Rearview mirror noise testing system, method and industrial control computer

Abstract

This application relates to the field of rearview mirror testing technology, and particularly to a rearview mirror noise testing system, method, and industrial control computer. The system includes: a rearview mirror stand clamping device for mounting the rearview mirror under test; a data acquisition device for acquiring sound signals emitted by the rearview mirror under test when performing a target action; and an industrial control computer for controlling the rearview mirror under test to perform the target action according to the user's target operation instructions, and inputting the sound signals into a pre-trained noise recognition model, which outputs the noise test results of the rearview mirror under test. This solves the problems of complex noise testing procedures and low accuracy of noise judgment results in related technologies.

Description

Technical Field

[0001] This application relates to the field of rearview mirror testing technology, and in particular to a rearview mirror noise testing system, method and industrial control computer. Background Technology

[0002] With the rapid development of the automotive industry and the increasing demands of consumers for driving experience, the stability of automotive exterior rearview mirrors, as commonly used components, as well as the accuracy of abnormal noise detection and the efficiency of diagnosis, have attracted much attention from automotive OEMs and rearview mirror manufacturers.

[0003] Most of the rearview mirror noise testing methods in related technologies are designed in a decentralized manner. The clamping, signal acquisition, and fault diagnosis processes require independent equipment to be completed. They lack an integrated system architecture, have cumbersome operation procedures, and involve a lot of human intervention, which greatly reduces the efficiency of noise testing. At the same time, they are highly dependent on the professional level of the personnel. Fault diagnosis often relies on manual listening or simple signal analysis, which is easily affected by the subjective judgment of the testers, resulting in a low accuracy rate of noise diagnosis. Summary of the Invention

[0004] This application provides a rearview mirror noise testing system, method, and industrial control computer to solve the problems of complex operation procedures and low accuracy of noise judgment results in related technologies.

[0005] The first aspect of this application provides a rearview mirror noise testing system, including: a rearview mirror stand clamping device for mounting the rearview mirror to be tested; a data acquisition device for acquiring sound signals emitted by the rearview mirror to be tested when performing a target action; and an industrial control computer for controlling the rearview mirror to be tested to perform the target action according to the user's target operation instructions, and inputting the sound signals into a pre-trained noise recognition model, wherein the noise recognition model outputs the noise test results of the rearview mirror to be tested.

[0006] Optionally, the data acquisition device includes a microphone, a data cable, and a data acquisition card. The microphone is used to acquire the physical sound signal emitted by the rearview mirror under test when it performs the target action, and converts the physical sound signal into an analog electrical signal. The data cable is used to transmit the analog electrical signal to the data acquisition card. The data acquisition card is used to convert the analog electrical signal into a digital sound signal.

[0007] Optionally, the industrial control computer includes: a data input module, a data processing module, a prediction module, and a result output module, wherein the data input module is used to receive digital audio signals; the data processing module is used to preprocess the digital audio signals, wherein the preprocessing includes at least one of noise reduction processing, short-time Fourier transform, and slicing processing; the prediction module is used to input the preprocessed digital audio signals into a pre-trained abnormal noise recognition model, and the abnormal noise recognition model outputs the abnormal noise test results of the rearview mirror under test; the result output module is used to output at least one of the abnormal noise result, abnormal noise type, and abnormal noise level of the rearview mirror under test based on the abnormal noise test results.

[0008] Optionally, the industrial computer is also equipped with multiple operation buttons, which are available for user operation.

[0009] Optionally, the rearview mirror mounting device includes: a rearview mirror mounting device body; a rearview mirror mounting base disposed on the device body for mounting the rearview mirror to be tested; and a power supply device disposed on the device body for supplying power to the rearview mirror to be tested.

[0010] Optionally, the normal distance between the microphone and the center of the lens of the rearview mirror under test is the target distance.

