cloud data belonging to the same obstacle.

[0054] With multiple data acquisition devices, data can be collected whenever any one of them detects a vehicle passing by on the road. This allows for the use of diverse data from multiple devices for vehicle detection, providing a wealth of data for reference and thus improving detection accuracy.

[0055] The aforementioned thermal imaging device may include a thermal imaging module for performing thermal imaging to obtain a heat map containing vehicle temperature information; different colors in the heat map represent different temperatures.

[0056] Step S102: Determine the target detection location of the vehicle in the collected data.

[0057] The target detection area is the part of the vehicle whose temperature is to be detected.

[0058] After identifying the data representing the vehicle in the collected data, the vehicle structure of the passing vehicle can be determined based on this data. Based on the vehicle structure, the various parts of the vehicle can be obtained, and the target detection parts can be determined from them. For specific implementation methods, please refer to the following embodiments, which will not be detailed here.

[0059] Different vehicle parts are prone to temperature anomalies in different road scenarios. For example, accidents at tunnel entrances are often caused by abnormal temperatures in the cargo compartment, while on long downhill sections, accidents are more likely to occur due to prolonged braking leading to abnormal temperatures in the wheel hubs and brake pads. Therefore, the type of road scenario can be used to determine different target detection locations, as detailed below:

[0060] Based on the scenario type of the road scene, determine the target type of the vehicle part to be temperature detected; determine the target detection part of the target type in the collected data.

[0061] In road scenarios where collected data and heat maps are obtained, the vehicle parts that are prone to temperature anomalies are classified as vehicle part target types.

[0062] Specifically, the type corresponding to each vehicle part can be determined, such as the type of carriage corresponding to the carriage part. Furthermore, given the types of each vehicle part, the target type of vehicle part can be found from the types of each vehicle part.

[0063] As can be seen from the above, different vehicle parts may experience temperature anomalies under different road scenarios. This allows us to identify the vehicle parts experiencing temperature anomalies in a road scenario by scene type, and more directly determine the target vehicle parts whose temperature needs to be detected from among the various vehicle parts. This enables us to locate potential temperature anomalies in different scenarios more quickly and achieve high temperature warnings for diverse scenarios.

[0064] In one scenario, vehicle parts of a preset type can be detected first in the collected data. Then, based on the detection time period and start time of each detected vehicle part, the target detection part can be determined from these vehicle parts. Detailed implementation methods will be provided later. Figure 2 Steps S202-S204 in the illustrated embodiment will not be described in detail here.

[0065] Step S103: Determine the thermal region corresponding to the target detection location in the heat map.

[0066] As can be seen from the aforementioned step S101, the relative positions of the thermal imaging device that obtains the heat map and the acquisition device that obtains the collected data can be fixed. In this way, there is a transformation relationship between the coordinate systems of the heat map and the collected data. This transformation relationship represents the offset between the actual position in the road scene and the corresponding position in the heat map and the corresponding position in the collected data. Thus, the data corresponding to the same position in the heat map and the collected data can be determined through the transformation relationship. For example, the position of the target detection part in the collected data can be transformed to the position in the heat map through the transformation relationship to determine its corresponding thermal area.

[0067] The transformation relationship can be obtained as follows:

[0068] For example, a preset number of markers can be set in a road scene to obtain the position of each marker in the heat map and collected data, and then the conversion relationship can be calculated based on the above positions.

[0069] For example, four calibration objects can be set up to determine the coordinates of each calibration object in the heat map coordinate system and the data acquisition coordinate system. The matrix representing the transformation relationship can be calculated using the SVD (Singular Value Decomposition) algorithm.

[0070] Step S104: Based on the area information of the thermal zone, detect whether the temperature of the target detection part of the vehicle is abnormal.

[0071] Specifically, an abnormal threshold representing temperature anomalies can be preset. If the temperature of the thermal area exceeds the abnormal threshold, the temperature anomaly of the target detection part will be detected.

[0072] As can be seen from the above, in the solution provided by the embodiments of this application, by collecting data of the road scene, determining the vehicles passing through the road scene, and obtaining the vehicle's heat map, which contains temperature information, the temperature of the vehicle can be detected based on the heat map.

[0073] Furthermore, by collecting data, the vehicle parts within the data can be identified, from which the target detection area can be derived. The corresponding thermal region of the target detection area on the heat map can then be determined, thus obtaining the regional information for each vehicle part. Based on this regional information, temperature detection can be performed on the target detection area, thereby identifying the specific vehicle part with an abnormal temperature and improving the accuracy of vehicle temperature detection.

[0074] exist Figure 1 Based on the illustrated embodiment, when determining the target detection location, the detection time period and the start time of each vehicle part detected in the collected data can be obtained. Then, the target detection location is determined based on the obtained detection time period and the start time of the detection time period. In view of the above, this application embodiment provides a second vehicle temperature detection method.

[0075] See Figure 2 This is a flowchart illustrating the second vehicle temperature detection method provided in this application embodiment. The method includes the following steps S201-S206.

[0076] Step S201: Obtain the collected data of vehicles passing through the road scene, and obtain the heat map obtained by thermal imaging of the vehicles.

[0077] The above steps are the same as those described above. Figure 1 Step S101 is the same in the illustrated embodiment, and will not be repeated here.

[0078] Step S202: Detect vehicle parts of the preset type in the collected data.

[0079] The above preset types can be flexibly set according to the actual road scenario.

[0080] As mentioned above Figure 1 Similarly, in the illustrated embodiments, the vehicle parts prone to abnormal temperatures differ depending on the road conditions. For example, at tunnel entrances, accidents are more likely to occur due to abnormal temperatures in the cargo compartment or windows. In this case, the aforementioned preset types could be cargo compartment type and window type. Likewise, on long downhill sections, abnormal temperatures in the wheel hubs and brake pads are more common due to prolonged braking. In this case, the aforementioned preset types could be wheel hub type and brake pad type.

[0081] For specific inspection methods for vehicle components, please refer to the following sections. Figure 3A Steps S303-S304 in the illustrated embodiment will not be described in detail here.

[0082] Step S203: Obtain the detection time period for each detected vehicle part.

[0083] First, let's explain the time period in which the above detections were made.

[0084] Because vehicles are constantly moving, the data collected by the acquisition device will vary at different times. For example, if the acquisition device is a camera, the video frames captured by the camera will be different as the vehicle moves from a distance towards the camera. Specifically, the proportion of the image occupied by different parts of the vehicle in each video frame will be different.

[0085] In the solution provided in this application embodiment, the electronic device, acting as the execution subject, can continuously acquire the aforementioned collected data and continuously detect vehicle parts of a preset type in the collected data. Since the collected data differs at different times, the vehicle parts detected at different times during the continuous detection process may be different.

[0086] For example, in the data collected earlier in the detection time, the windows occupy a larger proportion of the image, while the carriage occupies a smaller proportion; in this case, only the windows can be detected. In the data collected later in the detection time, the carriage occupies a larger proportion of the image, while the windows occupy a smaller proportion; in this case, only the carriage can be detected.

[0087] Therefore, in this step, for each detected vehicle part, the time period in which that part was detected can be determined.

[0088] The detection time period can be a continuous period of time. For example, if vehicle part p1 is detected during the period from time t1 to time t3, then the period from t1 to t3 is the detection time period of vehicle part p1.

[0089] Step S204: Determine the target detection parts from the detected vehicle parts whose detection time period is longer than a preset time threshold and whose relative order of the start time of the detection time period is consistent with the actual relative arrangement order of each preset type of vehicle parts in the vehicle.

[0090] The start time of the period when the vehicle part is detected can also be called the first time the vehicle part is detected.

[0091] The aforementioned preset time threshold can be set by staff according to the actual scenario, such as the speed limit of the road where the data collection equipment is deployed, the field of view of the data collection equipment, etc., which will not be elaborated here.

[0092] The above relative order is consistent with the actual relative arrangement order, which means that the relative trend of the first detection time is consistent with the actual relative arrangement trend of each preset type of vehicle part in the vehicle.

[0093] For example, under normal circumstances, the actual relative arrangement order of vehicle parts of the window type and vehicle parts of the body type in a vehicle is: vehicle parts of the window type precede vehicle parts of the body type. If the relative order of the first detection time of window p1 and body p2 is p1 before p2, that is, the relative order of the first detection time of window p1 and body p2 is consistent with the actual relative arrangement order of vehicle parts of the window type and vehicle parts of the body type in a vehicle.

