Tire single-side wear detection method and vehicle

Abstract

The invention relates to a tire single-side wear detection method and a vehicle, and is applied to the field of vehicle driving technology application. The tire single-side wear detection method comprises the following steps: acquiring an audio signal in the running process of a vehicle, and filtering the audio signal to obtain a target audio signal; further, performing feature extraction on the target audio signal to obtain an energy feature value of a preset feature frequency band; and detecting based on the energy characteristic value to obtain a detection result of single-side wear of the tire, and outputting early warning information. Therefore, through a complete automatic detection chain of acquisition, filtering, extraction, detection and output, tire detection is converted from regular and passive detection to real-time and active detection, early warning can be sent to a driver before the abrasion problem is worsened and safety accidents are caused, and serious accidents such as tire burst and out-of-control caused by serious abrasion of tires are effectively prevented.

Description

Technical Field

[0001] This disclosure relates to the field of vehicle driving technology applications, and in particular to a method for detecting tire wear on one side and a vehicle thereof. Background Technology

[0002] Tires are a key component that ensures vehicle driving safety. Unilateral wear (commonly known as "tire wear") is a common problem that can lead to reduced grip, increased driving noise, and even tire blowout accidents.

[0003] Currently, tire wear detection mainly relies on regular manual inspections, tire pressure monitoring systems (TPMS), or regular four-wheel alignments. Manual inspections are inefficient, cannot achieve real-time monitoring, and are prone to missing early wear; TPMS can only monitor tire pressure and temperature and cannot identify unilateral wear problems caused by misalignment of the four wheels; regular four-wheel alignments are preventative maintenance and cannot detect dynamic changes in tire wear in real time.

[0004] The methods mentioned above lack real-time capability, automation, and early warning capabilities, resulting in tire wear problems not being detected in a timely manner, thus increasing driving safety risks. Summary of the Invention

[0005] To address the aforementioned technical problems, this disclosure provides a method for detecting tire wear on one side and a vehicle.

[0006] The first aspect of this disclosure provides a method for detecting tire wear on one side, including: Collect audio signals while the vehicle is in motion; The audio signal is filtered to obtain a target audio signal, which is an audio signal related to tire wear. Feature extraction is performed on the target audio signal to obtain the energy feature value of a preset feature frequency band, which is used to analyze the wear condition of one side of the tire. Based on the energy characteristic value, the detection result of tire wear on one side is obtained, and a warning message is output.

[0007] In some embodiments of this disclosure, the filtering process of the audio signal to obtain the target audio signal includes: The audio signal is preprocessed to obtain a preprocessed audio signal; The preprocessed audio signal is subjected to high-pass filtering to obtain the first audio signal; The first audio signal is denoised using spectral subtraction to obtain the second audio signal. The second audio signal is subjected to bandpass filtering to obtain the target audio signal.

[0008] In some embodiments of this disclosure, the step of extracting features from the target audio signal to obtain energy feature values ​​of a preset feature frequency band includes: The target audio signal is transformed in the frequency domain to obtain a frequency domain signal; The average amplitude of each frequency component within a preset characteristic frequency band is calculated based on the frequency domain signal, and the average amplitude is determined as the energy characteristic value.

[0009] In some embodiments of this disclosure, before obtaining the detection result of tire unilateral wear based on the energy characteristic value, the method further includes: Determine whether the vehicle is currently in the target scene, where the target scene is a signal-to-noise ratio enhanced scene.

[0010] In some embodiments of this disclosure, the detection based on the energy characteristic value to obtain the detection result of tire unilateral wear includes: The energy characteristic values ​​are aggregated to obtain aggregated characteristic values; When the vehicle is currently in the target scenario, the aggregated feature value is compared with a preset first threshold. If the aggregated feature value is greater than or equal to the preset first threshold, it is determined that unilateral tire wear has occurred; or, If the vehicle is not currently in the target scenario, the aggregated feature value is compared with a preset second threshold. If the aggregated feature value is greater than or equal to the preset second threshold, it is determined that tire wear has occurred on one side. Wherein, the preset first threshold is less than the preset second threshold.

[0011] In some embodiments of this disclosure, determining whether the vehicle is currently in the target scenario includes: Scene recognition is performed based on the current image data collected by the vehicle to obtain a first recognition result and a first confidence level; or Based on the vehicle's current location information and a high-precision map, scene matching is performed to obtain a second recognition result and a second confidence level.

[0012] In some embodiments of this disclosure, determining whether the vehicle is currently in the target scenario includes: Obtain the first identification result, the first confidence level, the second identification result, and the second confidence level; Based on the current external environment information, the weights of the first identification result and the second identification result are dynamically allocated; A comprehensive confidence level is obtained by weighting and fusing the first confidence level, the second confidence level, and their respective assigned weights. If the overall confidence level exceeds a preset threshold, it is determined that the current location is the target scene.