[0011] Optionally, the preprocessed digital audio signal includes a two-dimensional time-frequency diagram and a time-series feature sequence. The abnormal noise recognition model includes an input layer, a first feature extraction layer, a second feature extraction layer, a feature fusion layer, and an output layer. The input layer is used to input the two-dimensional time-frequency diagram and the time-series feature sequence. The first feature extraction layer is used to extract frequency domain feature vectors from the two-dimensional time-frequency diagram. The second feature extraction layer is used to extract time-series feature vectors from the time-series feature sequence. The feature fusion layer is used to concatenate the frequency domain feature vectors and the time-series feature vectors to obtain a global feature vector. The output layer is used to output the abnormal noise test results of the rearview mirror under test based on the global feature vector.

[0012] Optionally, the training process of the abnormal noise recognition model includes: acquiring a training dataset, wherein the training dataset includes multiple sets of sound signals from rearview mirrors and corresponding abnormal noise test results; preprocessing the sound signals and enhancing the preprocessed signals, wherein the enhancement processing includes at least one of adding random noise, changing the volume gain, adjusting the volume, and time offset; and training the abnormal noise recognition model using the enhanced sound signals and corresponding abnormal noise test results.

[0013] A second aspect of this application provides a method for testing abnormal noises from a rearview mirror, implemented based on the rearview mirror abnormal noise testing system described in the above embodiments. The method includes the following steps: acquiring a user's target operation command; controlling the rearview mirror under test to perform a target action according to the target operation command, and acquiring the sound signal emitted by the rearview mirror under test when performing the target action; inputting the sound signal into a pre-trained abnormal noise recognition model, and the abnormal noise recognition model outputting the abnormal noise test result of the rearview mirror under test.

[0014] A third aspect of this application provides an industrial control computer, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the program to perform the rearview mirror noise testing method as described in the above embodiments.

[0015] A fourth aspect of this application provides a computer-readable storage medium having a computer program or instructions stored thereon, which is executed by a processor to perform the rearview mirror noise testing method as described above.

[0016] Therefore, this application has at least the following beneficial effects:

[0017] This application provides a rearview mirror noise testing system, comprising a rearview mirror stand clamping device, a data acquisition device, and an industrial control computer. The rearview mirror stand clamping device is used to mount and fix the rearview mirror under test. The data acquisition device collects sound signals generated during the rearview mirror's target action. The industrial control computer receives user operation commands and controls the rearview mirror to perform the target action, inputting the collected sound signals into a pre-trained noise recognition model. The model outputs the final noise test result, forming an integrated detection architecture of clamping, data acquisition, and judgment. This system requires no additional auxiliary equipment, is simple to operate, and the industrial control computer handles both action execution and noise judgment, reducing human intervention and improving testing efficiency. Furthermore, the pre-trained noise recognition model enables automatic noise judgment, replacing subjective human judgment and improving the accuracy of noise detection. Therefore, this system solves the technical problems of complex noise testing procedures and low accuracy of noise judgment results in related technologies.

[0018] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0019] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a schematic diagram of a rearview mirror noise testing system provided according to an embodiment of this application; Figure 2 This is a schematic diagram of the components of an industrial control computer provided according to an embodiment of this application; Figure 3 This is a schematic diagram illustrating the industrial control computer display results provided according to the embodiments of this application; Figure 4 This is a schematic diagram of the data acquisition device and industrial control computer provided according to the embodiments of this application; Figure 5 This is a schematic diagram of a rearview mirror mounting device according to an embodiment of this application; Figure 6 This is a schematic diagram illustrating the specific components of the rearview mirror noise testing system provided according to an embodiment of this application; Figure 7 This is a flowchart of a rearview mirror noise testing method provided according to an embodiment of this application; Figure 8 This is a schematic diagram of the structure of an industrial control computer provided according to an embodiment of this application. Detailed Implementation

[0020] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0021] The following description, with reference to the accompanying drawings, outlines a rearview mirror noise testing system, method, and industrial control computer according to embodiments of this application. Addressing the issues of complex testing procedures and low accuracy in noise assessment in related technologies mentioned in the background section, this application provides a rearview mirror noise testing system. This system forms an integrated and complete solution encompassing rearview mirror installation, testing, evaluation, and automatic assessment. It does not rely on the professional level of evaluators or additional equipment and facilities, while offering high testing and evaluation efficiency with minimal human intervention. Therefore, it solves the problems of complex testing procedures and low accuracy in noise assessment in related technologies.