[0094] The start and end times of the detection time period for the first detected target location will be referred to as the first start time and the first end time, respectively. The start and end times of the detection time period for the second detected target location will be referred to as the second start time and the second end time, respectively. The relationship between the start and end times of the detection time periods for each target location will be introduced.

[0095] 1. The first start time is earlier than the second start time.

[0096] As can be seen from the aforementioned steps, the relative order of the start times of the time period during which the target detection parts are detected is consistent with the actual relative arrangement order of the vehicle parts of each preset type in the vehicle. Therefore, the first start time is earlier than the second start time.

[0097] 2. The first start time is earlier than the second end time.

[0098] Since the first start time is earlier than the second start time, and the second end time is after the second start time, the first start time is earlier than the second end time.

[0099] 3. The first end time may be earlier or later than the second start time mentioned above.

[0100] For example, if the first target detection area is the car window and the second target detection area is the car body, in a real-world scenario, the car body may be detected for the first time in a subsequent period after the car window has been detected. At this point, the detection period for the car window has ended, meaning the first end time is earlier than the second start time. Alternatively, the car body may be detected for the first time during the process of detecting the car window. At this point, the detection period for the car window has not yet ended, meaning the first end time is later than the second start time.

[0101] 4. The first end time may be earlier or later than the second end time mentioned above.

[0102] For example, if the first target detection area is the car window and the second target detection area is the car body, in a real-world scenario, the car body may be detected for the first time in a subsequent period after the car window has been detected. At this point, the time period for detecting the car window has ended, meaning the first end time is earlier than the second end time.

[0103] For example, if the first target detection area is the carriage and the second target detection area is the front wheel, in a real scenario, if the front wheel is detected for the first time during the detection of the carriage, the detection of the wheel may have ended before the detection of the carriage has ended, due to the long length of the vehicle and the small diameter of the front wheel. In other words, the detection time of the carriage has not ended before the detection time of the front wheel has ended, and the first end time is later than the second end time.

[0104] Step S205: Determine the thermal region corresponding to the target detection location in the heat map.

[0105] As can be seen from the aforementioned step S204, the detection time of the target detection area is longer than the preset time threshold, that is, the target detection area is detected within a certain time period.

[0106] In this step, for each target detection location, we can obtain various heat maps generated within the detection time period of that target detection location, and then determine the thermal region corresponding to the target detection location in each heat map.

[0107] Specifically, based on the location of the target detection site detected each time within the detection period, and according to the conversion relationship between the coordinate systems of the heat map and the collected data, the corresponding thermal regions of the target detection site can be determined sequentially in each heat map.

[0108] Step S206: Based on the area information of the thermal zone, detect whether the temperature of the target detection part of the vehicle is abnormal.

[0109] As can be seen from the aforementioned steps, for each target detection location, the corresponding thermal region in each thermal map can be obtained, that is, the target detection location corresponds to multiple thermal regions.

[0110] In one scenario, for each target detection location, the temperature of each thermal region corresponding to the target detection can be determined, and the number of thermal regions whose temperature is greater than the aforementioned abnormal threshold can be determined. Based on the aforementioned number, it can be determined whether the temperature of the target location is abnormal.

[0111] For example, if the ratio of the above quantity to the total number of thermal areas corresponding to the target detection is greater than a preset ratio, then the temperature of the target location is determined to be abnormal; or if the above quantity is greater than a preset quantity, then the temperature of the target location is determined to be abnormal, etc.

[0112] Since heatmaps are generated within the detection time period of the target detection area, each heatmap corresponds to a different moment within that time period. The more heatmap regions in each heatmap that represent temperatures exceeding the abnormal threshold, the longer the temperature at the target detection area has been above the abnormal threshold. Therefore, based on these numbers, a more comprehensive and accurate determination of whether the temperature of the target detection area is abnormal can be made.

[0113] As can be seen from the above, when determining the target detection location using the scheme provided in the embodiments of this application, the vehicle parts of the preset type of vehicle in the collected data are first detected. Then, the target detection location can be determined from the detected vehicle parts where the duration of the detected time period is greater than the preset duration threshold and the relative order of the first detection time is consistent with the actual relative arrangement order of the vehicle parts of each preset type in the vehicle.

[0114] It can be seen that, on the one hand, the detection time of the target detection area is longer than the preset time, meaning that the target detection area is detected within a longer period of time. This reduces the possibility of misidentifying a vehicle part that was falsely detected at a certain moment as the target detection area due to noise within the number of samples, thus improving the accuracy of the identified target detection areas. On the other hand, the relative order of the first detection time of each target detection area is consistent with the actual relative arrangement order of each preset type of vehicle part in the vehicle. This can eliminate vehicle parts whose relative order of the first detection time does not match the actual relative arrangement order of the vehicle parts in the vehicle, making the identified target detection areas more consistent with the actual scene, further reducing false detections, and thus improving the accuracy of the identified target detection areas.

[0115] The following combination Figure 3AThe flowchart illustrating the vehicle temperature detection method explains how the target detection area is determined in step S102. Steps S302-S304 detail the specific implementation of determining the target detection area.

[0116] Step S301: Obtain the collected data of vehicles passing through the road scene, and obtain the heat map obtained by thermal imaging of the vehicles.

[0117] The above step S301 is the same as the aforementioned step S101, and will not be described in detail here.

[0118] Step S302: Extract the data features of the collected data.

[0119] Data features can be features representing vehicle data in the collected data; for example, if the collected data is a visible light image, the data features can be features representing texture and color; if the collected data is radar data, the data features can be geometric features such as straight lines, arcs, and corners.

[0120] Specifically, a pre-trained neural network model can be used for feature extraction. The aforementioned network model is trained using sample data of the same type as the collected data. The neural network model used can be a convolutional neural network model or a ViT (Vision Transformer) model, etc. This application embodiment does not limit this.

[0121] Step S303: Obtain vehicle structure information based on data characteristics.

[0122] Different vehicle parts have different attributes, shapes, and space occupies, meaning their attribute characteristics differ. In this case, by judging the similarity between the data characteristics and the preset attribute characteristics of corresponding vehicle parts, the vehicle parts that match the data characteristics can be identified. Then, based on the location of the data corresponding to the data characteristics in the collected data, the vehicle structure to which it belongs can be determined. Therefore, vehicle structure information can include the aforementioned identified vehicle parts and their locations.

[0123] In one implementation, the vehicle structure information may include: window information, vehicle outline information, and axle information.

[0124] in,

[0125] Window information can include the window's location or shape.

[0126] Vehicle outline information can include the shape of the vehicle outline, the area it occupies in the data collection, and other information.

[0127] Axle information can include information such as hub position, tire area, hub type, and number of hubs.

[0128] Step S304: Determine the target detection area of ​​the vehicle based on the vehicle structure information.

[0129] As described in step S303 above, the vehicle structure information includes information about each vehicle component. In this case, the vehicle component to be detected can be determined as the target detection component based on the scenario type or a manually specified method. The target detection component can be the driver's cab, engine, wheels, cargo box, etc.

[0130] Because the relative positions of various parts of a vehicle are relatively fixed—for example, the engine is always located at the front of the vehicle, and therefore closer to the front than the windows—the relative positional relationships between the various vehicle parts can be pre-recorded. Thus, given any given vehicle part, the target detection part with a relative positional relationship to it can be determined based on the position of that part.

[0131] For specific detection methods, please refer to the following embodiments, which will not be detailed here.

[0132] As can be seen from the above, the target detection area can be determined based on the vehicle structure information. In this way, when measuring the temperature, the temperature of each detected part can be measured, and when issuing an early warning, the vehicle part with abnormal temperature can be identified, thus achieving precise temperature measurement.

[0133] Step S305: Determine the thermal region corresponding to the target detection location in the heat map.

[0134] Step S306: Based on the area information of the thermal zone, detect whether the temperature of the target detection part of the vehicle is abnormal.

[0135] The steps S305-S306 described above are the same as the steps S103-S104 described above, and will not be described in detail here.

[0136] The following explains the specific implementation method of determining the target detection part of the vehicle based on the vehicle structure information in step S304.

[0137] In the first implementation method, the driver's cab area of ​​the vehicle can be determined based on the information from the vehicle windows, and the driver's cab area can be used as the target detection part of the vehicle.

