CN117782331A

CN117782331A - Method for detecting tire surface temperature and system for detecting tire surface temperature

Info

Publication number: CN117782331A
Application number: CN202311852158.7A
Authority: CN
Inventors: 袁嵩; 郭文鹏; 纪平; 吴玉林
Original assignee: Sailun Jinyu Group Co Ltd
Current assignee: Sailun Jinyu Group Co Ltd
Priority date: 2023-12-28
Filing date: 2023-12-28
Publication date: 2024-03-29

Abstract

The application discloses a method for detecting the surface temperature of a tire and a system for detecting the surface temperature of the tire. Wherein the method comprises the following steps: acquiring image information, wherein the image information comprises: a first optical image of a target object and a first thermal imaging image of the target object, the target object being a vehicle to which a plurality of tires belong; analyzing the first optical image by adopting a detection model, and identifying the position information of each tire; dividing the first thermal imaging image according to the position information of each tire to obtain a plurality of second thermal imaging images, wherein each second thermal imaging image only displays one tire; the surface temperature of each tire is determined from each second thermographic image. The method and the device solve the technical problems that the temperature of the whole tire surface cannot be detected due to the fact that the temperature of the inside of the tire or the temperature of a certain point of the surface of the tire can only be detected in the related art, and the detection result is inaccurate.

Description

Method for detecting tire surface temperature and system for detecting tire surface temperature

Technical Field

The application relates to the technical field of image processing, in particular to a method for detecting the surface temperature of a tire and a system for detecting the surface temperature of the tire.

Background

With the high-speed development of mining industry in China, logistics transportation tasks of mining vehicles are more and more frequent, and the mining vehicles play an increasingly important role in a mining transportation system. Mine tires are non-utility tires, and are more resistant to durability, wear resistance, puncture resistance, and traction than passenger tires, which allow mine truck tires to accommodate harsh working environments and heavy duty applications. In the heavy-load transportation process of the mine car, due to factors such as friction, extrusion, compression, expansion and the like, heat of the tire is continuously accumulated, the abrasion and aging process of the tire can be accelerated at high temperature, uneven abrasion of the tire and deterioration and thermal decomposition of sizing materials can be caused due to the fact that the surface temperature of the tire is too high, and the service life of the tire is shortened. Frequent local high temperature can also cause the tire to crack and damage, and the tire can be spontaneous or burst seriously, so that the service life of the mine car tire is influenced, and huge economic loss is brought. At present, an infrared detector and a sensor are adopted to detect the surface temperature of the tire in the temperature detection method of the tire; however, infrared detectors and sensors can only perform point temperature measurement, and cannot accurately detect the tire surface temperature.

In view of the above problems, no effective solution has been proposed at present.

Disclosure of Invention

The embodiment of the application provides a method for detecting the surface temperature of a tire and a system for detecting the surface temperature of the tire, which at least solve the technical problems that the temperature of the surface of the whole tire cannot be detected due to the fact that the temperature of the inside of the tire or the temperature of a certain point of the surface of the tire can be detected in the related technology, and the detection result is inaccurate.

According to one aspect of an embodiment of the present application, there is provided a method of detecting a tire surface temperature, comprising: acquiring image information, wherein the image information comprises: a first optical image of a target object and a first thermal imaging image of the target object, the target object being a vehicle to which a plurality of tires belong; analyzing the first optical image by using a detection model to identify the position information of each tire, wherein the position information of each tire comprises: the system comprises a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in a first optical image as a coordinate origin; dividing the first thermal imaging image according to the position information of each tire to obtain a plurality of second thermal imaging images, wherein each second thermal imaging image only displays one tire; the surface temperature of each tire is determined from each second thermographic image.

Optionally, analyzing the first optical image with a detection model to identify positional information of each tire, including: detecting the first optical image through a detection model to obtain a detection result, wherein the detection result is used for indicating whether a tire image exists in the first optical image; generating a plurality of block diagrams in the case that the detection result indicates that a tire image exists in the first optical image, wherein each block diagram only contains one tire; determining a confidence value of each block diagram, and determining the block diagram with the confidence value larger than a preset threshold value as a target block, wherein the confidence value is used for indicating the accuracy of a prediction result output by the detection model; and determining the abscissa of the center point of each target frame, the ordinate of the center point of each target frame, the width value of each target frame and the height value of each target frame as the position information of the tire corresponding to each target frame.

Optionally, the detection model is trained by the following method: acquiring a data set, wherein the data set is generated according to an optical image set of a target object, and the data set comprises: training data, verification data and test data, wherein the proportion of the training data, the verification data and the test data is a preset proportion, and the optical image set comprises: a history optical image of the target object, a real-time optical image of the target object, and a target optical image obtained by performing conversion processing on the history optical image and/or the real-time optical image; and determining a loss function according to the data set, and training a machine learning model based on the data set and the loss function to obtain a detection model.

Optionally, determining the loss function from the dataset includes: taking the training data as input information of a machine learning model to obtain predicted position information of a target frame corresponding to each tire output by the machine learning model, wherein the predicted position information comprises: a predicted abscissa, a predicted ordinate, a predicted height value, and a predicted width value of each target frame; acquiring actual position information of each target frame, wherein the actual position information is stored in verification data and comprises: the actual abscissa and the actual ordinate of each target frame; determining a loss function according to the association relation among the predicted position information, the actual position information and the test data, wherein the test data comprises: a plurality of preset parameters stored in a configuration file of the machine learning model.

Optionally, the predicted position information is obtained by: performing feature transformation on each second optical image in the training data to obtain a first feature image and a second feature image, wherein the first feature image is obtained by performing feature transformation on each second optical image in the up-down direction, and the second feature image is obtained by performing feature transformation on each second optical image in the left-right direction; determining a first weight feature vector of the second optical image in the up-down direction and a second weight feature vector of the second optical image in the left-right direction according to the first feature map and the second feature map; and determining the predicted position information according to the second optical image, the first weight feature vector and the second weight feature vector.

Optionally, determining the surface temperature of each tire from each second thermographic image comprises: cutting each second thermal imaging image to obtain third thermal imaging images, wherein the area of each third thermal imaging image is smaller than that of the corresponding second thermal imaging image, and each third thermal imaging image does not contain an image of a hub; and determining a pixel value of each pixel point in a plurality of pixel points of each third thermal imaging image, wherein the pixel value of each pixel point is the surface temperature of the tire area corresponding to each pixel point.