[0022] Specifically, Figure 1 This is a schematic diagram of a rearview mirror noise testing system provided in an embodiment of this application.

[0023] like Figure 1 As shown, the rearview mirror noise testing system 10 includes: a rearview mirror stand clamping device 101, a data acquisition device 102, and an industrial control computer 11.

[0024] The rearview mirror stand clamping device 101 is used to install the rearview mirror under test; the data acquisition device 102 is used to collect the sound signal emitted by the rearview mirror under test when performing the target action; the industrial control computer 11 is used to control the rearview mirror under test to perform the target action according to the user's target operation command, and input the sound signal into the pre-trained abnormal noise recognition model, and the abnormal noise recognition model outputs the abnormal noise test result of the rearview mirror under test.

[0025] It is understood that this application embodiment constructs a rearview mirror noise testing system 10, including a rearview mirror stand clamping device 101, a data acquisition device 102, and an industrial control computer 11. The rearview mirror stand clamping device 101 is used to install and fix the rearview mirror to be tested. The data acquisition device 102 is used to collect the sound signals generated during the rearview mirror's target action. The industrial control computer 11 receives user operation instructions and controls the rearview mirror to perform the target action. It also inputs the collected sound signals into a pre-trained noise recognition model, and the model outputs the final noise test result. This forms an integrated detection architecture of clamping, data acquisition, and judgment, which requires no additional auxiliary equipment, is simple to operate, and reduces human intervention to a certain extent by having the industrial control computer 11 control the execution of actions and determine noises. This improves testing efficiency. At the same time, the pre-trained noise recognition model is used to realize automatic noise judgment, replacing subjective human judgment, thereby improving the accuracy of noise detection.

[0026] The rearview mirror to be tested in this application embodiment is an electric exterior rearview mirror of a vehicle that needs to be tested for abnormal noises. It has electric functions such as folding and lens angle adjustment. The target action can be the folding of the rearview mirror, the up and down adjustment of the lens, the left and right adjustment, etc. The abnormal noise recognition model can be an AI model. The abnormal noise test results include whether there is an abnormal noise in the rearview mirror, the specific type of abnormal noise, and the degree of verification of the abnormal noise.

[0027] Furthermore, in one embodiment of this application, the data acquisition device 101 includes a microphone 303, a data cable 302, and a data acquisition card 304.

[0028] The microphone 303 is used to collect the physical sound signal emitted by the rearview mirror under test when it performs the target action, and convert the physical sound signal into an analog electrical signal; the data cable 302 is used to transmit the analog electrical signal to the data acquisition card; and the data acquisition card 304 is used to convert the analog electrical signal into a digital sound signal.

[0029] It is understood that the data acquisition device 20 in this application embodiment includes a microphone 303, a data cable 302, and a data acquisition card 304. The microphone 303 picks up the physical sound signal when the rearview mirror performs the target action and converts it into an analog electrical signal. The data cable 302 realizes the lossless transmission of the analog electrical signal from the microphone to the data acquisition card. The data acquisition card 304 completes the core analog-to-digital (A / D) conversion, converting the analog electrical signal into a digital sound signal that can be processed by the industrial control computer, so that the industrial control computer 11 can subsequently complete the abnormal noise identification result judgment based on the recognizable digital sound signal.

[0030] The microphone 303 in this embodiment is a microphone that has been calibrated and meets the test accuracy requirements; the microphone in this embodiment can also be replaced by a miniature acoustic array.

[0031] Furthermore, in one embodiment of this application, the industrial control computer 11 includes: a data input module 401, a data processing module 402, a prediction module 403, and a result output module 404.

[0032] The data input module 401 is used to receive digital audio signals; the data processing module 402 is used to preprocess the digital audio signals, wherein the preprocessing includes at least one of noise reduction processing, short-time Fourier transform and slicing processing; the prediction module 403 is used to input the preprocessed digital audio signals into a pre-trained abnormal noise recognition model, and the abnormal noise recognition model outputs the abnormal noise test results of the rearview mirror under test; the result output module 404 is used to output at least one of the abnormal noise results, abnormal noise type and abnormal noise level of the rearview mirror under test based on the abnormal noise test results.