[0138] In one embodiment of this application, when the data acquisition device is fixed, since vehicles always travel along the road in the road scene, the vehicles represented by the acquired data always face the acquisition device at the same or similar angle. The relative relationships between the positions of various vehicle parts presented in the acquired data are also relatively fixed. Thus, the driver's cab area can be obtained by displacement based on the window position recorded in the window information, using the fixed relative relationships as the starting point.

[0139] Taking the acquisition of vehicle images using a visible light acquisition device as an example, if the device is installed above the road in the middle of the road, then in all captured vehicle images, the angle at which the vehicle faces the acquisition device is fixed, and all vehicle images are top-down views. Figure 3B The image shown is of a vehicle captured by a data acquisition device. Figure 3B The vehicle image shows the front of the car facing down; however, considering that roads are usually two-way, there may also be situations where the front of the car faces up in the overhead view. This situation is consistent with... Figure 3B The processing method is the same; here we will only use... Figure 3B Please provide an explanation.

[0140] exist Figure 3B As can be seen, the windshield window located at the front of the vehicle, hereinafter referred to as the first window, is positioned slightly below the driver's cab area in the image. Regardless of the vehicle's model and size, in this road scene, captured by the same fixed acquisition device, the relative relationship between the area of ​​the first window and the driver's cab area in the image remains relatively fixed for any vehicle being photographed; that is, the area of ​​the first window is always slightly below.

[0141] In this way, the distance from the area containing the first window to the area containing the driver's cab can be calculated. Specifically, this can be done by calculating the distance between the center points of the aforementioned areas. Since the data acquisition device can continuously capture data, multiple sets of statistical values ​​for the distance from the area containing the first window to the area containing the driver's cab can be obtained from the collected data.

[0142] Furthermore, since the relative positions of the vehicle parts are fixed under this visible light acquisition device, the offset angle of the cab area relative to the first window can be determined in advance.

[0143] As mentioned above, the window information can include the window position. The distance corresponding to the statistical value of the displacement obtained from the offset angle of the window position can then be used to determine the driver's cab area.

[0144] The first window mentioned above is merely an example. If the data acquisition device has other installation angles, other window positions can be selected, and the distance from the selected window area to the driver's cab area can be calculated using the method described above to determine the location of the driver's cab area. This application does not limit the scope of the embodiments.

[0145] In the second implementation method, the license plate location of the vehicle can be obtained based on data features. Based on the window information, vehicle outline information and license plate location, the engine area of ​​the vehicle can be determined and the engine area can be used as the target detection part.

[0146] Similar to the first implementation, with the data acquisition equipment fixed, the positions of the car window and the engine area (where the vehicle engine is located), as well as the position of the license plate and the engine area, are relatively fixed in the car window information. Therefore, following the implementation process in the first method, statistical values ​​can be obtained by statistically analyzing the relative distances between the above positions using multiple data acquisitions. Based on these statistical values, the engine area can be obtained by shifting the positions of the car window and license plate.

[0147] Furthermore, the obtained engine area can be compared with the coverage area of ​​the vehicle outline in the vehicle outline information. If the engine area exceeds the coverage area, the engine area can be adjusted to be within the coverage area.

[0148] In the third implementation method, the tire area of ​​the vehicle is determined based on the hub position in the axle information, and the tire area is used as the target detection part.

[0149] Since the tire is connected to the rim, the area within a preset distance of the rim location is the tire area.

[0150] In the fourth implementation method, the vehicle compartment area is determined based on the hub position, hub type and number of hubs in the axle information, and the compartment area is used as the target detection part.

[0151] Combining the third and first implementation methods, the hub position, hub type, and number of hubs can be used to determine the overall tire area of ​​all tires. By statistically analyzing the relative positions of the overall tire area and the cargo box area, the cargo box area can be obtained by displacement based on the overall tire area determined by the axle information.

[0152] As can be seen from the above, when the vehicle structure is relatively fixed, the relative positions between various vehicle parts can be obtained based on the vehicle structure information representing the vehicle structure, and the target detection parts can be determined accordingly. This allows for detection even when some vehicle parts are not obvious from the outside, thus improving the applicability of this solution.

[0153] The vehicle structure information in step S304 above can include target information related to the road scene. Similar to step S102, different vehicle parts are prone to temperature anomalies, and correspondingly, the scene type of the road scene can be associated with the vehicle parts that need to be detected. In this case, target information for determining the target type can be selected from the vehicle structure information, where the target type is determined by the type of the road scene; based on the target information, the target detection part of the vehicle is determined.

[0154] As mentioned in step S102 above, different vehicle parts are prone to temperature anomalies in different road scenarios. Correspondingly, the target type of vehicle parts prone to temperature anomalies corresponds to the type of road scenario.

[0155] Target information can be descriptive information about vehicle parts, such as recording the name of the vehicle part and the target type corresponding to the vehicle part. As can be seen from the four implementation methods for determining the target detection parts of a vehicle in the above embodiments, the target detection parts can be obtained from partial information in the vehicle structure information, and there is a relative relationship between the target detection parts and the positions represented by the aforementioned partial information. Therefore, based on this relative relationship, a correspondence between target types and target information can be pre-established. For example, the cab type corresponding to the cab area corresponds to the window information, and the tire type corresponding to the tire area corresponds to the axle information, etc.

[0156] Thus, given the target information, the location of the target type that can be determined using the target information can be obtained based on the above correspondence.

[0157] In this way, when the vehicle structure information contains target information, the vehicle parts corresponding to the scene type can be directly determined as target detection parts based on the target information, which improves the efficiency of determining target detection parts.

[0158] In the aforementioned step S103, once the transformation relationship is obtained, it can be adjusted in the following manner.

[0159] Specifically, the time difference between the heat map acquisition time and the data acquisition time can be obtained; based on the time difference, the first transformation relationship between the coordinate system corresponding to the heat map and the coordinate system corresponding to the data acquisition is adjusted; based on the adjusted first transformation relationship, the thermal region corresponding to the target detection location is determined in the heat map.

[0160] Specifically, devices such as the main control board that collect data and heat maps can record the collection time when they collect the data and heat maps.

[0161] The first conversion relationship mentioned above is the conversion relationship obtained in step S103. Within the time difference between the heat map acquisition time and the data acquisition time, the vehicle speed is obtained, and the vehicle's translation in the road scene is calculated based on the acquired speed and the time difference. In this way, the translation can be used to adjust the first conversion relationship.

[0162] The method for determining the thermal region based on the adjusted first transformation relationship is the same as step S103, and will not be described in detail here.

[0163] As can be seen above, before the adjustment, if there is a time difference between the collection time of the data and the heat map, the first conversion relationship does not represent the relative position of the vehicle at the same location in the data and the heat map due to the displacement of the vehicle, resulting in a deviation. After the adjustment, the first conversion relationship represents the same location, reducing the original deviation and improving the accuracy of the obtained conversion relationship.

[0164] In one embodiment of this application, when the scene data includes radar data and visible light images, vehicle detection is performed on the scene data, including:

[0165] Based on the second transformation relationship between the radar coordinate system corresponding to the radar data and the image coordinate system corresponding to the visible light image, data fusion is performed on the radar data and the visible light image; vehicle detection is then performed based on the fused data.

[0166] The radar coordinate system and the visible light coordinate system can be determined based on the field of view information of the respective equipment that acquires radar data and visible light images.

[0167] The second conversion relationship can be achieved using existing automatic calibration algorithms, such as the TFAC (Target-Free Automatic Calibration)-Livox algorithm; or, it can be obtained by setting up calibration objects in the scene contained in the radar data and visible light image using the method for obtaining the first conversion relationship.

[0168] Data fusion refers to determining the radar data and visible light image regions corresponding to the same location through a second transformation relationship; it is essentially a data alignment process. In this way, the fused data includes both image information from the visible light image and depth information from the radar data, resulting in a more accurate description of the vehicle's position.

[0169] In this case, vehicle detection can be performed by separately detecting the image information and depth information in the fused data, and then jointly determining whether a vehicle has passed at any location in the road scene based on the results of the two detections.

[0170] As can be seen above, using fused data to detect vehicles at any location, and confirming the detection results through multiple pieces of information, ensures the accuracy of the detection results. The fusion of multi-target radar data with visible light and thermal imaging data enables real-time recording of target temperature, while also providing clear images and thermal maps as early warning information.

[0171] The following describes the automatic calibration algorithm for obtaining the second transformation relation; see [link to documentation]. Figure 4 , Figure 4A flowchart of a method for obtaining transformation relationships is provided, which includes the following steps S401-S407.