Optionally, the method of detecting the temperature of the surface of the tire further comprises: respectively comparing a plurality of pixel values of the plurality of pixel points with a preset temperature threshold value to obtain a comparison result; outputting alarm information when the comparison result indicates that a target pixel value larger than a preset temperature threshold exists in the pixel values, wherein the alarm information is used for prompting a user that the surface temperature of the tire is higher than the preset temperature threshold, and comprises first lamplight information, first image information and sound information; and outputting prompt information when the comparison result indicates that a target pixel value larger than a preset temperature threshold value does not exist in the pixel values, wherein the prompt information is used for prompting a user that the surface temperature of the tire is in a preset temperature interval, and the prompt information comprises second lamplight information and second image information.

According to another aspect of the embodiments of the present application, there is also provided a system for detecting a tire surface temperature, the system for detecting a tire surface temperature including: the system comprises an image acquisition module, an information processing module and a temperature early warning module, wherein the image acquisition module is used for acquiring image information, and the image acquisition module comprises: the camera system comprises a first camera system and a second camera system which is in parallel relation with the first camera system, wherein the included angles between the shooting directions of the first camera system and the second camera system and the horizontal direction are first preset angles, the included angles between the connecting line directions of the first camera system and the second camera system and the horizontal direction are second preset angles, and the first camera system comprises: a first camera and a second camera positioned adjacent to the first camera, the second camera system comprising: the first camera and the third camera are used for acquiring a first optical image of a target object, the second camera and the fourth camera are used for acquiring a first thermal imaging image of the target object, and the target object is a vehicle to which a plurality of tires belong; the information processing module is connected with the image acquisition module and comprises: and a detection model for analyzing the first optical image to identify positional information of each tire, wherein the positional information of each tire includes: the system comprises a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in a first optical image as a coordinate origin; the information processing module is further used for carrying out segmentation processing on the first thermal imaging image according to the position information of each tire to obtain a plurality of second thermal imaging images, wherein each second thermal imaging image only displays one tire; the temperature pre-warning module is used for determining the surface temperature of each tire according to each second thermal imaging image.

Optionally, the temperature early warning module is configured to perform cutting processing on each second thermal imaging image to obtain third thermal imaging images, where an area of each third thermal imaging image is smaller than an area of a second thermal imaging image corresponding to each third thermal imaging image, and each third thermal imaging image does not include an image of the hub; and determining a pixel value of each pixel point in a plurality of pixel points of each third thermal imaging image, wherein the pixel value of each pixel point is the surface temperature of the tire area corresponding to each pixel point.

Optionally, the temperature early warning module is further configured to compare a plurality of pixel values of the plurality of pixel points with a preset temperature threshold respectively, so as to obtain a comparison result; outputting alarm information when the comparison result indicates that a target pixel value larger than a preset temperature threshold exists in the pixel values, wherein the alarm information is used for prompting a user that the surface temperature of the tire is higher than the preset temperature threshold, and comprises first lamplight information, first image information and sound information; and outputting prompt information when the comparison result indicates that a target pixel value larger than a preset temperature threshold value does not exist in the pixel values, wherein the prompt information is used for prompting a user that the surface temperature of the tire is in a preset temperature interval, and the prompt information comprises second lamplight information and second image information.

According to another aspect of the embodiments of the present application, there is also provided an apparatus for detecting a tire surface temperature, including: the device comprises an acquisition module for acquiring image information, wherein the image information comprises: a first optical image of a target object and a first thermal imaging image of the target object, the target object being a vehicle to which a plurality of tires belong; the first determining module is configured to analyze the first optical image with a detection model, and identify position information of each tire, where the position information of each tire includes: the system comprises a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in a first optical image as a coordinate origin; the segmentation module is used for carrying out segmentation processing on the first thermal imaging image according to the position information of each tire to obtain a plurality of second thermal imaging images, wherein each second thermal imaging image only displays one tire; and a second determining module for determining the surface temperature of each tire from each second thermographic image.

According to another aspect of the embodiments of the present application, there is also provided a non-volatile storage medium, in which a computer program is stored, where the apparatus in which the non-volatile storage medium is located executes the above-mentioned method for detecting the tire surface temperature by running the computer program.

According to another aspect of the embodiments of the present application, there is also provided an electronic device comprising a memory in which a computer program is stored, and a processor arranged to perform the above-described method of detecting the temperature of a tyre surface by means of the computer program.

In the embodiment of the application, the image information is acquired, wherein the image information includes: a first optical image of a target object and a first thermal imaging image of the target object, the target object being a vehicle to which a plurality of tires belong; analyzing the first optical image by using a detection model to identify the position information of each tire, wherein the position information of each tire comprises: the system comprises a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in a first optical image as a coordinate origin; dividing the first thermal imaging image according to the position information of each tire to obtain a plurality of second thermal imaging images, wherein each second thermal imaging image only displays one tire; determining the surface temperature of each tire according to each second thermal imaging image, processing the image of the vehicle in which the tire is positioned by a deep learning-based algorithm, positioning the tire in the image, and processing the infrared thermal imaging image of the tire according to the positioning result by the deep learning-based algorithm to obtain a tire infrared thermal imaging image, namely a tire surface temperature field; finally, the tire surface temperature is determined according to the tire surface temperature field, the purpose of determining the tire surface temperature under the condition of no need of additional sensors and communication equipment is achieved by combining an algorithm and an infrared technology, the technical effects of engineering tire surface temperature detection and early warning are achieved, the technical effects of improving the safety and service life of an engineering tire are further achieved, and the technical problems that the whole tire surface temperature cannot be detected due to the fact that only the internal temperature of the tire or the temperature of a certain point on the tire surface can be detected in the related technology are solved, and the detection result is inaccurate are solved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:

FIG. 1 is a block diagram of the hardware architecture of a computer terminal for implementing a method of detecting tire surface temperature according to an embodiment of the present application;

FIG. 2 is a flowchart of steps of a method of detecting a tire surface temperature according to an embodiment of the present application;

FIG. 3 is a schematic diagram of a system for detecting tire surface temperature according to an embodiment of the present application;

FIG. 4 is a schematic installation diagram of a system for detecting tire surface temperature according to an embodiment of the present application;

FIG. 5 is a block diagram of an apparatus for detecting tire surface temperature according to an embodiment of the present application;

FIG. 6 is a flowchart of the operation of an apparatus for detecting the temperature of a tire surface according to an embodiment of the present application.

Detailed Description

In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that embodiments of the present application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

In the related art, the temperature of a tire is detected using a sensor, an infrared detector, and some communication devices, wherein the temperature sensor is installed on a rim to detect the temperature of the tire, and only the temperature inside the tire can be detected; and the surface temperature of the tire cannot be calculated based on the temperature inside the tire; the infrared detector is adopted to only perform point temperature measurement, and the whole tire surface cannot be detected; in addition, the sensor and the communication device are costly and cumbersome to install. The equipment such as the temperature sensor is high in damage risk under the off-highway environment, is difficult to maintain, has false alarm and fault risk, and therefore, the problem that the engineering tire surface temperature cannot be accurately detected exists. In order to solve this problem, related solutions are provided in the embodiments of the present application, and are described in detail below.