[0033] It is understood that the industrial control computer 11 in this embodiment of the application includes a data input module 401, a data processing module 402, a prediction module 403, and a result output module 404. The data input module 401 receives digital audio signals transmitted by the data acquisition card to provide data sources for subsequent processing. The data processing module 402 preprocesses the digital audio signals, including at least one of noise reduction, short-time Fourier transform, and slicing. The prediction module 403 inputs the preprocessed digital audio signals into a pre-trained abnormal noise recognition model. The result output module 404 analyzes the test results output by the model to clearly determine the abnormal noise result, abnormal noise type, and abnormal noise level of the rearview mirror.

[0034] The abnormal noise levels in this application embodiment can be set to multiple levels depending on the specific situation, such as setting it to level 3, where 0.1 represents slight abnormal noise, 0.3 represents moderate abnormal noise, and 1 represents severe abnormal noise.

[0035] Specifically, the industrial control computer 11 in this embodiment can preprocess sound data, provide a user interface, and display results. Its specific components are as follows: Figure 2As shown, the system includes a data input module 401, a data processing module 402, a prediction module 403, and a result output module 404. The execution flow between the various models is as follows: Step S1: Receive the data that has undergone simple processing from the data acquisition system through the data input module 401; Step S2: The data preprocessing module 402 (i.e., the data processing module) first performs wavelet packet noise reduction signal processing and short-time Fourier transform and slicing processing on the data, and outputs it to the AI ​​prediction module 403 (prediction module). Step S3: The preprocessed data is transmitted to the AI ​​prediction module 403. The AI ​​prediction module is equipped with an abnormal noise recognition model. This model is trained on a large amount of rearview mirror abnormal noise test data and has a prediction accuracy of more than 95%. The prediction model has two core features: During model training, the model uses data enhancement methods such as adding random noise, changing volume gain, adjusting pitch, and time offset to improve the model's generalization ability and prediction accuracy; and it uses a CNN (Convolutional Neural Network) + LSTM (Long Short-Term Memory) dual-stream feature extraction network for feature extraction and fault identification.

[0036] Step S4: The result output module 404 outputs the fault determination result of the electric rearview mirror; Step S5: As Figure 3 As shown, the result output module 404 determines the result based on the abnormal noise identification model and displays whether there is an abnormal noise in the tested sample. If there is an abnormal noise, the type and severity of the abnormal noise are determined. The severity of the abnormal noise is divided into three levels: 0.1 indicates a slight abnormal noise, which needs to be improved but will not cause strong complaints from users; 0.3 indicates a moderate abnormal noise, which needs to be improved; and 1 indicates a severe abnormal noise, which will cause strong complaints from users and must be improved.

[0037] Furthermore, in one embodiment of this application, the industrial control computer 11 is also provided with multiple operation buttons, which are operated by the user.

[0038] It is understood that in this embodiment of the application, multiple operation buttons can be set on the industrial control computer 11 as the only operation entry point for the user. The button functions are bound to the target actions of the rearview mirror. The user can perform the operation of the rearview mirror by clicking the button, which simplifies the operation process and reduces the need for manual intervention to adjust the rearview mirror.

[0039] The operation buttons in this application embodiment can be virtual buttons or physical buttons.

[0040] Furthermore, it should be noted that the industrial control computer 11 and the aforementioned data acquisition card 304 in this embodiment of the application are both located in... Figure 4 In the data acquisition front-end 301 shown, the microphone 303, the data cable 302 and the data acquisition card 304 constitute the data acquisition device 20. The data acquisition card 303 is responsible for acquiring sound signals, and the industrial control computer 11 is responsible for preprocessing the sound data and providing a user interface and displaying the results.

[0041] In this embodiment, parameters such as sampling frequency and frequency response range can be set on the industrial control computer 11. The sampling frequency is recommended to be 48kHz and the frequency response range is recommended to be 20Hz-20kHz to ensure complete sound signal pickup. The corresponding buttons on the control computer can be pressed to perform operations such as rearview mirror folding and lens adjustment, and data acquisition and storage can be performed simultaneously.