[0172] Step S401: Obtain the first calibration radar data for the road scene.

[0173] The method for obtaining the first calibration radar data is the same as that for obtaining radar data in step S101. Alternatively, a preset static object can be used as the calibration object, and the radar data of that calibration object can be obtained as the first calibration radar data.

[0174] Step S402: Determine the first driving trajectory of the vehicle in the road scene based on the first calibration radar data.

[0175] The vehicle's position at different acquisition times can be obtained using the first calibration radar data. Based on the displacement between positions at different times, the vehicle's driving trajectory can be determined and used as the first driving trajectory.

[0176] Step S403: Obtain the angle between the first driving trajectory and the road in the road scene.

[0177] The direction of the road can be represented by the direction of the lane lines in the middle and on the side of the road. In this case, the angle between the tangent direction of the first driving trajectory and the direction of the lane lines can be calculated.

[0178] Step S404: Correct the radar coordinate system corresponding to the included angle radar data.

[0179] Specifically, the radar coordinate system can be adjusted by rotation, translation, etc., until the aforementioned included angle changes and is less than the preset included angle threshold, and the adjusted coordinate system is determined as the new coordinate system.

[0180] Step S405: Continuously acquire calibration data sets for road scenarios.

[0181] The calibration data set includes: synchronously acquired second calibration radar data and calibration visible light images.

[0182] Calibration Data Set: The second calibration radar data is similar to the first calibration radar data, differing only in the acquisition time. The acquisition method for the calibration visible light image is the same as that for the visible light image in the aforementioned embodiments, and will not be described in detail here. The calibration object for the second calibration radar data and the calibration visible light image is similar to that for the first calibration radar data; it can be a vehicle or a preset static calibration object.

[0183] Step S406: For each set of calibration data obtained, vehicle detection is performed on the second calibration radar data and the calibration visible light image respectively to obtain radar vehicle detection results and visible light vehicle detection results. Based on the radar vehicle detection results and visible light vehicle detection results, the coordinate transformation relationship between the radar coordinate system and the image coordinate system is determined.

[0184] The vehicle detection method is the same as the aforementioned embodiment. The difference is that when it is determined that the same vehicle or the same static calibration object exists, the calibration position in the two types of data detected corresponds to the same actual position. In this way, the conversion relationship between the calibration positions is used as the above coordinate conversion relationship.

[0185] Step S407: If the difference between the coordinate transformation relationships corresponding to the two adjacent calibration data sets is less than the preset error, then the second transformation relationship is obtained based on the coordinate transformation relationships corresponding to the two adjacent calibration data sets.

[0186] The differences between coordinate transformation relationships can be obtained by calculating the differences between the displacements represented by each transformation relationship. In this case, the preset error is a pre-set difference threshold.

[0187] If the difference between coordinate transformation relationships is less than the preset error, then the transformation relationship corresponding to one of the two adjacent calibration data sets can be selected, or the transformation relationship corresponding to the two adjacent calibration data sets can be selected if they are similar. The selected transformation relationship is used as the second transformation relationship. This application embodiment does not limit this.

[0188] In addition, if the difference between the coordinate transformation relationships corresponding to two adjacent sets of calibration data is not less than the preset error, a new set of calibration data can be selected again to achieve multiple iterations.

[0189] As can be seen above, a transformation relationship with smaller errors was determined by multiple calibration data sets. The resulting matrix is ​​applicable to calibration data sets collected multiple times, which means that the transformation can be performed more accurately in multiple collected data sets, thereby improving the accuracy of obtaining the transformation relationship.

[0190] The following explains the warning method when an abnormal temperature is detected.

[0191] In one embodiment of this application, after detecting an abnormal temperature at a target detection area of ​​a vehicle, the vehicle's license plate number can be obtained; based on the license plate number, an early warning can be issued for the target detection area; based on the collected data, the vehicle's driving information can be monitored; based on the driving information, it can be determined whether the vehicle has responded to the early warning; if not, an upgraded early warning can be issued for the target detection area based on the license plate number.

[0192] Upon receiving a warning message, if the driver responds, the vehicle's driving status will change, such as in terms of direction and speed. This driving information may include the vehicle's trajectory calculated from continuously acquired data after the warning. Based on this trajectory, it can be determined whether the vehicle's driving status has changed, and a warning message can be sent to the target detection area. The warning message may include the license plate number, allowing drivers of vehicles experiencing abnormal temperatures to confirm that their vehicle is the problem. The warning message can be sent via directional speakers and screens, or it can be displayed using warning lights. This application does not limit the scope of this application.

[0193] During the above process, multi-target radar can be used to collect data on vehicles in real time, around the clock, and update the vehicle's driving trajectory.

[0194] Upgrading early warning measures could involve sending more prominent warning messages. For example, increasing the volume of directional sound systems or brightening hazard lights.

[0195] As can be seen from the above, by issuing warnings and escalating them if no response is received, drivers can receive warning information and take appropriate measures as much as possible, thus reducing the occurrence of dangerous situations.

[0196] The following explains how to determine whether a vehicle should issue a warning response.

[0197] In one implementation, if the driving information includes a second driving trajectory, it is determined whether the second driving trajectory deviates from the original driving trajectory and heads towards a preset abnormal handling location. If so, it is determined that the vehicle has issued a warning response.

[0198] The original driving trajectory can be considered as the vehicle's driving trajectory before the warning was issued. The system determines whether the direction of the original driving trajectory and the driving trajectory after the warning information was sent has changed. If the degree of change exceeds a preset threshold, a deviation is confirmed.

[0199] The anomaly handling location can have a preset location. Therefore, by obtaining the relative position of the vehicle and the anomaly handling location at each time based on the vehicle trajectory, it can be determined whether the vehicle is heading towards the anomaly handling location.

[0200] In another implementation, if the driving information includes driving speed, it is determined whether the vehicle is decelerating based on the driving speed. If so, it is determined that the vehicle has issued a warning response.

[0201] Specifically, the calculation can determine the speed difference between the vehicle's speed before and after the warning message is sent, and whether the vehicle has slowed down.

[0202] In addition, after confirming deceleration, a speed threshold can be preset, and the vehicle is deemed to have issued a warning response when the driving speed drops below the speed threshold.

[0203] As can be seen from the above, by detecting changes in the vehicle's driving status, it is possible to determine whether a warning has been responded to, and thus decide whether to escalate the warning.

[0204] Corresponding to the above-described vehicle temperature detection method, one embodiment of this application also provides a vehicle temperature detection device.

[0205] See Figure 5 In one embodiment of this application, a structural schematic diagram of a first vehicle temperature detection device is provided, the device comprising:

[0206] Collector 501 is used to acquire data on passing vehicles in a road scene and send the acquired data to the main control board.

[0207] The thermal imaging module 502 is used to obtain a thermal map of the vehicle by performing thermal imaging; and to send the thermal map to the main control board.

[0208] The main control board 503 is used to determine the target detection part of the vehicle in the collected data; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region.

[0209] In one embodiment of this application, the main control board 503 is specifically used to detect vehicle parts of a preset type in the collected data; obtain the detection time period of each detected vehicle part; and determine the target detection parts from the detected vehicle parts whose detection time period is longer than a preset time period threshold and whose relative order of the start time of the detection time period is consistent with the actual relative arrangement order of each preset type of vehicle part in the vehicle.

[0210] As can be seen from the above, when determining the target detection location using the scheme provided in the embodiments of this application, the vehicle parts of the preset type of vehicle in the collected data are first detected. Then, the target detection location can be determined from the detected vehicle parts where the duration of the detected time period is greater than the preset duration threshold and the relative order of the start time of the detected time period is consistent with the actual relative arrangement order of the vehicle parts of each preset type in the vehicle.

[0211] It can be seen that, on the one hand, the detection time of the target detection area is longer than the preset time, meaning that the target detection area is detected within a longer period of time. This reduces the possibility of misidentifying a vehicle part that was falsely detected at a certain moment as the target detection area due to noise within the number of samples, thus improving the accuracy of the identified target detection areas. On the other hand, the relative order of the first detection time of each target detection area is consistent with the actual relative arrangement order of each preset type of vehicle part in the vehicle. This can eliminate vehicle parts whose relative order of the first detection time does not match the actual relative arrangement order of the vehicle parts in the vehicle, making the identified target detection areas more consistent with the actual scene, further reducing false detections, and thus improving the accuracy of the identified target detection areas.