In accordance with the embodiments of the present application, there is provided an embodiment of a method of detecting tire surface temperature, it being noted that the steps illustrated in the flowchart of the figures may be performed in a computer system, such as a set of computer executable instructions, and, although a logical sequence is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in a different order than that illustrated herein.

The method embodiments provided by the embodiments of the present application may be performed in a mobile terminal, a computer terminal, or similar computing device. Fig. 1 shows a block diagram of the hardware architecture of a computer terminal for implementing a method of detecting the temperature of the surface of a tyre. As shown in fig. 1, the computer terminal 10 may include one or more processors 102 (shown as 102a, 102b, … …,102 n) 102 (the processor 102 may include, but is not limited to, a microprocessor MCU or a processing device such as a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, the method may further include: a display, an input/output interface (I/O interface), a Universal Serial BUS (USB) port (which may be included as one of the ports of the BUS), a network interface, a power supply, and/or a camera. It will be appreciated by those of ordinary skill in the art that the configuration shown in fig. 1 is merely illustrative and is not intended to limit the configuration of the electronic device described above. For example, the computer terminal 10 may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.

It should be noted that the one or more processors 102 and/or other data processing circuits described above may be referred to generally herein as "data processing circuits. The data processing circuit may be embodied in whole or in part in software, hardware, firmware, or any other combination. Furthermore, the data processing circuitry may be a single stand-alone processing module or incorporated, in whole or in part, into any of the other elements in the computer terminal 10. As referred to in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of the path of the variable resistor termination to interface).

The memory 104 may be used to store software programs and modules of application software, such as program instructions/data storage devices corresponding to the method for detecting tire surface temperature in the embodiments of the present application, and the processor 102 executes the software programs and modules stored in the memory 104 to perform various functional applications and data processing, i.e., implement the method for detecting tire surface temperature described above. Memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission means 106 is arranged to receive or transmit data via a network. The specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC) that can connect to other network devices through a base station to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module for communicating with the internet wirelessly.

The display may be, for example, a touch screen type Liquid Crystal Display (LCD) that may enable a user to interact with a user interface of the computer terminal 10.

An embodiment of the present application provides a method for detecting a tire surface temperature capable of operating in the above-described operating environment, and fig. 2 is a flowchart of steps of the method for detecting a tire surface temperature according to the embodiment of the present application, as shown in fig. 2, and the method includes the following steps:

step S202, obtaining image information, where the image information includes: a first optical image of a target object and a first thermal imaging image of the target object, the target object being a vehicle to which a plurality of tires belong.

The method for detecting the tire surface temperature provided by the embodiment of the invention is applied to a mine giant tire (engineering tire), and the tread temperature (namely the tire surface temperature) of the engineering tire is detected based on the image of the engineering tire, so in step S202, image information of a mine vehicle (namely a target object) provided with the engineering tire is firstly obtained, wherein the image information of the mine vehicle (namely the target object) at least comprises a real-time optical high-definition image (namely a first optical image) of the mine vehicle (namely the target object) and a real-time infrared thermal imaging image (namely the first thermal imaging image) of the mine vehicle (namely the target object).

Step S204, analyzing the first optical image by adopting a detection model to identify the position information of each tire, wherein the position information of each tire comprises: and a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in the first optical image as a coordinate origin.

After the image information is obtained, in step S204, a detection model is first adopted to analyze a real-time optical high-definition image (i.e., a first optical image) of the mining vehicle (i.e., a target object) in the image information, and the image of the tire in the real-time optical high-definition image (i.e., the first optical image) is positioned, so as to obtain the position information of each tire in the real-time optical high-definition image; in this embodiment, each tire (engineering tire) is positioned by using a plurality of coordinates in a coordinate system, so when each tire in the real-time optical high-definition image is positioned by using the detection model, the positioning result (i.e., the position information of each tire) obtained is a plurality of coordinates, and the coordinates are all assigned to the same (target) coordinate system, and the (target) coordinate system is established by taking a preset pixel point in the real-time optical high-definition image as a coordinate origin; for example, the preset pixel point may be a pixel point in the lower left corner of the real-time optical high-definition image.

In this embodiment, when the detection model performs detection processing on the real-time optical high-definition image (i.e., the first optical image) of the input mining vehicle (i.e., the target object), a detection result for indicating whether a tire (engineering tire) exists in the real-time optical high-definition image is output, if the detection model processes the real-time optical high-definition image to confirm that the tire exists, each tire is selected in one block frame, and since one mining vehicle (i.e., the target object) is not only provided with one tire, if the detection model detects that the image of the real-time optical high-definition image includes a plurality of tires, the plurality of tires are selected in a plurality of block frames respectively, and each block frame is ensured to be selected only for one tire. Meanwhile, if the detection result is that tires exist in the real-time optical high-definition image, when the detection model selects a plurality of tires by using a plurality of block frames, outputting a confidence coefficient (conf) value for each block frame, wherein the confidence coefficient value is the evaluation of the detection model on the positioning result (namely the prediction result) output by the detection model and is used for evaluating the accuracy of the tire as the object selected by the block frames; in this embodiment, only the block diagram (i.e., the target frame) with the confidence value larger than the preset threshold is used as the positioning result for the subsequent segmentation of the real-time thermal imaging image, i.e., only the positioning result of the tire in the target frame is considered to be correct. After each tire is positioned, calculating to obtain the position information of the tire selected by the frame in each target frame according to the related information of each target frame, and specifically, taking the coordinate information of the center point of each target frame (including the abscissa xi of the center point and the ordinate yi of the center point), the height value hi of each target frame and the width value wi of each target frame as the position information of the tire selected by the frame in each target frame.

According to an alternative embodiment of the present application, the detection model is trained by the following method: acquiring a data set, wherein the data set is generated according to an optical image set of a target object, and the data set comprises: training data, verification data and test data, wherein the proportion of the training data, the verification data and the test data is a preset proportion, and the optical image set comprises: a history optical image of the target object, a real-time optical image of the target object, and a target optical image obtained by performing conversion processing on the history optical image and/or the real-time optical image; and determining a loss function according to the data set, and training a machine learning model based on the data set and the loss function to obtain a detection model.