[0013] In some embodiments of this disclosure, the output of warning information includes: Within a preset time window, determine the frequency and duration of events that cause unilateral tire wear. Based on the frequency and duration of the events, the severity level of tire wear on one side is determined, and a corresponding warning message is generated and output based on the severity level.

[0014] In some embodiments of this disclosure, generating and outputting warning information corresponding to the severity level includes: When the severity level is Level 1, a tire check reminder icon will be displayed on the dashboard; When the severity level is Level 2, a voice prompt will be output, and tire inspection suggestions will be displayed on the central control screen; When the severity level is Level 3, a hazard warning signal is output, and a speed limit is recommended.

[0015] A second aspect of this disclosure provides a tire unilateral wear detection device, comprising: The acquisition module is used to acquire audio signals while the vehicle is in motion; The module is used to filter the audio signal to obtain a target audio signal, wherein the target audio signal is an audio signal related to tire wear. The extraction module is used to extract features from the target audio signal to obtain energy feature values ​​of a preset feature frequency band, which is used to analyze the wear condition of one side of the tire. The detection module is used to perform detection based on the energy characteristic value, obtain the detection result of tire wear on one side, and output warning information.

[0016] In some embodiments of this disclosure, when the obtaining module filters the audio signal to obtain the target audio signal, it is specifically used for: The audio signal is preprocessed to obtain a preprocessed audio signal; The preprocessed audio signal is subjected to high-pass filtering to obtain the first audio signal; The first audio signal is denoised using spectral subtraction to obtain the second audio signal. The second audio signal is subjected to bandpass filtering to obtain the target audio signal.

[0017] In some embodiments of this disclosure, when the extraction module extracts features from the target audio signal to obtain energy feature values ​​of a preset feature frequency band, it is specifically used for: The target audio signal is transformed in the frequency domain to obtain a frequency domain signal; The average amplitude of each frequency component within a preset characteristic frequency band is calculated based on the frequency domain signal, and the average amplitude is determined as the energy characteristic value.

[0018] In some embodiments of this disclosure, before obtaining the detection result of tire unilateral wear based on the energy characteristic value, the device further includes: The judgment module is used to determine whether the vehicle is currently in the target scene, which is a signal-to-noise ratio enhanced scene.

[0019] In some embodiments of this disclosure, when the detection module performs detection based on the energy characteristic value to obtain the detection result of tire wear on one side, it is specifically used for: The energy characteristic values ​​are aggregated to obtain aggregated characteristic values; When the vehicle is currently in the target scenario, the aggregated feature value is compared with a preset first threshold. If the aggregated feature value is greater than or equal to the preset first threshold, it is determined that unilateral tire wear has occurred; or, If the vehicle is not currently in the target scenario, the aggregated feature value is compared with a preset second threshold. If the aggregated feature value is greater than or equal to the preset second threshold, it is determined that tire wear has occurred on one side. Wherein, the preset first threshold is less than the preset second threshold.

[0020] In some embodiments of this disclosure, when the determination module determines whether the vehicle is currently in the target scene, it is specifically used for: Scene recognition is performed based on the current image data collected by the vehicle to obtain a first recognition result and a first confidence level; or Based on the vehicle's current location information and a high-precision map, scene matching is performed to obtain a second recognition result and a second confidence level.

[0021] In some embodiments of this disclosure, when the determination module determines whether the vehicle is currently in the target scene, it is specifically used for: Obtain the first identification result, the first confidence level, the second identification result, and the second confidence level; Based on the current external environment information, the weights of the first identification result and the second identification result are dynamically allocated; A comprehensive confidence level is obtained by weighting and fusing the first confidence level, the second confidence level, and their respective assigned weights. If the overall confidence level exceeds a preset threshold, it is determined that the current location is the target scene.

[0022] In some embodiments of this disclosure, when the detection module outputs warning information, it is specifically used for: Within a preset time window, determine the frequency and duration of events that cause unilateral tire wear. Based on the frequency and duration of the events, the severity level of tire wear on one side is determined, and a corresponding warning message is generated and output based on the severity level.

[0023] In some embodiments of this disclosure, when the detection module generates and outputs warning information corresponding to the severity level, it is specifically used for: When the severity level is Level 1, a tire check reminder icon will be displayed on the dashboard; When the severity level is Level 2, a voice prompt will be output, and tire inspection suggestions will be displayed on the central control screen; When the severity level is Level 3, a hazard warning signal is output, and a speed limit is recommended.

[0024] A third aspect of this disclosure provides an electronic device, including: processor; Memory, used to store executable instructions; The processor is used to read executable instructions from memory and execute the executable instructions to implement the tire unilateral wear detection method provided in the first aspect above.

[0025] A fourth aspect of this disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the tire unilateral wear detection method provided in the first aspect.

[0026] A fifth aspect of this disclosure provides a computer program product comprising a computer program or instructions that, when executed by a processor, implement the tire unilateral wear detection method of the first aspect described above.