[0042] Furthermore, in one embodiment of this application, the rearview mirror mounting device 101 includes: a rearview mirror mounting device body 201; a rearview mirror mounting base 202 disposed on the device body for mounting the rearview mirror to be tested; and a power supply device 203 disposed on the device body for supplying power to the rearview mirror to be tested.

[0043] Among them, the rearview mirror stand clamping body 201 can also be called the experimental stand basic frame; the rearview mirror mounting base 202 is a clamping mechanism with adjustable position.

[0044] It is understood that the rearview mirror stand clamping device 101 of this application consists of a rearview mirror stand clamping device body 201, a rearview mirror mounting base 202, and a power supply device 203. The device body 201 provides physical support and installation foundation for the entire stand. The rearview mirror mounting base 202 is set on the body and used to fix the rearview mirror under test. The power supply device 203 is set on the body and provides working power to the electric rearview mirror under test, ensuring that it can normally perform target actions such as folding and lens adjustment. This realizes the integrated design of stand clamping and power supply.

[0045] The rearview mirror mounting base 202 in this embodiment is an adjustable mounting base, which can be adapted to different models of rearview mirrors to improve the versatility of the rearview mirror stand clamping device and reduce repeated investment in equipment. In practical application scenarios, this embodiment can arrange multiple similar structures without mutual interference of sound, so as to achieve the purpose of detecting multiple rearview mirror noise faults at one time.

[0046] Specifically, the rearview mirror mounting device of this application embodiment is as follows: Figure 5 As shown, the test bench includes a basic frame 201 (i.e., the rearview mirror mounting body), a rearview mirror mounting base 202, and a power supply module 203 (i.e., the power supply equipment). The rearview mirror mounting base 202 has an adjustable clamping mechanism, which enables the rapid installation of the rearview mirror.

[0047] The bench clamping mechanism in this application embodiment can also use automated clamping components such as pneumatic grippers and electric adjustable guide rails to replace the manually adjustable clamping mechanism. As long as it can achieve rapid positioning and installation of the rearview mirror, ensure clamping stability, and not change the relative positional accuracy of the microphone and the specimen, the same clamping effect can be achieved.

[0048] Furthermore, in one embodiment of this application, the normal distance between the microphone 303 and the center of the lens of the rearview mirror to be tested is the target distance.

[0049] Among them, the normal distance between the microphone 303 and the center of the lens of the rearview mirror under test is the vertical distance between the microphone pickup end and the center of the lens of the rearview mirror under test; the target distance is the optimal normal distance, which can be 30cm. It can be finely adjusted, but it must be consistent with the distance when the abnormal noise recognition model is trained.

[0050] Furthermore, in one embodiment of this application, the preprocessed digital audio signal includes a two-dimensional time-frequency diagram and a time-series feature sequence. The abnormal noise recognition model includes an input layer, a first feature extraction layer, a second feature extraction layer, a feature fusion layer, and an output layer. The input layer is used to input the two-dimensional time-frequency diagram and the time-series feature sequence. The first feature extraction layer is used to extract a frequency domain feature vector from the two-dimensional time-frequency diagram. The second feature extraction layer is used to extract a time-series feature vector from the time-series feature sequence. The feature fusion layer is used to concatenate the frequency domain feature vector and the time-series feature vector to obtain a global feature vector. The output layer is used to output the abnormal noise test result of the rearview mirror under test based on the global feature vector.

[0051] The first feature extraction layer can be a CNN layer, and the second feature extraction layer can be an LSTM layer.

[0052] It is understood that the abnormal noise recognition model in this application adopts a dual-stream feature extraction architecture, which simultaneously captures the frequency domain and temporal features of the sound to solve the problem of missed fault features caused by single feature extraction, improve the accuracy of abnormal noise recognition of the model, and the feature fusion layer integrates the dual features to form a complete fault feature description, avoids the one-sidedness of single-dimensional features, and further improves the accuracy of model judgment.