The detection model adopted in the step S204 is obtained by training a machine model by using an improved detection algorithm (YOLOv 8 algorithm), wherein the data for training the machine learning model is a data set formed by an optical image set of a mining vehicle (i.e. a target object), and the optical image set of the mining vehicle (i.e. the target object) comprises a mining vehicle high-definition image (i.e. a real-time optical image of the target object) shot in real time on site, a mining vehicle high-definition image (i.e. a historical optical image of the target object) downloaded from a public data source and a transformed optical image (i.e. a target optical image) obtained by transforming any optical image (either the real-time optical image or the historical optical image) in the optical image set; wherein, the purpose of the transformation processing is to carry out data enhancement on the data set, and the quantity of the data in the data set is amplified, and the transformation processing comprises the following steps: random cropping (Random Crop), random Rotation (Random Rotation), random Scaling (Random Scaling), random Flip (Random Flip), random brightness, contrast and tone adjustment (Random Brightness, contrast, and Hue Adjustment), noise Addition (Noise Addition), and the like, wherein Random cropping is to Crop any optical image in an optical image set, and sub-images of different areas of the image can increase the diversity of sample data by cropping; the random rotation is to rotate any optical image in the optical image set, and the robustness of the detection model to rotation transformation can be increased by using the randomly rotated image as training data; the random scaling is to scale any optical image in the optical image set, and the randomly scaled image can simulate objects with different scales, so that the adaptability of the detection model to scale changes is improved; the random overturn is to horizontally or vertically overturn any optical image in the optical image set, and the image after random overturn can simulate mirror image, symmetry and other scenes, so that the robustness of the detection model to mirror image transformation is improved; the image subjected to random brightness, contrast and tone adjustment is used as training data, so that the adaptability of the detection model to different illumination conditions and color changes can be improved; noise is added to any optical image in the optical image set, noise interference in a real scene can be simulated, and the robustness of a detection model can be increased by adopting the image subjected to noise addition processing as training data. In this embodiment, the data in the data set is classified into training data according to a preset proportion, and the verification data and the test data may be divided according to a proportion of 8:1:1, for example. Before the data set is divided, labeling the high-definition optical image in the data set, labeling the tire surface in the high-definition optical image with a label of tire (tre), and storing the high-definition optical image with the label of tire (tre) and a source image before the label corresponding to the high-definition optical image as txt labels according to the data set labeling format of YOLOv 8; the training data, the verification data and the test data obtained by dividing according to a preset proportion are stored in one folder among folders storing the training data of the detection model, for example, in an images folder, and simultaneously, in another folder among folders storing the training data of the detection model, for example, in a "labels" folder, wherein the folder storing the training data, the verification data and the test data and the folder storing the data stored as txt labels are different folders under the same folder. In training the machine learning model with the modified detection algorithm (YOLOv 8 algorithm), a loss function is used in addition to the dataset, where the loss function is not the CIoU loss function commonly used by YOLOv8 algorithm. But rather a Wise-IOU penalty function generated from the data set described above.

In this embodiment, the loss function used in training the detection model is L _WIOU (Wise-IOU penalty function), L _WIOU Generated based on the data set and the prediction result (i.e., the predicted position information) output by the machine learning model in the training process; during training, the machine learning model receives input training data and outputs a prediction result, wherein the prediction result comprises: prediction abscissa x _p Predicted aspect y of each target frame _p The machine learning model such as the predicted height value H of each target frame and the predicted width value W of each target frame aims at the predicted position information which is output by each target frame which is used for framing each tire in the high-definition optical image and has a confidence value larger than a preset value. Next, the actual abscissa x of each target frame recorded in the validation data of the dataset is acquired _gt And the actual ordinate y of each target frame _gt . Determining a loss function L from predicted location information in training data and actual location information in validation data and test data stored in a configuration file of a machine learning model _WIOU Specifically, the area a of the intersection area (first target area) of the predicted target frame (first target frame) corresponding to the predicted position information and the actual target frame (second target frame) corresponding to the actual position information and the area B of the union area (second target area) of the first target frame and the second target frame are first determined, and the ratio of the area a to the area B is recorded as the current predicted frame intersection ratio loss valueAccording to multiple->Determining a prediction frame dynamic cross ratio loss mean +.>Finally obtaining the outlier degree beta, < ->Next, the hyper-parameters α and δ in the test data are obtained, according to the formula r=β/δα ^β-δ Obtaining the gain r of the current prediction frame; at the same time, on-machineThe machine learning model screens out a predicted height value H with the smallest value from a predicted height value H of each target frame and a predicted width value W of each target frame in a plurality of target frames output by a currently input high-definition image _g Predictive width value W with minimum he value _g According to the formula->Determining the difference R between a predicted frame and an actual target frame output by a machine learning model _WIoU Wherein exp represents the exponential function, and finally, the simultaneous equationsI.e. the loss function L can be obtained _WIoU 。

According to another alternative embodiment of the present application, the predicted position information is obtained by: performing feature transformation on each second optical image in the training data to obtain a first feature image and a second feature image, wherein the first feature image is obtained by performing feature transformation on each second optical image in the up-down direction, and the second feature image is obtained by performing feature transformation on each second optical image in the left-right direction; determining a first weight feature vector of the second optical image in the up-down direction and a second weight feature vector of the second optical image in the left-right direction according to the first feature map and the second feature map; and determining the predicted position information according to the second optical image, the first weight feature vector and the second weight feature vector.