[0212] In one embodiment of this application, the main control board 503 is specifically used to extract data features from the collected data; obtain vehicle structure information of the vehicle based on the data features, wherein the vehicle structure information includes at least one of the following: window information, vehicle outline information, and axle information; determine the target detection part of the vehicle based on the vehicle structure information; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region.

[0213] The data acquisition unit 501 and the thermal imaging module 502 can be connected to the main control board 503 through a preset physical interface. Furthermore, the data acquisition unit 501 and the thermal imaging module 502 can be packaged with the main control board 503 in the same housing to form an integrated unit, and the positions of the data acquisition unit 501, the thermal imaging module 502, and the main control board 503 can be fixed during packaging.

[0214] In the solution provided in this application embodiment, by collecting data of the road scene, it is determined that vehicles pass through the road scene and a heat map of the vehicles is obtained. The heat map contains temperature information, so the temperature of the vehicles can be detected based on the heat map.

[0215] In one embodiment of this application, the main control board 503 is specifically used to extract data features from the collected data; obtain vehicle structure information of the vehicle based on the data features, wherein the vehicle structure information includes at least one of the following: window information, vehicle outline information, and axle information; determine the driver's cab area of ​​the vehicle based on the window information, and use the driver's cab area as the target detection part of the vehicle; and / or obtain the license plate position of the vehicle based on the data features, determine the engine area of ​​the vehicle based on the window information, vehicle outline information, and license plate position, and use the engine area as the target detection part; and / or determine the tire area of ​​the vehicle based on the hub position in the axle information, and use the tire area as the target detection part; and / or determine the passenger compartment area of ​​the vehicle based on the hub position, hub type, and number of hubs in the axle information, and use the passenger compartment area as the target detection part; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the area information of the thermal region.

[0216] Based on the vehicle's structural information, the target detection area can be determined. In this way, when measuring the temperature, the temperature of each detected area can be measured, and when issuing an early warning, the vehicle part with abnormal temperature can be identified, thus achieving precise temperature measurement.

[0217] In one embodiment of this application, the main control board 503 is specifically used to extract data features from the collected data; obtain vehicle structure information of the vehicle based on the data features, wherein the vehicle structure information includes at least one of the following: window information, vehicle outline information, and axle information; select target information from the vehicle structure information for determining the target type, wherein the target type is determined based on the type of the road scene; determine the target detection part of the vehicle based on the target information; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region.

[0218] With a relatively fixed vehicle structure, the relative positions of each vehicle part can be obtained based on the vehicle structure information representing the vehicle structure, and the target detection part can be determined accordingly. This allows for detection even when some vehicle parts are not obvious from the outside, thus improving the applicability of this solution.

[0219] In one embodiment of this application, the main control board 503 is specifically used to determine the target type of the vehicle part to be temperature detected based on the scene type of the road scene; determine the target detection part of the target type in the collected data; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region.

[0220] In this way, when the vehicle structure information contains target information, the vehicle parts corresponding to the scene type can be directly determined as target detection parts based on the target information, which improves the efficiency of determining target detection parts.

[0221] In one embodiment of this application, the main control board 503 is specifically used to determine the target detection part of the vehicle in the collected data; obtain the time difference between the acquisition time of the heat map and the acquisition time of the collected data; adjust the first transformation relationship between the coordinate system corresponding to the heat map and the coordinate system corresponding to the collected data according to the time difference; determine the thermal region corresponding to the target detection part in the heat map based on the adjusted first transformation relationship; and detect whether the temperature of the target detection part of the vehicle is abnormal according to the regional information of the thermal region.

[0222] Before the adjustment, if there was a time difference between the data collection time and the heat map collection time, the first conversion relationship would not represent the relative position of the vehicle at the same location in the data collection data and the heat map due to the vehicle's displacement, resulting in a deviation. After the adjustment, the first conversion relationship represents the same location, improving the accuracy of the obtained conversion relationship.

[0223] See Figure 6 The diagram shown is a structural schematic of the second type of vehicle temperature detection device. Figure 6 Another main control board access method is provided, wherein the data acquisition device includes: radar and / or visible light sensor, wherein the radar and the visible light sensor are fixedly installed in the housing of the vehicle temperature detection device; Figure 6 Take the example of radar and visible light sensors coexisting.

[0224] The radar 601 is specifically used to acquire radar data collected for road scenes and to send the radar data to the main control board.

[0225] The visible light sensor is specifically used to acquire visible light images of a road scene and send the visible light images to the main control board.

[0226] The main control board 603 is specifically used to perform vehicle detection on scene data including radar data and / or visible light images to obtain detection results; if the detection results indicate that a vehicle has passed through the road scene, then the scene data is determined as the collection data for the vehicle passing through the road scene; the target detection part of the vehicle in the collection data is determined; the thermal region corresponding to the target detection part is determined in the heat map; and the temperature of the target detection part of the vehicle is detected as abnormal based on the regional information of the thermal region.

[0227] In this embodiment, the radar 601 and the thermal imaging module 602 can still be connected to the main control board through a physical interface, while the visible light sensor can be built into the device in which the main control board is installed.

[0228] This allows for the use of various data acquisition devices to collect data for vehicle detection, providing a wealth of data for reference during the detection process and thus improving the accuracy of the detection.

[0229] In one embodiment of this application, when the scene data includes radar data and visible light images...

[0230] The main control board is specifically used to perform data fusion on the radar data and the visible light image based on a second transformation relationship between the radar coordinate system corresponding to the radar data and the image coordinate system corresponding to the visible light image; perform vehicle detection based on the fused data to obtain a detection result; if the detection result indicates that a vehicle has passed through the road scene, then the scene data is determined as the collection data for the vehicle passing through the road scene; determine the target detection part of the vehicle in the collection data; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region.

[0231] Vehicles at any location are detected using fused data, and the detection results are confirmed through multiple pieces of information, thus ensuring the accuracy of the detection results. The fusion of multi-target radar data with visible light and thermal imaging data enables real-time recording of target temperature, while also providing clear images and thermal maps as early warning information.

[0232] In one embodiment of this application, the second conversion relationship is obtained in the following manner:

[0233] Obtain first calibration radar data for the road scenario;

[0234] The first driving trajectory of the vehicle in the road scene is determined based on the first calibration radar data;

[0235] Obtain the angle between the first driving trajectory and the road in the road scene;

[0236] The radar coordinate system corresponding to the radar data is corrected based on the included angle.

[0237] Continuously acquire calibration data sets for the road scene, wherein the calibration data sets include: synchronously acquired second calibration radar data and calibration visible light images;

[0238] For each set of calibration data obtained, vehicle detection is performed on the second calibration radar data and the calibration visible light image respectively to obtain radar vehicle detection results and visible light vehicle detection results. Based on the radar vehicle detection results and visible light vehicle detection results, the coordinate transformation relationship between the radar coordinate system and the image coordinate system is determined.

[0239] If the difference between the coordinate transformation relationships corresponding to two adjacent sets of calibration data is less than a preset error, then the second transformation relationship is obtained based on the coordinate transformation relationships corresponding to the two adjacent sets of calibration data.

[0240] By using multiple sets of calibration data, a transformation relationship with smaller errors was determined. The resulting matrix is ​​applicable to multiple sets of calibration data, meaning that the transformation can be performed more accurately across multiple sets of collected data, thereby improving the accuracy of the transformation relationship.

[0241] In one embodiment of this application, the main control board 503 is specifically used to determine the target detection part of the vehicle in the collected data; determine the thermal region corresponding to the target detection part in the heat map; detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region; after detecting the abnormal temperature of the target detection part of the vehicle, obtain the license plate number of the vehicle; issue an early warning for the target detection part based on the license plate number; monitor the driving information of the vehicle based on the collected data; determine whether the vehicle has issued an early warning response based on the driving information; if not, issue an upgraded early warning for the target detection part based on the license plate number.

[0242] By issuing warnings and escalating them if no response is received, drivers can receive warning information and take appropriate measures as much as possible, thus reducing the occurrence of dangerous situations.