[0027] A sixth aspect of this disclosure provides a vehicle that includes electronic equipment provided in the third aspect.

[0028] The technical solution provided in this disclosure has the following advantages: The tire unilateral wear detection method and vehicle provided in this disclosure can collect audio signals during vehicle operation, filter the audio signals to obtain a target audio signal, and further extract features from the target audio signal to obtain energy feature values ​​of a preset feature frequency band. Based on these energy feature values, detection is performed to obtain the tire unilateral wear detection result and output a warning message. Thus, by continuously collecting and automatically analyzing audio signals during vehicle operation, tire detection is transformed from periodic, passive detection to real-time, proactive detection. This allows for timely detection of abnormalities in the early stages of wear, avoiding missed optimal maintenance opportunities due to excessively long inspection intervals. By accurately extracting the energy feature values ​​of the preset feature frequency band, the system can sensitively identify characteristic signals of tire unilateral wear, thereby issuing warnings to the driver before wear deteriorates and causes safety accidents, effectively preventing serious accidents such as tire blowouts and loss of control caused by severe tire wear. A complete automated detection chain of acquisition, filtering, extraction, detection, and output is constructed, automatically completing tire condition assessment and warning without driver intervention, improving the intelligence level of vehicle maintenance and reducing the driver's burden. Attached Figure Description

[0029] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0030] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 This is a flowchart of a tire unilateral wear detection method provided in an embodiment of this disclosure; Figure 2 This is a flowchart of another tire unilateral wear detection method provided in this disclosure embodiment; Figure 3 This is a flowchart of another tire unilateral wear detection method provided in this disclosure embodiment; Figure 4 This is a schematic diagram of the overall architecture of a tire single-sided wear detection system provided in an embodiment of this disclosure; Figure 5 This is a schematic diagram of the structure of a tire unilateral wear detection device provided in an embodiment of this disclosure; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0032] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0033] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0034] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0035] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0036] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0037] Tires are a key component that ensures vehicle driving safety. Unilateral wear (commonly known as "tire wear") is a common problem that can lead to reduced grip, increased driving noise, and even tire blowout accidents.

[0038] Currently, tire wear detection mainly relies on regular manual inspections, tire pressure monitoring systems (TPMS), or regular four-wheel alignments. Manual inspections are inefficient, cannot achieve real-time monitoring, and are prone to missing early wear; TPMS can only monitor tire pressure and temperature and cannot identify unilateral wear problems caused by misalignment of the four wheels; regular four-wheel alignments are preventative maintenance and cannot detect dynamic changes in tire wear in real time.

[0039] The aforementioned methods lack real-time performance, automation, and early warning capabilities, resulting in tire wear problems going undetected and increasing driving safety risks. Therefore, this disclosure provides a method for detecting unilateral tire wear, which will be described below with reference to specific embodiments.

[0040] Figure 1 This is a flowchart of a tire unilateral wear detection method provided in an embodiment of the present disclosure. The method can be executed by a tire unilateral wear detection device, which can be implemented in software and / or hardware. The tire unilateral wear detection device can be configured in an electronic device, such as a server or terminal, wherein the terminal specifically includes an in-vehicle terminal, a computer or a tablet computer, etc.

[0041] like Figure 1 As shown, the tire unilateral wear detection method provided in this disclosure can be applied to the field of vehicle driving technology applications. For example, it can be used to detect or warn of tire unilateral wear. The tire unilateral wear detection method may include the following steps: S110, Collect audio signals while the vehicle is in motion.

[0042] In this embodiment of the disclosure, the electronic device continuously collects audio signals from the vehicle while it is in motion via a microphone. It is understood that the collected audio signal x(t) contains a mixture of engine noise, wind noise, road noise, in-vehicle voices / music, and possibly a specific low-frequency humming sound caused by unilateral tire wear. When a tire experiences unilateral wear, its contact patch becomes irregular, changing from smooth rolling to partial slippage and periodic impacts. This vibration is transmitted through the axle and body, producing a continuous, low-frequency humming sound inside the vehicle. The intensity of this sound is positively correlated with the severity of the wear.

[0043] S120. Filter the audio signal to obtain the target audio signal, which is an audio signal related to tire wear.

[0044] In this embodiment, after acquiring the audio signal during vehicle movement, the electronic device can filter the audio signal to obtain the target audio signal. The target audio signal is an audio signal related to tire wear. The purpose of this step is to extract the effective signal related to tire wear from a noisy background. Specifically, the electronic device can filter the audio data within the current time window (e.g., 2 seconds). This is a crucial step in improving detection accuracy and can employ a multi-level filtering strategy. The input is the original audio signal x(t), which includes engine noise, wind noise, road noise, human voice, music, and tire wear noise. The output is the filtered audio signal y(t) with a significantly improved signal-to-noise ratio, facilitating subsequent feature extraction.