In this embodiment, an attention mechanism (Coordinate attention) is introduced when training a machine learning model, and in the training process, the machine learning model outputs a prediction result (i.e., predicted position information) based on the attention mechanism, and the specific method is as follows, after receiving training data, the machine learning model performs feature transformation in a height (H) direction (i.e., an up-down direction) and a width (W) (i.e., a left-right direction) respectively, and the specific feature transformation process includes pooling, stitching and convolution. The machine learning model is first based on the formulaAnd->Pooling an optical image (i.e., a second optical image) of the mining vehicle in the training data, wherein W _num Indicating the total number of pixels in the W direction, H _num The total number of pixels in the H direction; x is x _c (H, i) indicating to pool the H direction,/>The result is obtained by carrying out H-direction pooling operation on training data; x is x _c (j, W) indicates pooling of W direction,>the result obtained by the pooling operation in the W direction is obtained for the training data. Next, according to the formula f=δ (F ₁ ([z ^h ,z ^w ]) According to +.>And->Splicing together, sending the two to a convolution module with a convolution kernel of 1 to reduce the dimension of the convolution module, finally carrying out batch normalization processing, and sending to sequential processing of activating function delta activation to obtain a feature diagram f; wherein F is ₁ Representing a batch normalization function, delta being an activation function, +.>Wherein x represents any one of unknowns, and x is replaced with (F) when calculating the feature map F ₁ ([z ^h ,z ^w ])). Finally, the characteristic diagram f obtained in the splicing step is subjected to a formula g ^h ＝δ(F _h (f ^h ) A) convolving the characteristic diagram F in the H direction to obtain the characteristic diagram F in the H direction _h Reuse of activation function delta to handle F _h Obtaining F _h Weight in H direction g ^h Wherein f ^h The method comprises the steps that a convolution kernel of the feature map f is 1 multiplied by 1 according to an original H value, namely a height value of a target frame in training data; meanwhile, according to formula g ^w ＝δ(F _w (f ^w ) Convolution of the feature map F in the W direction to obtain a feature map F in the W direction _w Reuse of activation function delta to handle F _w Obtaining F _w Weight in W direction g ^w Wherein f ^w The original W value is the width value of the target frame in the training data, and is the result obtained by convolving the feature map f with the convolution kernel of 1×1 according to the original W value. After the above-described convolution, stitching and pooling operations, the method is performed according to the formula +.>Calculating to obtain a final prediction result y _c (i, j), wherein y _c (i, j) is x _c (x, j) weight vector with attention in the H and W directions, equivalent to the predicted ordinate y in the above embodiment _p ；x _c (x, j) is a vector corresponding to the inputted training data (x, j),/ >Is x _c The weight feature vector of (x, j) in the H direction can be obtained by the steps of pooling, stitching and convolution described above, < >>Is x _c The weight feature vector of (x, j) in the W direction can also be obtained by the steps of pooling, stitching and convolution described above.

In step S206, the first thermal imaging image is segmented according to the position information of each tire to obtain a plurality of second thermal imaging images, where each second thermal imaging image only shows one tire.

In step S206, the real-time ir thermal imaging image obtained in step S202 is segmented according to the positioning result (i.e., the position information of each tire) output by the detection model in step S204, and the real-time ir thermal imaging image (i.e., the first thermal imaging image) of the mine car (i.e., the target object) is segmented into a plurality of small thermal imaging images (i.e., the second thermal imaging images), wherein each small thermal imaging image (i.e., the second thermal imaging image) only displays one tire (i.e., the engineering tire), i.e., the thermal imaging image of each tire is obtained after segmentation in step S206.

Step S208, determining the surface temperature of each tire according to each second thermal imaging image.

In step S208, the thermal imaging image of each tire (i.e., each second thermal imaging image) is analyzed and processed to obtain the tread temperature (i.e., the surface temperature) of each tire.

In the present embodiment, since the whole tire is shown in the thermal imaging image of each tire (i.e., each second thermal imaging image), one tire includes both the hub portion and the tread portion, the (third) thermal imaging image of the tread is first divided from the thermal imaging image (second thermal imaging image) of the tire before the tread temperature is detected, and specifically, cut from the quarter height value or the fifth height value of the second thermal imaging image, resulting in the (third) thermal imaging image as the tread. When detecting the tread temperature (i.e., the surface temperature) of each tire, traversing each pixel point in the (third) thermal imaging image of the tread, and determining the obtained pixel value of each pixel point as the temperature of the tire tread position (i.e., the tire area) corresponding to each pixel point. For example, the height value of the (first) optical Image of the mine car is image_h, the width value of the (first) optical Image of the mine car is image_h, and the detection model is output for selecting middle frames of the high-definition optical Image The center coordinates of the target frame of one tire are (x, y), the width value of the target frame is w, the height value of the target frame is h, the detection model is processed, and the detection model is processed in the following stepsThe treatment of the disease>The cutting is performed to obtain an upper left corner coordinate (X1, Y1) and a lower right corner coordinate (X2, Y2) of a target frame included in the (third) thermal imaging Image including only the tread portion, where x1= (X-0.5×w) image_w, y1= (y+0.5×h) image_h, x2= (x+0.5×w) image_w, y2= (Y-0.5×h) image_h). The tread temperature is predicted by traversing pixel values for a target box of upper left corner coordinates (X1, Y1) and lower right corner coordinates (X2, Y2).

According to some optional embodiments of the present application, the method of detecting a tire surface temperature further comprises: respectively comparing a plurality of pixel values of the plurality of pixel points with a preset temperature threshold value to obtain a comparison result; outputting alarm information when the comparison result indicates that a target pixel value larger than a preset temperature threshold exists in the pixel values, wherein the alarm information is used for prompting a user that the surface temperature of the tire is higher than the preset temperature threshold, and comprises first lamplight information, first image information and sound information; and outputting prompt information when the comparison result indicates that a target pixel value larger than a preset temperature threshold value does not exist in the pixel values, wherein the prompt information is used for prompting a user that the surface temperature of the tire is in a preset temperature interval, and the prompt information comprises second lamplight information and second image information.

The method provided by the embodiment of the application can also realize high-temperature early warning, and when the tread temperature of the tire is detected according to the (third) thermal imaging image of the tread of the tire in the embodiment, the pixel value of each pixel point is compared with a dangerous temperature value (namely a preset temperature threshold value); if, upon comparing one of the pixel values with the preset temperature threshold, the result of the comparison is that the pixel value is higher than the preset temperature value, an alert message is sent to the user prompting the user that the tire tread temperature is in a dangerous temperature interval, e.g., in the interval [100 ℃, + -infinity), wherein, the alarm information comprises (first) lamplight information, (first) image information and (first) sound information, wherein the (first) lamplight information is an early warning lamp, the (first) image information is image information in a display screen, and the (first) sound information can be sound emitted by a buzzer. For example, when detecting that the pixel value of one pixel point is higher than a preset temperature threshold value, the warning lamp is displayed in red, the highest temperature of the tread is displayed on the display screen, and the buzzer sounds an alarm. If one of the pixel values is compared with the preset temperature threshold value, the obtained comparison result is that the pixel value is lower than the preset temperature value, which indicates that the tread temperature is in a normal temperature interval, for example, in an interval [0 ℃,100 ℃), at this time, prompt information is output to the user, and the prompt information simultaneously comprises (second) lamplight information and (second) image information and is used for prompting the user about the specific value of the current tread temperature and notifying the user that the current tread temperature is in the normal temperature interval (i.e., the preset temperature interval); the (second) image information is image information in a display screen, and the (second) light information may be an alarm lamp, for example, when the pixel value of one pixel point is detected to be lower than a preset temperature threshold value, the highest temperature of the current tread is displayed in the display screen, and the color displayed by the alarm lamp is different from that of the alarm lamp in the alarm information, for example, the alarm lamp is green.