[0243] In one embodiment of this application, the main control board 503 is specifically used to determine the target detection part of the vehicle in the collected data; determine the thermal region corresponding to the target detection part in the heat map; detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region; after detecting the abnormal temperature of the target detection part of the vehicle, obtain the license plate number of the vehicle; issue an early warning for the target detection part based on the license plate number; monitor the driving information of the vehicle based on the collected data; if the driving information includes a second driving trajectory, determine whether the second driving trajectory deviates from the original driving trajectory and heads towards a preset abnormal handling location; if so, determine that the vehicle has issued an early warning response; and / or, if the driving information includes driving speed, determine whether the vehicle is decelerating based on the driving speed; if so, determine that the vehicle has issued an early warning response; if not, issue an upgraded early warning for the target detection part based on the license plate number.

[0244] By detecting changes in the vehicle's driving status, it is possible to determine whether a warning has been responded to, and thus decide whether to escalate the warning.

[0245] Corresponding to the above-described vehicle temperature detection method, this application also provides a vehicle temperature detection device, which will be described below.

[0246] In one embodiment of this application, see Figure 7 A schematic diagram of a vehicle temperature detection device is provided, the device comprising:

[0247] The data acquisition module 701 is used to acquire collected data of vehicles passing through a road scene and to obtain a thermal map obtained by thermal imaging of the vehicles.

[0248] The detection location determination module 702 is used to determine the target detection location of the vehicle in the collected data;

[0249] The thermal region determination module 703 is used to determine the thermal region corresponding to the target detection location in the thermal map;

[0250] The temperature anomaly detection module 704 is used to detect whether the temperature of the target detection part of the vehicle is abnormal based on the area information of the thermal zone.

[0251] In the solution provided in this application embodiment, by collecting data of the road scene, it is determined that vehicles pass through the road scene and a heat map of the vehicles is obtained. The heat map contains temperature information, so the temperature of the vehicles can be detected based on the heat map.

[0252] Furthermore, by collecting data, the vehicle parts within the data can be identified, from which the target detection area can be derived. The corresponding thermal region of the target detection area on the heat map can then be determined, thus obtaining the regional information for each vehicle part. Based on this regional information, temperature detection can be performed on the target detection area, thereby identifying the specific vehicle part with an abnormal temperature and improving the accuracy of vehicle temperature detection.

[0253] In one embodiment of this application, the detection location determination module includes:

[0254] A data feature extraction unit is used to extract data features from the collected data;

[0255] The structural information acquisition unit is used to obtain the vehicle structural information of the vehicle based on the data features, wherein the vehicle structural information includes at least one of the following: window information, vehicle outline information, and axle information.

[0256] The detection location determination unit is used to determine the target detection location of the vehicle based on the vehicle structure information.

[0257] Based on the vehicle's structural information, the target detection area can be determined. In this way, when measuring the temperature, the temperature of each detected area can be measured, and when issuing an early warning, the vehicle part with abnormal temperature can be identified, thus achieving precise temperature measurement.

[0258] In one embodiment of this application, the detection location determination unit is specifically configured to: determine the driver's cab area of ​​the vehicle based on the window information, and use the driver's cab area as the target detection location of the vehicle; and / or obtain the license plate position of the vehicle based on the data features, and determine the engine area of ​​the vehicle based on the window information, vehicle outline information, and the license plate position, and use the engine area as the target detection location; and / or determine the tire area of ​​the vehicle based on the hub position in the axle information, and use the tire area as the target detection location; and / or determine the passenger compartment area of ​​the vehicle based on the hub position, hub type, and number of hubs in the axle information, and use the passenger compartment area as the target detection location.

[0259] With a relatively fixed vehicle structure, the relative positions of each vehicle part can be obtained based on the vehicle structure information representing the vehicle structure, and the target detection part can be determined accordingly. This allows for detection even when some vehicle parts are not obvious from the outside, thus improving the applicability of this solution.

[0260] In one embodiment of this application, the detection location determination unit is specifically used to select target information for determining the target type from the vehicle structure information, wherein the target type is determined based on the type of the road scene; and to determine the target detection location of the vehicle based on the target information.

[0261] In this way, when the vehicle structure information contains target information, the vehicle parts corresponding to the scene type can be directly determined as target detection parts based on the target information, which improves the efficiency of determining target detection parts.

[0262] In one embodiment of this application, the detection location determination module is specifically used to determine the target type of the vehicle part to be temperature detected based on the scene type of the road scene; and to determine the target detection location of the target type in the collected data.

[0263] Before the adjustment, if there was a time difference between the data collection time and the heat map collection time, the first conversion relationship would not represent the relative position of the vehicle at the same location in the data collection data and the heat map due to the vehicle's displacement, resulting in a deviation. After the adjustment, the first conversion relationship represents the same location, improving the accuracy of the obtained conversion relationship.

[0264] In one embodiment of this application, the thermal region determination module is specifically used to obtain the time difference between the acquisition time of the heat map and the acquisition time of the acquired data; adjust the first transformation relationship between the coordinate system corresponding to the heat map and the coordinate system corresponding to the acquired data based on the time difference; and determine the thermal region corresponding to the target detection part in the heat map based on the adjusted first transformation relationship.

[0265] This allows for the use of various data acquisition devices to collect data for vehicle detection, providing a wealth of data for reference during the detection process and thus improving the accuracy of the detection.

[0266] In one embodiment of this application, the data acquisition module is specifically used to acquire scene data, including radar data and / or visible light images, collected for a road scene; perform vehicle detection on the scene data to obtain detection results; if the detection results indicate that a vehicle has passed through the road scene, then the scene data is determined as the data collected for the vehicle passing through the road scene.

[0267] Vehicles at any location are detected using fused data, and the detection results are confirmed through multiple pieces of information, thus ensuring the accuracy of the detection results. The fusion of multi-target radar data with visible light and thermal imaging data enables real-time recording of target temperature, while also providing clear images and thermal maps as early warning information.

[0268] In one embodiment of this application, when the scene data includes radar data and a visible light image, the data acquisition module is specifically used to acquire scene data for a road scene, including radar data and / or a visible light image; based on a second transformation relationship between the radar coordinate system corresponding to the radar data and the image coordinate system corresponding to the visible light image, the radar data and the visible light image are fused; vehicle detection is performed based on the fused data to obtain a detection result; if the detection result indicates that a vehicle has passed through the road scene, then the scene data is determined as the data collected for the vehicle passing through the road scene.

[0269] By using multiple sets of calibration data, a transformation relationship with smaller errors was determined. The resulting matrix is ​​applicable to multiple sets of calibration data, meaning that the transformation can be performed more accurately across multiple sets of collected data, thereby improving the accuracy of the transformation relationship.

[0270] In one embodiment of this application, the second conversion relationship is obtained in the following manner:

[0271] The system acquires first calibration radar data for the road scene; determines a first driving trajectory of a vehicle in the road scene based on the first calibration radar data; obtains the angle between the first driving trajectory and the road in the road scene; corrects the radar coordinate system corresponding to the radar data based on the angle; continuously acquires calibration data sets for the road scene, wherein the calibration data sets include: synchronously acquired second calibration radar data and calibration visible light images; for each acquired calibration data set, vehicle detection is performed on the second calibration radar data and the calibration visible light image respectively to obtain radar vehicle detection results and visible light vehicle detection results; based on the radar vehicle detection results and visible light vehicle detection results, a coordinate transformation relationship between the radar coordinate system and the image coordinates is determined; if the difference between the coordinate transformation relationships corresponding to two adjacent calibration data sets is less than a preset error, the second transformation relationship is obtained based on the coordinate transformation relationships corresponding to the two adjacent calibration data sets.

[0272] In one embodiment of this application, the apparatus further includes:

[0273] A license plate number acquisition module is used to obtain the license plate number of the vehicle.

[0274] The early warning module is used to issue an early warning based on the license plate number and the target detection location.

[0275] A driving information monitoring module is used to monitor the driving information of the vehicle based on the collected data;

[0276] The warning response judgment module is used to determine whether the vehicle has issued a warning response based on the driving information.

[0277] An upgraded early warning module is used to issue an upgraded early warning for the target detection area based on the license plate number if the condition is not met.

[0278] By issuing warnings and escalating them if no response is received, drivers can receive warning information and take appropriate measures as much as possible, thus reducing the occurrence of dangerous situations.

[0279] In one embodiment of this application, the warning response judgment module is specifically used to determine whether the second driving trajectory deviates from the original driving trajectory and heads towards a preset abnormal handling location when the driving information includes a second driving trajectory; if so, the vehicle is determined to have issued a warning response; and / or, when the driving information includes a driving speed, to determine whether the vehicle is decelerating based on the driving speed; if so, the vehicle is determined to have issued a warning response.