[0045] S130. Extract features from the target audio signal to obtain the energy feature value of the preset feature frequency band. The preset feature frequency band is used to analyze the wear of one side of the tire.

[0046] In this embodiment, the electronic device can extract features from the target audio signal to obtain energy feature values ​​for a preset feature frequency band. The preset feature frequency band is used to analyze tire wear on one side. The goal of this step is to quantify the target audio signal into an evaluable index, namely, the energy feature value.

[0047] S140. Based on energy characteristic values, the detection results of tire wear on one side are obtained, and warning information is output.

[0048] In this step, the electronic device will perform detection based on energy characteristic values ​​to obtain the detection result of tire wear on one side and output warning information. Specifically, if the detection result is tire wear on one side, corresponding warning information will be output according to the degree of wear.

[0049] Therefore, in this embodiment, audio signals from a moving vehicle can be collected, filtered, and a target audio signal obtained. Further, feature extraction is performed on the target audio signal to obtain energy feature values ​​for a preset feature frequency band. Then, detection is performed based on these energy feature values ​​to obtain the detection result of tire wear on one side and output a warning message. Thus, by continuously collecting and automatically analyzing audio signals while the vehicle is in motion, tire detection is transformed from periodic, passive detection to real-time, proactive detection. This allows for timely detection of abnormalities in the early stages of wear, avoiding missed optimal maintenance opportunities due to excessively long inspection intervals. By accurately extracting the energy feature values ​​from the preset feature frequency band, the system can sensitively identify characteristic signals of tire wear on one side, thereby issuing warnings to the driver before wear deteriorates and causes safety accidents, effectively preventing serious accidents such as tire blowouts and loss of control caused by severe tire wear. A complete automated detection chain of acquisition-filtering-extraction-detection-output is constructed, automatically completing tire condition assessment and warnings without driver intervention, improving the intelligence level of vehicle maintenance and reducing the driver's burden.

[0050] Figure 2 This is a flowchart of another tire unilateral wear detection method provided in this embodiment.

[0051] like Figure 2 As shown, the tire unilateral wear detection method may include the following steps: S310, Collect audio signals while the vehicle is in motion.

[0052] Specifically, the implementation process and principle of S310 and S110 are the same, and will not be repeated here.

[0053] S320. Preprocess the audio signal to obtain the preprocessed audio signal.

[0054] In this step, the electronic device preprocesses the audio signal to obtain a preprocessed audio signal. Specifically, preprocessing includes sampling and quantization, framing, and windowing. Sampling and quantization: The analog signal collected by the in-vehicle microphone is converted into a digital signal by an analog-to-digital converter (ADC). The sampling rate is typically set to 8kHz or 16kHz, which is sufficient to cover the main frequency band of tire noise (usually below 1kHz) while also considering computational efficiency. Framing: The continuous audio signal is divided into short frames, each typically 20-40 milliseconds long (e.g., 256 or 512 sampling points per frame at a 16kHz sampling rate). This is because the audio signal can be considered quasi-stationary over a short period. Windowing: A window function (such as a Hamming window) is applied to each frame of data to reduce spectral leakage caused by discontinuities at both ends of the signal due to framing.

[0055] S330. Perform high-pass filtering on the preprocessed audio signal to obtain the first audio signal.

[0056] In this step, the electronic equipment performs a high-pass filter on the pre-processed audio signal to obtain the first audio signal. The purpose is to filter out the strong low-frequency roaring sound (usually below 80Hz) generated when the engine is idling and running at low speeds, thus removing low-frequency noise. Specifically, a digital high-pass filter is used, with a cutoff frequency set to 80Hz, as a preliminary cleaning step.

[0057] S340. The first audio signal is denoised using spectral subtraction to obtain the second audio signal.

[0058] In this step, the electronic device performs noise reduction on the first audio signal based on spectral subtraction to obtain the second audio signal. The purpose is to estimate and subtract the background steady-state noise.

[0059] Specifically, 1) During the initial stage of vehicle startup or during a period of steady, uniform driving with no significant tire noise, collect several frames of audio data and calculate the average power spectrum |N(f)| of these frames of audio data. 2 , as an estimate of the noise spectrum.

[0060] Optional, |N(f)| 2 The calculation method is as follows: Physical meaning: |N(f)|² represents the average power of the background noise at each frequency point f. Subtracting this estimate from the noisy signal is equivalent to removing the portion of energy that the noise would normally occupy at each frequency point, thus highlighting the target signal.

[0061] Step 1: From continuous signal to discrete frame Noise signal acquisition: During a set pure noise period (such as when the vehicle is just starting and idling, or when it is driving at a constant speed and without abnormal noise), a continuous time-domain noise signal n(t) is acquired.

[0062] Analog-to-digital conversion: n(t) is converted into a discrete sequence n[m] by an ADC, where m is the sampling point index.

[0063] Framing: Divide the long sequence n[m] into M short, possibly partially overlapping frames. Let the length of each frame be L (e.g., L = 512 sampling points). The time-domain signal of the i-th frame is denoted as n_i[m], where m = 0, 1, ..., L-1.