Through the steps, the sensor is not required to be installed on the mine truck, the whole tread of the tire of the mine truck can be accurately positioned and the temperature is accurately detected as far as possible through the image of the mine truck, the defect that the tread temperature cannot be detected in the prior art can be overcome, and the equipment cost and the maintenance cost are reduced; further, when the tread temperature is detected to be higher than the temperature threshold value, early warning is timely carried out, so that the service life of the tire can be prolonged, and the use cost of the tire can be reduced.

FIG. 3 is a schematic diagram of a system for detecting tire surface temperature according to an embodiment of the present application, as shown in FIG. 3, the system for detecting tire surface temperature comprising: the system comprises an image acquisition module 30, an information processing module 32 and a temperature early warning module 34, wherein the image acquisition module 30 is used for acquiring image information, and the image acquisition module 30 comprises: the first camera system 302 and the second camera system 304 having a parallel relationship with the first camera system 302, wherein the included angles between the shooting directions of the first camera system 302 and the second camera system 304 and the horizontal direction are both a first preset angle, the included angle between the connecting line direction of the first camera system 302 and the second camera system 304 and the horizontal direction is a second preset angle, and the first camera system 302 comprises: a first camera 3022 and a second camera 3024 in a position adjacent to the first camera 3022, the second camera system 304 comprising: a third camera 3042 and a fourth camera 3044 positioned adjacent to the third camera 3042, the first camera 3022 and the third camera 3042 being configured to acquire a first optical image of a target object, the second camera 3024 and the fourth camera 3044 being configured to acquire a first thermal image of the target object, the target object being a vehicle to which a plurality of tires belong; an information processing module 32, connected to the image acquisition module 30, comprising: and a detection model for analyzing the first optical image to identify positional information of each tire, wherein the positional information of each tire includes: the system comprises a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in a first optical image as a coordinate origin; the information processing module 32 is further configured to perform segmentation processing on the first thermal imaging image according to the position information of each tire to obtain a plurality of second thermal imaging images, where each second thermal imaging image only shows one tire; the temperature pre-warning module 34 is configured to determine a surface temperature of each tire from each of the second thermographic images.

FIG. 4 is a schematic diagram of the installation of a system for detecting the temperature of the tire surface, wherein the system for detecting the temperature of the tire surface is installed on a flat road and the height from the road surface is 100 meters, and the first camera system 302 and the second camera system 304 are respectively located at two sides of the road, as shown in FIG. 4, the line direction of the first camera system 302 and the second camera system 304 forms an angle of ninety degrees with the driving direction of the road (i.e. a second preset angle), the shooting direction of the first camera system 302 forms an angle of 30 degrees with the edge of the road (i.e. a first preset angle), and the shooting direction of the second camera system 304 forms an angle of 30 degrees with the edge of the road (i.e. a first preset angle); the first camera 3022 in the first camera system 302 is an optical camera C1, the second camera 3024 is an infrared thermal imaging camera C2, and the first camera system 302 further includes a light supplementing lamp L for supplementing light to a photographed object (such as a mine car) so as to improve the definition of the photographed image. The third camera 3042 in the second camera system 304 is an optical camera C3, the fourth camera 3044 is an infrared thermal imaging camera C4, and a light filling lamp L is also included in the second camera system 304. The shooting direction of the first camera system 302 and the shooting direction of the second camera system 304 may also be adjusted to different angles, for example, to a shooting direction that is at a right angle to a road.

Optionally, the temperature pre-warning module 34 is configured to perform a cutting process on each second thermal imaging image to obtain third thermal imaging images, where an area of each third thermal imaging image is smaller than an area of a second thermal imaging image corresponding to each third thermal imaging image, and each third thermal imaging image does not include an image of the hub; and determining a pixel value of each pixel point in a plurality of pixel points of each third thermal imaging image, wherein the pixel value of each pixel point is the surface temperature of the tire area corresponding to each pixel point.

The temperature warning module 34 cuts out the (third) thermal imaging image of only the tread portion of each tire based on the predicted position information of the target frame framing each tire output from the detection model in the information processing module 32 when predicting the tread temperature of each tire (engineering tire) based on the (second) thermal imaging image of each tire. For example, if the height value of the (first) optical Image of the mine car is image_h, the width value is image_h, the center coordinates of the target frame for selecting one tire in the high-definition optical Image output by the detection model are (x, y), the width value of the target frame is w, and the height value of the target frame is H, the detection model is processed, and the Image is displayed in the Image display device The treatment of the disease>The pair is cut to obtain upper left corner coordinates (X1, Y1) and of a target frame contained in a (third) thermal imaging image containing only the tread partThe lower right hand corner is (X2, Y2), x1= (X-0.5X W) image_w, y1= (y+0.5X H) image_h, x2= (x+0.5X W) image_w, y2= (Y-0.5X H) image_h. The temperature pre-warning module 34 predicts the tread temperature for each pixel through the target frame with the upper left corner coordinates (X1, Y1) and the lower right corner coordinates (X2, Y2) in the (third) thermal imaging image, and determines the pixel value of each pixel as the temperature value of the tread area corresponding to the pixel.

According to an optional embodiment of the present application, the temperature early warning module 34 is further configured to compare a plurality of pixel values of a plurality of pixel points with a preset temperature threshold value, respectively, to obtain a comparison result; outputting alarm information when the comparison result indicates that a target pixel value larger than a preset temperature threshold exists in the pixel values, wherein the alarm information is used for prompting a user that the surface temperature of the tire is higher than the preset temperature threshold, and comprises first lamplight information, first image information and sound information; and outputting prompt information when the comparison result indicates that a target pixel value larger than a preset temperature threshold value does not exist in the pixel values, wherein the prompt information is used for prompting a user that the surface temperature of the tire is in a preset temperature interval, and the prompt information comprises second lamplight information and second image information.

The temperature early warning module 34 can also realize a high-temperature early warning function, and when the temperature early warning module 34 detects the tread temperature of the tire according to the (third) thermal imaging image of the tread of the tire, the pixel value of each pixel point is compared with a dangerous temperature value (namely a preset temperature threshold value); if, upon comparing one of the pixel values with the preset temperature threshold, the result of the comparison is that the pixel value is higher than the preset temperature value, an alert message is sent to the user prompting the user that the tire tread temperature is in a dangerous temperature interval, e.g., in the interval [100 ℃, + -infinity), wherein, the alarm information comprises (first) lamplight information, (first) image information and (first) sound information, wherein the (first) lamplight information is an early warning lamp, the (first) image information is image information in a display screen, and the (first) sound information can be sound emitted by a buzzer. For example, when detecting that the pixel value of one pixel point is higher than a preset temperature threshold value, the warning lamp is displayed in red, the highest temperature of the tread is displayed on the display screen, and the buzzer sounds an alarm. If one of the pixel values is compared with the preset temperature threshold value, the obtained comparison result is that the pixel value is lower than the preset temperature value, which indicates that the tread temperature is in a normal temperature interval, for example, in an interval [0 ℃,100 ℃), at this time, prompt information is output to the user, and the prompt information simultaneously comprises (second) lamplight information and (second) image information and is used for prompting the user about the specific value of the current tread temperature and notifying the user that the current tread temperature is in the normal temperature interval (i.e., the preset temperature interval); the (second) image information is image information in a display screen, and the (second) light information may be an alarm lamp, for example, when the pixel value of one pixel point is detected to be lower than a preset temperature threshold value, the highest temperature of the current tread is displayed in the display screen, and the color displayed by the alarm lamp is different from that of the alarm lamp in the alarm information, for example, the alarm lamp is green.