[0280] By detecting changes in the vehicle's driving status, it is possible to determine whether a warning has been responded to, and thus decide whether to escalate the warning.

[0281] This application also provides an electronic device, such as... Figure 8 As shown, it includes:

[0282] Memory 801 is used to store computer programs;

[0283] When the processor 802 executes the program stored in the memory 801, it implements the vehicle temperature detection steps described in any of the above embodiments:

[0284] Furthermore, the aforementioned electronic device may also include a communication bus and / or a communication interface, with the processor 802, the communication interface, and the memory 801 communicating with each other via the communication bus.

[0285] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0286] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0287] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0288] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0289] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of any of the above-described vehicle temperature detection methods.

[0290] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the vehicle temperature detection methods described above.

[0291] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or other media (e.g., solid state disk (SSD)).

[0292] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0293] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for devices, apparatuses, electronic devices, and storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0294] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A vehicle temperature detection method characterized by, The method includes: Data is collected from vehicles passing through a road scene, and a thermal image is obtained by thermal imaging of the vehicles. Determine the target detection location of the vehicle in the collected data; The thermal region corresponding to the target detection location is determined in the thermal map; Based on the regional information of the thermal zone, detect whether the temperature of the target detection part of the vehicle is abnormal; Determining the target detection location of the vehicle in the collected data includes: Extract data features from the collected data; based on the data features, obtain vehicle structure information of the vehicle, wherein the vehicle structure information includes at least one of the following: window information, vehicle outline information, and axle information, and the vehicle structure information can represent the relative positions between the window information, vehicle outline information, and axle information; determine the target detection area of ​​the vehicle based on the vehicle structure information, wherein the target detection area includes: the driver's cab area and / or the engine area; The process of obtaining data on vehicles passing through a road scene includes: obtaining scene data on the road scene, including radar data collected by radar and visible light images collected by a visible light sensor, wherein the radar and the visible light sensor are fixedly installed inside the housing of a vehicle temperature detection device; performing vehicle detection on the scene data to obtain detection results; if the detection results indicate that a vehicle has passed through the road scene, then the scene data is determined as data collected on vehicles passing through the road scene. The process of detecting vehicles from the scene data includes: fusing the radar data and the visible light image based on a second transformation relationship between the radar coordinate system corresponding to the radar data and the image coordinate system corresponding to the visible light image; and detecting vehicles based on the fused data. The second transformation relationship is obtained as follows: first calibration radar data for the road scene is obtained; a first driving trajectory of a vehicle in the road scene is determined based on the first calibration radar data; the angle between the first driving trajectory and the road in the road scene is obtained; the radar coordinate system corresponding to the radar data is corrected based on the angle; calibration data sets for the road scene are continuously obtained, wherein the calibration data sets include: second calibration radar data and calibration visible light images obtained synchronously; for each obtained calibration data set, vehicle detection is performed on the second calibration radar data and the calibration visible light image respectively to obtain radar vehicle detection results and visible light vehicle detection results; based on the radar vehicle detection results and visible light vehicle detection results, the coordinate transformation relationship between the radar coordinate system and the image coordinates is determined; if the difference between the coordinate transformation relationships corresponding to two adjacent calibration data sets is less than a preset error, the second transformation relationship is obtained based on the coordinate transformation relationships corresponding to the two adjacent calibration data sets.

2. The method according to claim 1, characterized in that, Determining the target detection location of the vehicle in the collected data includes: Detect vehicle parts of a preset type in the collected data; Obtain the detection time period for each detected vehicle part; The target detection parts are determined from the detected vehicle parts whose detection time period is longer than a preset time threshold and whose relative order of the start time of the detection time period is consistent with the actual relative arrangement order of each preset type of vehicle parts in the vehicle.

3. The method according to claim 1, characterized in that, Determining the target detection location of the vehicle based on the vehicle structure information includes: Based on the window information, the driver's cab area of ​​the vehicle is determined, and the driver's cab area is designated as the target detection area of ​​the vehicle; and / or the license plate position of the vehicle is obtained based on the data features, and the engine area of ​​the vehicle is determined based on the window information, vehicle outline information, and the license plate position, and the engine area is designated as the target detection area; and / or the tire area of ​​the vehicle is determined based on the hub position in the axle information, and the tire area is designated as the target detection area; and / or the passenger compartment area of ​​the vehicle is determined based on the hub position, hub type, and number of hubs in the axle information, and the passenger compartment area is designated as the target detection area; or Determining the target detection location of the vehicle based on the vehicle structure information includes: Select target information for determining the target type from the vehicle structure information, wherein the target type is determined based on the type of the road scene; determine the target detection part of the vehicle based on the target information.

4. The method according to claim 1, characterized in that, Determining the target detection part of the vehicle in the collected data includes: determining the target type of the vehicle part to be temperature detected based on the scene type of the road scene; and determining the target detection part of the target type in the collected data. or The step of determining the thermal region corresponding to the target detection location in the thermal map includes: obtaining the time difference between the acquisition time of the thermal map and the acquisition time of the acquired data; adjusting the first transformation relationship between the coordinate system corresponding to the thermal map and the coordinate system corresponding to the acquired data based on the time difference; and determining the thermal region corresponding to the target detection location in the thermal map based on the adjusted first transformation relationship.

5. A method for detecting vehicle temperature, characterized in that, The method includes: Data is collected from vehicles passing through a road scene, and a thermal image is obtained by thermal imaging of the vehicles. Determine the target detection location of the vehicle in the collected data; The thermal region corresponding to the target detection location is determined in the thermal map; Based on the regional information of the thermal zone, detect whether the temperature of the target detection part of the vehicle is abnormal; Obtain the license plate number of the vehicle; Based on the license plate number, an early warning is issued for the target detection area; Based on the collected data, the vehicle's driving information is monitored; Based on the driving information, determine whether the vehicle has issued a warning response; If not, then based on the license plate number, an upgraded warning will be issued for the target detection area; Determining the target detection location of the vehicle in the collected data includes: Extract data features from the collected data; based on the data features, obtain vehicle structure information of the vehicle, wherein the vehicle structure information includes at least one of the following: window information, vehicle outline information, and axle information, and the vehicle structure information can represent the relative positions between the window information, vehicle outline information, and axle information; determine the target detection part of the vehicle based on the vehicle structure information; The process of obtaining data on vehicles passing through a road scene includes: obtaining scene data on the road scene, including radar data collected by radar and visible light images collected by a visible light sensor, wherein the radar and the visible light sensor are fixedly installed inside the housing of a vehicle temperature detection device; performing vehicle detection on the scene data to obtain detection results; if the detection results indicate that a vehicle has passed through the road scene, then the scene data is determined as data collected on vehicles passing through the road scene. The process of detecting vehicles from the scene data includes: fusing the radar data and the visible light image based on a second transformation relationship between the radar coordinate system corresponding to the radar data and the image coordinate system corresponding to the visible light image; and detecting vehicles based on the fused data. The second transformation relationship is obtained as follows: first calibration radar data for the road scene is obtained; a first driving trajectory of a vehicle in the road scene is determined based on the first calibration radar data; the angle between the first driving trajectory and the road in the road scene is obtained; the radar coordinate system corresponding to the radar data is corrected based on the angle; calibration data sets for the road scene are continuously obtained, wherein the calibration data sets include: second calibration radar data and calibration visible light images obtained synchronously; for each obtained calibration data set, vehicle detection is performed on the second calibration radar data and the calibration visible light image respectively to obtain radar vehicle detection results and visible light vehicle detection results; based on the radar vehicle detection results and visible light vehicle detection results, the coordinate transformation relationship between the radar coordinate system and the image coordinates is determined; if the difference between the coordinate transformation relationships corresponding to two adjacent calibration data sets is less than a preset error, the second transformation relationship is obtained based on the coordinate transformation relationships corresponding to the two adjacent calibration data sets.

6. The method according to claim 5, characterized in that, The step of determining whether the vehicle has issued a warning response based on the driving information includes: If the driving information includes a second driving trajectory, determine whether the second driving trajectory deviates from the original driving trajectory and heads towards a preset abnormal handling location. If so, determine that the vehicle has issued a warning response. And / or, If the driving information includes driving speed, it is determined whether the vehicle is decelerating based on the driving speed. If so, it is determined that the vehicle has issued a warning response.