[0064] Step 2: Single-frame power spectrum calculation For each frame n_i[m], perform the following operation: Windowing: To reduce spectral leakage, a window function w[m] (such as a Hamming window) is applied to the frame data.

[0065] n_i_w[m] = n_i[m]•w[m] Discrete Fourier Transform: The Fourier Transform (FFT) algorithm is used to transform the windowed signal from the time domain to the frequency domain.

[0066] N_i[k]=DFT{n_i_w[m]} Here, k is the frequency index (k=0,1,...,L-1), corresponding to the actual frequency f=k•F_s / L, where F_s is the sampling frequency.

[0067] Calculating a single-frame periodogram: The power spectrum estimate of a single frame is called a periodogram. It is given by the squared magnitude of the spectral coefficients of that frame.

[0068] Formula: P_i[k] = |N_i[k]|² To elaborate further: |N_i[k]|²=(Real{N_i[k]})²+(Imag{N_i[k]})² Where Real and Imag represent taking the real and imaginary parts of the complex number, respectively.

[0069] Step 3: Averaging to obtain the final noise power spectrum estimate The periodogram P_i[k] of a single frame has a large variance, resulting in a coarse power spectrum estimate. By averaging the periodograms of M consecutive, non-overlapping (or partially overlapping) noise frames, we can obtain a smooth and statistically stable noise power spectrum estimate.

[0070] Final formula: |N[k]|²=(1 / M)•Σ_{i=1}^{M}P_i[k]=(1 / M)•Σ_{i=1}^{M}|N_i[k]|² This |N[k]|² is the final required average noise power spectrum estimate |N(f)| 2 It is an array of length L, where each position k stores the average background noise power corresponding to that frequency point.

[0071] 2) For each frame of the audio signal to be analyzed, calculate its power spectrum |X(f)| 2 .

[0072] 3) Subtract the estimated noise spectrum from the noisy signal spectrum: |Y(f)| 2 =∣X(f)∣ 2 α |N(f)| 2 α is an over-subtraction factor (usually ≥1) used to compensate for the bias in noise estimation.

[0073] 4) For |Y(f)| 2 Perform half-wave rectification (set negative values ​​to 0), and then reconstruct the denoised time-domain signal y(t) using inverse FFT.

[0074] This completes the noise reduction process and yields the second audio signal.

[0075] S350. Bandpass filtering is performed on the second audio signal to obtain the target audio signal.

[0076] In this step, the electronic device performs bandpass filtering on the second audio signal to obtain the target audio signal. The purpose is to confine the signal to the characteristic frequency band of interest and eliminate irrelevant interference to the greatest extent possible. Specifically, a bandpass filter is designed, for example, with a passband frequency of 100Hz-600Hz. This range is set based on the physical characteristics of tire wear noise. After this filtering, the target audio signal ybp(t) is obtained, which mainly contains the characteristic humming sound of tire wear.

[0077] This embodiment systematically filters out low-frequency engine noise, steady-state background noise, and other irrelevant frequency band interference through a multi-stage signal processing flow including preprocessing, high-pass filtering, spectral subtraction noise reduction, and band-pass filtering. This significantly improves the signal-to-noise ratio of the target signal and lays a solid foundation for accurate feature extraction in the future.

[0078] S360. Perform frequency domain transformation on the target audio signal to obtain the frequency domain signal.

[0079] In this step, the electronic device performs a frequency domain transformation on the target audio signal to obtain a frequency domain signal. Specifically, a Fast Fourier Transform (FFT) is performed on each frame ybp(t) to transform it from the time domain to the frequency domain, obtaining its amplitude spectrum Y(f). In the frequency domain, the frequency components corresponding to different sound sources are immediately apparent. The "humming" sound of tire wear will manifest as one or more prominent "spectral peaks" within a specific frequency range.

[0080] S370. Calculate the average amplitude of each frequency component in the preset characteristic frequency band based on the frequency domain signal, and determine the average amplitude as the energy characteristic value.

[0081] One or more key frequency windows are predefined, such as [f low f high The frequency band [150Hz, 250Hz] is the characteristic frequency band of tire wear on one side, i.e., the preset characteristic frequency band. Specifically, the preset characteristic frequency band [f] is calculated on the amplitude spectrum Y(f). low f high The average amplitude of all frequency components within the frame can also be the root mean square (RMS) value, which can be used as the energy characteristic value A of the frame data.

[0082] formula: , where N is the number of frequency points within the frequency band.

[0083] The energy characteristic value A directly reflects the intensity of the "humming" sound. The larger the A value, the more intense the vibration at that characteristic frequency, and the higher the likelihood and severity of tire wear. Essentially, it converts the target audio signal into a numerical indicator that can quantify the degree of wear.