Fig. 5 is a block diagram of an apparatus for detecting a temperature of a tire surface according to an embodiment of the present application, as shown in fig. 5, the apparatus includes: an acquisition module 50, configured to acquire image information, where the image information includes: a first optical image of a target object and a first thermal imaging image of the target object, the target object being a vehicle to which a plurality of tires belong; a first determining module 52, configured to analyze the first optical image using the detection model, and identify position information of each tire, where the position information of each tire includes: the system comprises a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in a first optical image as a coordinate origin; the segmentation module 54 is configured to perform segmentation processing on the first thermal imaging image according to the position information of each tire to obtain a plurality of second thermal imaging images, where each second thermal imaging image only shows one tire; a second determination module 56 for determining a surface temperature of each tire from each second thermographic image.

Fig. 6 is a flowchart of the operation of the apparatus for detecting the tire surface temperature, and as shown in fig. 6, the apparatus for detecting the tire surface temperature starts to operate, and a (first) optical image and a (first) infrared thermal imaging image of the mining vehicle (i.e., the target object) on which the tire is mounted are acquired by the acquisition module 50 at the detection area. The first determination module 52 invokes the detection model to identify tires in the (first) optical image of the mine car and if the detection model does not identify tires in the (first) optical image of the mine car, continues to detect other acquired optical images of the mine car. If the detection model identifies a tire in the (first) optical image of the mining vehicle, further locating the position information of the tire; the segmentation module 54 further processes the coordinate information (i.e. the position information of the tire) output by the detection model, maps the coordinate information into a (first) thermal imaging image shot at the same time as the (first) optical image, and cuts the (second) thermal imaging image containing only one tire according to the mapping result; the second determining module 56 cuts a (third) thermal imaging image containing only the tread portion of the tire from the (second) thermal imaging image, and traverses the pixel value of each pixel in the (third) thermal imaging image, determining the pixel value of each pixel as the temperature value of the tread region corresponding to each pixel; meanwhile, comparing each temperature value in the temperature field (namely, the third thermal imaging image) with a set temperature threshold (namely, a preset temperature threshold) in traversing, and judging whether the temperature of the temperature field exceeds the set threshold; and if the temperature of the temperature field exceeds the set threshold value, sending out alarm information, wherein the alarm information comprises: the warning lamp is displayed as a red lamp (first light information), the display screen displays the highest temperature of the tread (first image information) and the buzzer alarms (first sound information). If the temperature of the temperature field does not exceed the set threshold value, only prompt information is sent, wherein the prompt information comprises: the warning lamp is displayed as a green light (second lamplight information), and the display screen is displayed at the current highest tread temperature (second image information).

It should be noted that, the preferred implementation manner of the embodiment shown in fig. 5 may refer to the related description of the embodiment shown in fig. 2, which is not repeated herein.

The embodiment of the application also provides a nonvolatile storage medium, wherein the nonvolatile storage medium stores a computer program, and the equipment where the nonvolatile storage medium is located executes the method for detecting the tire surface temperature by running the computer program.

The above-described nonvolatile storage medium is used to store a program that performs the following functions: acquiring image information, wherein the image information comprises: a first optical image of a target object and a first thermal imaging image of the target object, the target object being a vehicle to which a plurality of tires belong; analyzing the first optical image by using a detection model to identify the position information of each tire, wherein the position information of each tire comprises: the system comprises a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in a first optical image as a coordinate origin; dividing the first thermal imaging image according to the position information of each tire to obtain a plurality of second thermal imaging images, wherein each second thermal imaging image only displays one tire; the surface temperature of each tire is determined from each second thermographic image.

The embodiment of the application also provides an electronic device comprising a memory in which a computer program is stored and a processor arranged to perform the above method of detecting the temperature of a tyre surface by means of the computer program.

The processor in the electronic device is configured to execute a program that performs the following functions: acquiring image information, wherein the image information comprises: a first optical image of a target object and a first thermal imaging image of the target object, the target object being a vehicle to which a plurality of tires belong; analyzing the first optical image by using a detection model to identify the position information of each tire, wherein the position information of each tire comprises: the system comprises a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in a first optical image as a coordinate origin; dividing the first thermal imaging image according to the position information of each tire to obtain a plurality of second thermal imaging images, wherein each second thermal imaging image only displays one tire; the surface temperature of each tire is determined from each second thermographic image.

The respective modules in the above-described apparatus for detecting the tire surface temperature may be program modules (for example, a set of program instructions for realizing a specific function), or may be hardware modules, and the latter may be represented by the following forms, but are not limited thereto: the expression forms of the modules are all a processor, or the functions of the modules are realized by one processor.

The foregoing embodiment numbers of the present application are merely for describing, and do not represent advantages or disadvantages of the embodiments.

In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.

In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, for example, may be a logic function division, and may be implemented in another manner, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.

The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be essentially or a part contributing to the related art or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium, including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.

The foregoing is merely a preferred embodiment of the present application and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present application and are intended to be comprehended within the scope of the present application.

Claims

1. A method of detecting the temperature of a tire surface, comprising:

acquiring image information, wherein the image information comprises: a first optical image of a target object and a first thermal imaging image of the target object, wherein the target object is a vehicle to which a plurality of tires belong;

analyzing the first optical image by adopting a detection model to identify the position information of each tire, wherein the position information of each tire comprises: the tire comprises a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in the first optical image as a coordinate origin;

dividing the first thermal imaging image according to the position information of each tire to obtain a plurality of second thermal imaging images, wherein each second thermal imaging image only displays one tire;

Determining a surface temperature of each of the tires from each of the second thermographic images.