7. A vehicle temperature detection device, characterized in that, The device includes: The data collector is used to acquire data on vehicles passing through a road scene and to send the acquired data to the main control board. A thermal imaging module is used to obtain a thermal map of the vehicle by performing thermal imaging; and to send the thermal map to the main control board. The main control board is used to determine the target detection part of the vehicle in the collected data; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region. The main control board is specifically used to extract data features from the collected data; obtain vehicle structure information of the vehicle based on the data features, wherein the vehicle structure information includes at least one of the following: window information, vehicle outline information, and axle information, and the vehicle structure information can represent the relative positions between the window information, vehicle outline information, and axle information; determine the target detection part of the vehicle based on the vehicle structure information, the target detection part including: the driver's cab area and / or the engine area; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region. The data acquisition unit includes a radar and a visible light sensor, which are fixedly installed inside the housing of the vehicle temperature detection device. Specifically, the radar is used to acquire radar data for a road scene and send the radar data to the main control board. The visible light sensor is specifically used to acquire a visible light image for a road scene and send the visible light image to the main control board. The main control board is specifically used to perform vehicle detection on the scene data including the radar data and the visible light image, and obtain a detection result. If the detection result indicates that a vehicle has passed through the road scene, then the scene data is determined as the data collected for the vehicle that passed through the road scene. The main control board is specifically used to perform data fusion on the radar data and the visible light image based on a second transformation relationship between the radar coordinate system corresponding to the radar data and the image coordinate system corresponding to the visible light image; and to perform vehicle detection based on the fused data to obtain detection results. The second transformation relationship is obtained as follows: first calibration radar data for the road scene is obtained; a first driving trajectory of a vehicle in the road scene is determined based on the first calibration radar data; the angle between the first driving trajectory and the road in the road scene is obtained; the radar coordinate system corresponding to the radar data is corrected based on the angle; calibration data sets for the road scene are continuously obtained, wherein the calibration data sets include: second calibration radar data and calibration visible light images obtained synchronously; for each obtained calibration data set, vehicle detection is performed on the second calibration radar data and the calibration visible light image respectively to obtain radar vehicle detection results and visible light vehicle detection results; based on the radar vehicle detection results and visible light vehicle detection results, the coordinate transformation relationship between the radar coordinate system and the image coordinates is determined; if the difference between the coordinate transformation relationships corresponding to two adjacent calibration data sets is less than a preset error, the second transformation relationship is obtained based on the coordinate transformation relationships corresponding to the two adjacent calibration data sets.

8. The device according to claim 7, characterized in that, The main control board is specifically used to detect vehicle parts of a preset type in the collected data; obtain the detection time period of each detected vehicle part; and determine the target detection parts from the detected vehicle parts whose detection time period is longer than a preset time threshold and whose relative order of the start time of the detection time period is consistent with the actual relative arrangement order of each preset type of vehicle part in the vehicle. or The main control board is specifically used to extract data features from the collected data; obtain vehicle structure information of the vehicle based on the data features, wherein the vehicle structure information includes at least one of the following: window information, vehicle outline information, and axle information; determine the driver's cab area of ​​the vehicle based on the window information, and use the driver's cab area as the target detection part of the vehicle; and / or obtain the license plate position of the vehicle based on the data features, and determine the engine area of ​​the vehicle based on the window information, vehicle outline information, and license plate position, and use the engine area as the target detection part; and / or determine the tire area of ​​the vehicle based on the hub position in the axle information, and use the tire area as the target detection part; and / or determine the passenger compartment area of ​​the vehicle based on the hub position, hub type, and number of hubs in the axle information, and use the passenger compartment area as the target detection part; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region. or The main control board is specifically used to extract data features from the collected data; obtain vehicle structure information of the vehicle based on the data features, wherein the vehicle structure information includes at least one of the following: window information, vehicle outline information, and axle information; select target information from the vehicle structure information for determining the target type, wherein the target type is determined based on the type of the road scene; determine the target detection part of the vehicle based on the target information; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region. or The main control board is specifically used to determine the target type of the vehicle part to be temperature detected based on the scene type of the road scene; determine the target detection part of the target type in the collected data; determine the thermal region corresponding to the target detection part in the heat map; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region. or The main control board is specifically used to determine the target detection part of the vehicle in the collected data; obtain the time difference between the heat map acquisition time and the data acquisition time; adjust the first transformation relationship between the coordinate system corresponding to the heat map and the coordinate system corresponding to the collected data based on the time difference; determine the thermal region corresponding to the target detection part in the heat map based on the adjusted first transformation relationship; and detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region. or The main control board is specifically used to determine the target detection part of the vehicle in the collected data; determine the thermal region corresponding to the target detection part in the heat map; detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region; after detecting the abnormal temperature of the target detection part of the vehicle, obtain the vehicle's license plate number; issue an early warning for the target detection part based on the license plate number; monitor the vehicle's driving information based on the collected data; determine whether the vehicle has issued an early warning response based on the driving information; if not, issue an upgraded early warning for the target detection part based on the license plate number. or The main control board is specifically used to determine the target detection part of the vehicle in the collected data; determine the thermal region corresponding to the target detection part in the heat map; detect whether the temperature of the target detection part of the vehicle is abnormal based on the regional information of the thermal region; after detecting the temperature abnormality of the target detection part of the vehicle, obtain the vehicle's license plate number; issue a warning for the target detection part based on the license plate number; monitor the vehicle's driving information based on the collected data; if the driving information includes a second driving trajectory, determine whether the second driving trajectory deviates from the original driving trajectory and heads towards a preset abnormal handling location; if so, determine that the vehicle has responded to a warning; and / or, if the driving information includes driving speed, determine whether the vehicle is decelerating based on the driving speed; if so, determine that the vehicle has responded to a warning; if not, issue an upgraded warning for the target detection part based on the license plate number.

9. A vehicle temperature detection device, characterized in that, The device includes: The data acquisition module is used to acquire collected data of vehicles passing through the road scene and to obtain a thermal map obtained by thermal imaging of the vehicles. The detection location determination module is used to determine the target detection location of the vehicle in the collected data; A thermal region determination module is used to determine the thermal region corresponding to the target detection location in the thermal map. The temperature anomaly detection module is used to detect whether the temperature of the target detection part of the vehicle is abnormal based on the area information of the thermal zone. The detection site determination module includes: A data feature extraction unit is used to extract data features from the collected data; a structural information acquisition unit is used to obtain vehicle structural information of the vehicle based on the data features, wherein the vehicle structural information includes at least one of the following: window information, vehicle outline information, and axle information, and the vehicle structural information can represent the relative positions between the window information, vehicle outline information, and axle information; a detection location determination unit is used to determine the target detection location of the vehicle based on the vehicle structural information, wherein the target detection location includes: the driver's cab area and / or the engine area; The data acquisition module is specifically used to acquire scene data, including radar data and visible light images, collected for a road scene; to perform vehicle detection on the scene data and obtain detection results; if the detection results indicate that a vehicle has passed through the road scene, then the scene data is determined as the data collected for the vehicle that passed through the road scene. The data acquisition module is specifically used to acquire scene data, including radar data and visible light images, for a road scene; based on a second transformation relationship between the radar coordinate system corresponding to the radar data and the image coordinate system corresponding to the visible light image, the radar data and the visible light image are fused; vehicle detection is performed based on the fused data to obtain a detection result; if the detection result indicates that a vehicle has passed through the road scene, the scene data is determined as the data collected for the vehicle that passed through the road scene. The second transformation relationship is obtained as follows: first calibration radar data for the road scene is obtained; a first driving trajectory of a vehicle in the road scene is determined based on the first calibration radar data; the angle between the first driving trajectory and the road in the road scene is obtained; the radar coordinate system corresponding to the radar data is corrected based on the angle; calibration data sets for the road scene are continuously obtained, wherein the calibration data sets include: second calibration radar data and calibration visible light images obtained synchronously; for each obtained calibration data set, vehicle detection is performed on the second calibration radar data and the calibration visible light image respectively to obtain radar vehicle detection results and visible light vehicle detection results; based on the radar vehicle detection results and visible light vehicle detection results, the coordinate transformation relationship between the radar coordinate system and the image coordinates is determined; if the difference between the coordinate transformation relationships corresponding to two adjacent calibration data sets is less than a preset error, the second transformation relationship is obtained based on the coordinate transformation relationships corresponding to the two adjacent calibration data sets.

10. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method of any one of claims 1-4 or 5-6.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-4 or 5-6.

Images