[0084] Optionally, the preset characteristic frequency bands are derived from the tire's physical structure, vibration modes, and driving dynamics. Specifically, 1) analysis of the tire's pass frequencies and orders: Passing frequency: When a tire rolls, every irregular point on the tire tread (such as a flattened area caused by wear) periodically impacts the ground, generating periodic excitation. The frequency of this is called the passing frequency (or rotational frequency).

[0085] Calculation formula: ffundamental = vehicle speed (m / s) / tire rolling circumference (m).

[0086] For example, the tire specification is 225 / 65R17, and its rolling circumference is approximately 2.25 meters; the vehicle speed is 80km / h≈22.22m / s, so the basic passing frequency ffundamental=22.22 / 2.25≈9.9.

[0087] Harmonic resonance: Unilateral wear is usually not a single point, but rather uniform wear around the entire tire shoulder. This type of wear can excite structural resonance in the tire. This resonance frequency is much higher than the baseline passing frequency, often tens of times higher. For passenger car tires, the typical range of their radial (perpendicular to the ground) first or second order resonance frequencies is between 80Hz and 300Hz, especially concentrated in the 150-250Hz range.

[0088] When one side of the tire wears down, the stiffness distribution changes, and the impact when it touches the ground is more intense, which strongly excites this resonant frequency band, thus producing the "humming" sound that can be heard.

[0089] 2) Acoustic research and experimental verification Extensive actual recordings and analysis: By recording and analyzing the sound spectrum of vehicles known to have unilateral wear, it can be clearly observed that the energy amplitude in the 150-250Hz frequency band of the sound spectrum of worn vehicles is significantly higher than that of normal tires.

[0090] Industry experience: Tire "rumbling" or "wear noise" is typically associated with the 100-500Hz frequency range. Among these, 100-200Hz is often reported as a critical sensitive frequency caused by uneven tire wear or dynamic imbalance.

[0091] 3) In order to avoid the main sources of interference Engine: The combustion and mechanical noise of the engine are mainly concentrated in the <150Hz range at low to medium speeds. Setting the characteristic frequency band above 150Hz can effectively avoid most engine interference.

[0092] Road noise: The roughness of ordinary asphalt roads generates a wide noise spectrum, but the energy usually decreases as the frequency increases. In the 150-250Hz range, as long as the road surface is not particularly rough, the resonance peak of tire wear can be highlighted.

[0093] Therefore, setting the preset characteristic frequency band to [150Hz, 250Hz] is a comprehensive choice based on tire dynamics theory, industry experience, and actual experimental data. It is a highly sensitive and specific range that can most effectively separate the noise from unilateral tire wear from the complex background of in-vehicle sound.

[0094] This embodiment quantifies features by using FFT transformation and calculating the average amplitude of a specific frequency band, which can effectively capture wear characteristics while avoiding major interference sources, greatly improving the targeting and accuracy of detection.

[0095] S380 uses energy characteristic values ​​to detect tire wear on one side and outputs warning information.

[0096] Specifically, the implementation process and principle of S380 and S140 are the same, and will not be repeated here.

[0097] This embodiment of the disclosure acquires audio signals from a moving vehicle, preprocesses the audio signals to obtain preprocessed audio signals, and then performs high-pass filtering on the preprocessed audio signals to obtain a first audio signal. Further, the first audio signal is denoised using spectral subtraction to obtain a second audio signal, which is then band-pass filtered to obtain a target audio signal. Then, the target audio signal is frequency-domain transformed to obtain a frequency-domain signal. Based on the frequency-domain signal, the average amplitude of each frequency component within a preset characteristic frequency band is calculated, and the average amplitude is determined as an energy feature value. Subsequently, detection is performed based on the energy feature value to obtain the detection result of tire wear on one side, and a warning message is output. Thus, through a multi-stage audio signal processing flow, low-frequency engine noise, steady-state background noise, and other irrelevant frequency band interference are systematically filtered out, significantly improving the signal-to-noise ratio of the target audio signal and laying a solid foundation for subsequent accurate feature extraction. By quantifying features through frequency domain transformation and calculation of the average amplitude of specific frequency bands, wear characteristics can be effectively captured while avoiding major interference sources. This greatly improves the targeting and accuracy of detection, enabling automated and real-time detection of tire wear on one side, and solving the problem of lag caused by reliance on manual labor and periodic inspections.

[0098] Figure 3 This is a flowchart of another tire unilateral wear detection method provided in this disclosure embodiment.

[0099] like Figure 3 As shown, the tire unilateral wear detection method may include the following steps: S410: Collect audio signals while the vehicle is in motion.

[0100] Specifically, the implementation process and principle of S410 and S110 are the same, and will not be repeated here.

[0101] S420. Filter the audio signal to obtain the target audio signal, which is an audio signal related to tire wear.

[0102] Specifically, the implementation process and principle of S420 and S120 are the same, and will not be repeated here.

[0103] S430. Extract features from the target audio signal to obtain the energy feature value of the preset feature frequency band. The preset feature frequency band is used to analyze the wear of one side of the tire.