[0053] The output layer of the abnormal noise recognition model in this application embodiment can be a classification and regression fusion output layer, which can realize multiple task outputs (classification judgment and regression judgment). Among them, the classification judgment includes the determination of the presence or absence of abnormal noise (binary classification) and the determination of the type of abnormal noise (multi-class classification). The determination of the presence or absence of abnormal noise can be output by a fully connected layer + sigmoid activation function, and the determination of the type of abnormal noise can be output by a fully connected layer + softmax activation function. The regression judgment can be output by a fully connected layer + linear activation function, which outputs a quantified severity value, thereby determining the level of abnormal noise.

[0054] Furthermore, it should be noted that the feature extraction network constructed by the first feature extraction layer and the second feature extraction layer in the embodiments of this application can also adopt a Transformer+CNN hybrid network, an improved residual network, or a bidirectional LSTM network. There is no specific limitation on this; the above is merely an example of a possible implementation.

[0055] Furthermore, in one embodiment of this application, the training process of the abnormal noise recognition model includes: acquiring a training dataset, wherein the training dataset includes multiple sets of sound signals from rearview mirrors and corresponding abnormal noise test results; preprocessing the sound signals and enhancing the preprocessed signals, wherein the enhancement processing includes at least one of adding random noise, changing the volume gain, adjusting the volume, and time offset; and training the abnormal noise recognition model using the enhanced sound signals and corresponding abnormal noise test results.

[0056] It is understood that the embodiments of this application can obtain a training dataset containing multiple sets of rearview mirror sound signals and corresponding abnormal noise test results. The sound signals in the training dataset are preprocessed in the same way as in the testing stage, and the preprocessed signals are enhanced. The effective training sample size is expanded through data augmentation, thereby improving the generalization ability and anti-interference ability of the model. Then, the enhanced sound signals are used as input, and the corresponding abnormal noise test results are used as labels to train the abnormal noise recognition model, thus obtaining the pre-trained model.

[0057] Specifically, the rearview mirror noise testing system of this application embodiment has the following specific components: Figure 6 As shown, the system includes: a rearview mirror stand clamping module 101 (i.e., a rearview mirror stand clamping device), which enables rapid clamping of the rearview mirror under test and provides power for rearview mirror adjustment; a data acquisition module 102 (i.e., a data acquisition device), which acquires, stores, and outputs the results to the next module; a machine learning-based intelligent fault identification system 103, which preprocesses the data and then transmits it to an AI intelligent identification system that has been trained with big data and whose accuracy meets the requirements. This system can automatically identify the presence, type, and severity of faults based on the characteristics of the input data; and a fault output module 104, which displays the fault status of the test piece on the industrial control computer interface and can also store relevant analysis results and raw data. It features high integration and intelligence, does not require professional evaluation personnel, does not rely on additional equipment and facilities, has high testing and evaluation efficiency, and requires minimal human intervention. It provides rearview mirror suppliers and automotive OEMs with a time-saving and labor-saving (manual) quality inspection and fault judgment system. Through the fully automated design of clamping-acquisition-identification-output, it achieves accurate detection and intelligent judgment of abnormal noises. Among them, the machine learning-based intelligent fault identification system 103 and the fault output module 104 are both components of the industrial control computer.

[0058] In summary, the rearview mirror noise testing system of this application embodiment can achieve the following: 1. An integrated, fully automated solution that innovatively integrates four core modules: rearview mirror stand clamping, data acquisition, intelligent noise recognition, and result output. This forms an integrated closed-loop system from test piece installation to fault diagnosis, eliminating the need for additional auxiliary equipment, significantly reducing human intervention, and solving the problems of traditional testing relying on multiple devices and cumbersome operations. 2. The highly adaptable bench clamping structure design features an adjustable mounting base that supports quick clamping and positioning of electric rearview mirrors. Multiple workstations can be arranged without mutual noise interference, enabling simultaneous testing of multiple specimens. The microphone is synchronously fixed to the clamping device, facilitating the adjustment of the normal distance between the microphone and the center of the rearview mirror lens under test, ensuring consistency of the testing environment and reliability of the data.