2. The method of claim 1, wherein analyzing the first optical image using a detection model to identify positional information for each of the tires comprises:

detecting the first optical image through the detection model to obtain a detection result, wherein the detection result is used for indicating whether a tire image exists in the first optical image;

generating a plurality of block diagrams in the case that the detection result indicates that the tire image exists in the first optical image, wherein each block diagram only contains one tire;

determining a confidence value of each block diagram, and determining a block diagram with the confidence value larger than a preset threshold value as a target frame, wherein the confidence value is used for indicating the accuracy of a prediction result output by the detection model;

and determining the abscissa of the center point of each target frame, the ordinate of the center point of each target frame, the width value of each target frame and the height value of each target frame as the position information of the tire corresponding to each target frame.

3. The method according to claim 1, wherein the detection model is trained by:

acquiring a dataset, wherein the dataset is generated from a set of optical images of the target object, the dataset comprising: training data, verification data and test data, wherein the proportion of the training data, the verification data and the test data is a preset proportion, and the optical image set comprises: the target object comprises a historical optical image of the target object, a real-time optical image of the target object and a target optical image obtained by carrying out conversion processing on the historical optical image and/or the real-time optical image;

and determining a loss function according to the data set, and training a machine learning model based on the data set and the loss function to obtain the detection model.

4. A method according to claim 3, wherein determining a loss function from the dataset comprises:

and taking the training data as input information of the machine learning model to obtain predicted position information of a target frame corresponding to each tire output by the machine learning model, wherein the predicted position information comprises: a predicted abscissa, a predicted ordinate, a predicted height value, and a predicted width value of each of the target frames;

Acquiring actual position information of each target frame, wherein the actual position information is stored in the verification data, and the actual position information comprises: the actual abscissa and the actual ordinate of each target frame;

determining the loss function according to the association relation among the predicted position information, the actual position information and the test data, wherein the test data comprises: a plurality of preset parameters stored in a configuration file of the machine learning model.

5. The method of claim 4, wherein the predicted location information is obtained by:

performing feature transformation on each second optical image in the training data to obtain a first feature map and a second feature map, wherein the first feature map is obtained by performing the feature transformation on each second optical image in the up-down direction, and the second feature map is obtained by performing the feature transformation on each second optical image in the left-right direction;

determining a first weight feature vector of the second optical image in the up-down direction and a second weight feature vector of the second optical image in the left-right direction according to the first feature map and the second feature map;

And determining the predicted position information according to the second optical image, the first weight feature vector and the second weight feature vector.

6. The method of claim 1, wherein determining the surface temperature of each of the tires from each of the second thermographic images comprises:

cutting each second thermal imaging image to obtain third thermal imaging images, wherein the area of each third thermal imaging image is smaller than that of the corresponding second thermal imaging image, and each third thermal imaging image does not contain an image of a hub;

and determining a pixel value of each pixel point in a plurality of pixel points of each third thermal imaging image, wherein the pixel value of each pixel point is the surface temperature of the tire area corresponding to each pixel point.

7. The method of claim 6, wherein the method further comprises:

comparing a plurality of pixel values of the pixel points with a preset temperature threshold value respectively to obtain a comparison result;

outputting alarm information when the comparison result indicates that a target pixel value larger than the preset temperature threshold exists in the pixel values, wherein the alarm information is used for prompting a user that the surface temperature of the tire is higher than the preset temperature threshold, and comprises first lamplight information, first image information and sound information;

And outputting prompt information when the comparison result indicates that a target pixel value which is larger than the preset temperature threshold value does not exist in the pixel values, wherein the prompt information is used for prompting the user that the surface temperature of the tire is in a preset temperature interval, and the prompt information comprises second lamplight information and second image information.

8. A system for detecting the temperature of a tire surface, the system comprising: an image acquisition module, an information processing module and a temperature early warning module, wherein,

the image acquisition module is used for acquiring image information, wherein the image acquisition module comprises: the camera system comprises a first camera system and a second camera system which is in parallel relation with the first camera system, wherein the included angles between the shooting directions of the first camera system and the second camera system and the horizontal direction are first preset angles, the included angles between the connecting line directions of the first camera system and the second camera system and the horizontal direction are second preset angles, and the first camera system comprises: a first camera and a second camera in a position adjacent to the first camera, the second camera system comprising: a third camera and a fourth camera positioned adjacent to the third camera, wherein the first camera and the third camera are used for acquiring a first optical image of a target object, the second camera and the fourth camera are used for acquiring a first thermal imaging image of the target object, and the target object is a vehicle to which a plurality of tires belong;

The information processing module is connected with the image acquisition module and comprises: a detection model for analyzing the first optical image to identify positional information of each of the tires, wherein the positional information of each of the tires includes: the tire comprises a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in the first optical image as a coordinate origin;

the information processing module is further used for dividing the first thermal imaging image according to the position information of each tire to obtain a plurality of second thermal imaging images, wherein each second thermal imaging image only displays one tire;

the temperature early warning module is used for determining the surface temperature of each tire according to each second thermal imaging image.

9. The system for detecting the surface temperature of a tire according to claim 8, wherein the temperature pre-warning module is configured to perform a cutting process on each of the second thermal imaging images to obtain third thermal imaging images, where an area of each of the third thermal imaging images is smaller than an area of a corresponding second thermal imaging image of each of the third thermal imaging images, and each of the third thermal imaging images does not include an image of a hub;

10. The system for detecting the surface temperature of a tire according to claim 9, wherein the temperature pre-warning module is further configured to compare a plurality of pixel values of the plurality of pixel points with a preset temperature threshold value, respectively, to obtain a comparison result;

11. An apparatus for detecting the temperature of a tire surface, comprising:

the device comprises an acquisition module for acquiring image information, wherein the image information comprises: a first optical image of a target object and a first thermal imaging image of the target object, wherein the target object is a vehicle to which a plurality of tires belong;

the first determining module is configured to analyze the first optical image with a detection model, and identify position information of each tire, where the position information of each tire includes: the tire comprises a plurality of coordinates of each tire in a target coordinate system, wherein the target coordinate system is established by taking a preset pixel point in the first optical image as a coordinate origin;

the segmentation module is used for carrying out segmentation processing on the first thermal imaging images according to the position information of each tire to obtain a plurality of second thermal imaging images, wherein each second thermal imaging image only displays one tire;

and a second determining module for determining a surface temperature of each tire from each of the second thermal imaging images.

12. A non-volatile storage medium, wherein a computer program is stored in the non-volatile storage medium, and wherein the method of detecting the tire surface temperature of any one of claims 1 to 7 is performed by running the computer program on a device in which the non-volatile storage medium is located.

13. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, the processor being arranged to perform the method of detecting the temperature of a tyre surface as claimed in any one of claims 1 to 7 by means of the computer program.