[0104] Specifically, the implementation process and principle of S430 and S130 are the same, and will not be repeated here.

[0105] S440. Determine whether the vehicle is currently in the target scene, which is a signal-to-noise ratio enhanced scene.

[0106] In this step, the electronic equipment determines whether the vehicle is currently in the target scenario. The target scenario is a signal-to-noise ratio enhanced scenario, also known as a sound reflection enhanced scenario. This scenario refers to an environment where the vehicle is driving near obstacles that can reflect sound, such as medians, bridge piers, and guardrails. In this scenario, the in-vehicle microphones can receive the superposition of direct sound and reflected sound, thus effectively increasing the sound signal-to-noise ratio. Specifically, real-time data from cameras, GPS / maps, and the vehicle bus (CAN) can be used to determine whether the vehicle is currently in the target scenario.

[0107] By introducing scene judgment and identifying the target scene, the system cleverly utilizes the physical characteristics of the environment to amplify the effective signal, enabling the detection of problems when wear is less and the signal is weaker, thus achieving early warning.

[0108] In some embodiments, S440 may include S4401 and S4402: S4401. Based on the current image data collected by the vehicle, scene recognition is performed to obtain the first recognition result and the first confidence level.

[0109] In this step, the electronic device can perform scene recognition based on the current image data collected by the vehicle to obtain a first recognition result and a first confidence level. Specifically, video streams captured by cameras on the side of the vehicle (such as surround-view cameras or rearview mirror cameras) are used to perform real-time image semantic segmentation or target detection using a lightweight convolutional neural network (CNN) model to identify targets as reflective objects: such as bridge piers, walls of viaducts, tunnel walls, highway guardrails (especially concrete sections), continuous medians, and adjacent lane crash barriers. For example, video frames are captured at a certain frequency (e.g., 10fps), and the images are input into the CNN model. The model is trained to recognize the aforementioned reflective objects and outputs their position and confidence level in the image. If a reflective object with a confidence level > confidence threshold (e.g., 0.7) is detected located on the side of the vehicle (e.g., within 0.5-3 meters to the left or right), it is determined that the target scene has been entered. The confidence threshold can be determined based on a large number of experimental results. The first recognition result can be the position of the reflective object in the image.

[0110] S4402. Based on the vehicle's current location information and a high-precision map, scene matching is performed to obtain a second recognition result and a second confidence level.

[0111] In this step, the electronic device can perform scene matching based on the vehicle's current location information and a high-precision map to obtain a second recognition result and a second confidence level. Specifically, it acquires the vehicle's GPS coordinates and high-precision map data. Targets are identified based on map matching and geofencing technology. For example, tunnels are marked with tunnel entrances and exits in the map data; bridges / elevated roads are marked with major bridge sections in the map data. For instance, GPS coordinates are continuously acquired, and the coordinates are matched with points of interest (POIs) on the map. When the vehicle's location enters a polygonal area marked as a "tunnel" or "specific bridge" on the map, a target scene marker is triggered, indicating that the vehicle has entered the target scene.

[0112] By employing two methods for scene matching, the flexibility and accuracy of scene matching are improved.

[0113] In some embodiments, S440 may include S44001, S44002, S44003, and S44004: S44001, Obtain the first recognition result, the first confidence level, the second recognition result, and the second confidence level.

[0114] In this step, the first identification result, the first confidence level, the second identification result, and the second confidence level can be obtained.

[0115] S44002. Based on the current external environment information, dynamically allocate the weights of the first recognition result and the second recognition result.

[0116] The reliability of the first recognition result of the vision system and the second recognition result of the position system varies significantly under different weather and lighting conditions. By evaluating the confidence of each sensor in real time and dynamically assigning weights accordingly, a comprehensive confidence level for determining whether to enter a reflection-enhanced scene is finally calculated.

[0117] The visual system's base weight W_v_base is set to 0.8 because the visual system is most reliable under ideal conditions (sunny daytime).

[0118] The base weight of the position system, W_p_base, is set to 0.5 because the position system is not affected by optical conditions but has limited accuracy (it cannot identify temporary obstacles).

[0119] Table 1 defines the attenuation coefficient α_v (between 0 and 1, where 1 indicates no attenuation) for the reliability of the vision system under different environmental conditions.

[0120] Table 1. Attenuation coefficients of visual system reliability under different environmental conditions

[0121] The final dynamic weight is the result of the base weight decaying through all current environment states.

[0122] The final weights W_v of the visual system are determined by the following formula: W_v=W_v_base*α_v_night*α_v_rain*α_v_fog*α_v_snow Example 1: Sunny daytime: W_v = 0.8 * 1.0 * 1.0 * 1.0 * 1.0 = 0.8; Example 2: Rainy night: W_v=0.8*0.4*0.5*1.0*1.0=0.16.