[0059] 3. An integrated approach to data acquisition, preprocessing, and analysis, with built-in data acquisition, storage, preprocessing, and analysis modules. Utilizing the collaborative work of the acquisition card and industrial control computer, it achieves a high degree of integration in the complete acquisition, preprocessing, and storage of sound signals. No professional background is required for testing and evaluation personnel, thus improving testing efficiency. 4. A machine learning recognition system with high generalization ability, which adopts the core technology of data augmentation + dual-stream network. By adding random noise, changing volume gain, adjusting pitch and time offset, the model's adaptability is improved. Based on CNN+LSTM dual-stream feature extraction network, fault feature mining and recognition are performed. After training on massive data, the model has a prediction accuracy of over 95% and can automatically determine the presence, type and severity of abnormal noises.

[0060] 5. A graded fault output and traceability mechanism is used to intuitively display fault results through the industrial control computer interface. The severity of abnormal noise is divided into three levels: slight (0.1), moderate (0.3), and severe (1), which clarifies the priority of improvement. At the same time, it supports the storage of original sound data, preprocessing results and fault judgment reports to meet the needs of quality traceability and data analysis.

[0061] The rearview mirror noise testing system proposed in this application includes a rearview mirror stand clamping device, a data acquisition device, and an industrial control computer. The rearview mirror stand clamping device is used to install and fix the rearview mirror under test. The data acquisition device is used to collect the sound signals generated during the rearview mirror's target action. The industrial control computer receives user operation instructions and controls the rearview mirror to perform the target action. It also inputs the collected sound signals into a pre-trained noise recognition model, which outputs the final noise test result. This forms an integrated detection architecture of clamping, data acquisition, and judgment. It requires no additional auxiliary equipment, is easy to operate, and the industrial control computer controls the execution of actions and the noise judgment, reducing human intervention to a certain extent and improving testing efficiency. At the same time, the pre-trained noise recognition model enables automatic noise judgment, replacing subjective human judgment and improving the accuracy of noise detection.

[0062] Next, the method for testing rearview mirror noise according to the embodiments of this application is described with reference to the accompanying drawings.

[0063] This method for testing rearview mirror noise is based on the rearview mirror noise testing system described above.

[0064] Figure 7 This is a flowchart of a rearview mirror noise testing method according to an embodiment of this application.

[0065] like Figure 7 As shown, the method for testing rearview mirror noise includes the following steps: In step S101, the user's target operation instruction is obtained.

[0066] The target operation command is the command to adjust the rearview mirror function generated by the user's operation of the operation button.

[0067] In step S102, the rearview mirror under test is controlled to perform the target action according to the target operation command, and the sound signal emitted by the rearview mirror under test when performing the target action is acquired.

[0068] In step S103, the sound signal is input to the pre-trained abnormal noise recognition model, and the abnormal noise recognition model outputs the abnormal noise test results of the rearview mirror under test.

[0069] It should be noted that the foregoing explanation of the rearview mirror noise testing system embodiment also applies to the rearview mirror noise testing method of this embodiment, and will not be repeated here.

[0070] Figure 8 A schematic diagram of the structure of an industrial control computer provided in an embodiment of this application. The industrial control computer may include: The memory 801, the processor 802, and the computer program stored on the memory 801 and capable of running on the processor 802.

[0071] When the processor 802 executes the program, it implements the rearview mirror noise test method provided in the above embodiments.

[0072] Furthermore, industrial control computers also include: Communication interface 803 is used for communication between memory 801 and processor 802.

[0073] The memory 801 is used to store computer programs that can run on the processor 802.

[0074] The memory 801 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0075] If the memory 801, processor 802, and communication interface 803 are implemented independently, then the communication interface 803, memory 801, and processor 802 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized into address buses, data buses, control buses, etc. For ease of representation, Figure 8 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0076] Optionally, in a specific implementation, if the memory 801, processor 802, and communication interface 803 are integrated on a single chip, then the memory 801, processor 802, and communication interface 803 can communicate with each other through an internal interface.

[0077] The processor 802 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0078] This application also provides a computer-readable storage medium storing a computer program or instructions thereon, which, when executed by a processor, implements the above-described method for testing rearview mirror noise.

[0079] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0080] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0081] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0082] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (FPGAs), field-programmable gate arrays (FPGAs), etc.

[0083] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

Claims

1. A rearview mirror noise testing system, characterized in that, include: A rearview mirror mounting device for mounting the rearview mirror to be tested; The data acquisition device is used to acquire the sound signal emitted by the rearview mirror under test when it performs the target action; An industrial control computer is used to control the rearview mirror under test to perform target actions according to the user's target operation instructions, and input the sound signal into a pre-trained abnormal noise recognition model. The abnormal noise recognition model outputs the abnormal noise test results of the rearview mirror under test.

2. The rearview mirror noise testing system according to claim 1, characterized in that, The data acquisition device includes a microphone, a data cable, and a data acquisition card, wherein... The microphone is used to collect the physical sound signal emitted by the rearview mirror under test when it performs the target action, and convert the physical sound signal into an analog electrical signal. The data cable is used to transmit the analog electrical signal to the data acquisition card; The data acquisition card is used to convert the analog electrical signal into a digital sound signal.

3. The rearview mirror noise testing system according to claim 2, characterized in that, The industrial control computer includes: a data input module, a data processing module, a prediction module, and a result output module, wherein... The data input module is used to receive the digital audio signal; A data processing module is used to preprocess the digital audio signal, wherein the preprocessing includes at least one of noise reduction processing, short-time Fourier transform, and slicing processing; The prediction module is used to input the preprocessed digital sound signal into a pre-trained abnormal noise recognition model, and the abnormal noise recognition model outputs the abnormal noise test result of the rearview mirror under test. The result output module is used to output at least one of the following based on the abnormal noise test results: abnormal noise result, abnormal noise type, and abnormal noise level of the rearview mirror under test.

4. The rearview mirror noise testing system according to claim 1, characterized in that, The industrial computer is also equipped with multiple operation buttons, which are operated by the user.

5. The rearview mirror noise testing system according to claim 1, characterized in that, The rearview mirror stand clamping device includes: The main body of the rearview mirror mounting bracket; A rearview mirror mounting bracket is provided on the main body of the device for mounting the rearview mirror to be tested. The power supply device installed on the main body of the device is used to supply power to the rearview mirror under test.

6. The rearview mirror noise testing system according to claim 2, characterized in that, The target distance is the normal distance between the microphone and the center of the lens of the rearview mirror under test.

7. The rearview mirror noise testing system according to claim 3, characterized in that, The preprocessed digital audio signal includes a two-dimensional time-frequency graph and a temporal feature sequence. The abnormal noise recognition model includes an input layer, a first feature extraction layer, a second feature extraction layer, a feature fusion layer, and an output layer. The input layer is used to input the two-dimensional time-frequency graph and the time-series feature sequence; The first feature extraction layer is used to extract frequency domain feature vectors from the two-dimensional time-frequency graph; The second feature extraction layer is used to extract temporal feature vectors from the temporal feature sequence; The feature fusion layer is used to concatenate the frequency domain feature vector and the time-series feature vector to obtain a global feature vector; The output layer is used to output the abnormal noise test results of the rearview mirror under test based on the global feature vector.

8. The rearview mirror noise testing system according to claim 1, characterized in that, The training process of the abnormal noise recognition model includes: Obtain a training dataset, wherein the training dataset includes multiple sets of sound signals from rearview mirrors and corresponding abnormal noise test results; The audio signal is preprocessed, and the preprocessed signal is enhanced, wherein the enhancement process includes at least one of adding random noise, changing the volume gain, adjusting the volume, and time offset. The abnormal noise recognition model is trained using the enhanced sound signal and the corresponding abnormal noise test results.

9. A method for testing abnormal noises from a rearview mirror, characterized in that, The method is implemented based on the rearview mirror noise testing system as described in any one of claims 1-8, wherein the method includes the following steps: Obtain the user's target operation command; The system controls the rearview mirror under test to perform the target action according to the target operation command, and acquires the sound signal emitted by the rearview mirror under test when performing the target action; The sound signal is input into a pre-trained abnormal noise recognition model, which outputs the abnormal noise test results of the rearview mirror under test.

10. An industrial control computer, characterized in that, include: The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the rearview mirror noise testing method as described in claim 9.

Images