[0123] The location system is relatively reliable, but to prevent over-reliance on it in the event of complete visual failure, an upper limit is set for its reliability. The final weight W_p of the location system is determined by the following formula: W_p = min(W_p_base, 1.0 - W_v) This formula ensures that when the visual system's weight is very low, the position system's weight will automatically compensate, but the sum will not exceed 1. This forms a complementary relationship where one increases at the expense of the other.

[0124] Example 1: Sunny daytime: W_p=min(0.5,1.0-0.8)=0.2(total 1.0); Example 2: Rainy night: W_p=min(0.5,1.0-0.16)=0.5 (total 0.66).

[0125] Thus, the weights of the first and second identification results are determined.

[0126] S44003. Based on the first confidence level, the second confidence level and their respective assigned weights, a weighted fusion is performed to obtain the comprehensive confidence level.

[0127] In this step, the electronic device performs a weighted fusion based on a first confidence level, a second confidence level, and their respective assigned weights to obtain a comprehensive confidence level. The first confidence level V_score of the reflected scene output by the vision system and the second confidence level P_score of the reflected scene output by the position system are used. Dynamic weights W_v and W_p are calculated based on the current environmental state.

[0128] Weighted calculation: V = V_score * W_v, P = P_score * W_p.

[0129] Calculate the overall confidence level: S = V + P.

[0130] S44004. If the overall confidence level exceeds the preset threshold, it is determined that the current location is the target scene.

[0131] In this step, if the overall confidence level exceeds a preset threshold, it is determined that the current location is in the target scene. For example, if a reflector with an overall confidence level S > the threshold (e.g., 0.7) is detected located to the side of the vehicle (e.g., within 0.5-3 meters to the left or right), it is determined that the vehicle has entered a reflection scene.

[0132] By fusing information from multiple sensors (visual and location) and making decisions based on dynamically allocated weights according to the environment, the scene judgment maintains high reliability and robustness even under adverse conditions such as nighttime, rain, snow, and fog, ensuring the effective operation of the core detection mechanism around the clock.

[0133] S450. Aggregate the energy eigenvalues ​​to obtain aggregated eigenvalues.

[0134] In this step, the electronic device aggregates the energy feature values ​​to obtain an aggregated feature value. Specifically, the calculation for a single frame may be affected by a momentary impact (such as running over a stone). Therefore, it is necessary to perform a moving average or median filtering on the feature values ​​A calculated over multiple consecutive frames (e.g., all frames within 1-2 seconds) to obtain a more stable and reliable aggregated feature value A. avg A avg =(A1+A2+A3+…A N ) / N.

[0135] S460. When the vehicle is currently in the target scenario, the aggregated feature value is compared with a preset first threshold. If the aggregated feature value is greater than or equal to the preset first threshold, it is determined that tire wear has occurred on one side.

[0136] In this embodiment, a fixed threshold is not used; instead, the threshold is dynamically adjusted according to the scenario. A preset first threshold Th... reflect The threshold for judgment in the target scenario; a second threshold Th is preset. normal This is the threshold for judgment in normal scenarios. Wherein, Th reflect <Th normal Because in the target scenario—namely, a scenario with enhanced sound reflection—the same amount of wear will produce a stronger effective signal, a more sensitive threshold can be used to detect problems earlier. reflect and Th normal The set value can be given in advance within a range based on actual scenario testing, and optimized with a large number of experimental results.

[0137] In this step, if the vehicle is currently in the target scene, the aggregated feature value will be compared with a preset first threshold Th. reflect The comparison is performed, and if the aggregated feature value is greater than or equal to the preset first threshold, it is determined that tire wear has occurred on one side.

[0138] S470. If the vehicle is not currently in the target scenario, compare the aggregated feature value with a preset second threshold. If the aggregated feature value is greater than or equal to the preset second threshold, determine that tire wear has occurred on one side.

[0139] In this step, if the vehicle is not currently in the target scenario, i.e., in a normal scenario, the aggregated feature value is compared with a preset second threshold Th. normal The comparison is performed, and if the aggregated feature value is greater than or equal to the preset second threshold, it is determined that tire wear has occurred on one side.

[0140] This embodiment introduces scene recognition and a dynamic threshold mechanism. By identifying the target scene and applying a lower detection threshold, it cleverly utilizes the physical characteristics of the environment to amplify the effective signal, enabling the detection of problems when wear is less and the signal is weaker, thus achieving early warning. Simultaneously, the dual threshold mechanism prevents false alarms in noisy, ordinary environments.

[0141] S480. Within a preset time window, determine the frequency and duration of events that cause unilateral tire wear.

[0142] In this step, the frequency and duration of events involving unilateral tire wear are determined within a preset time window. Specifically, a single wear event is insufficient to trigger an immediate alarm, to prevent false alarms (e.g., driving over a rough road surface). The electronic device will count the frequency and duration of unilateral tire wear events within a time window (e.g., 5 minutes of driving distance).

[0143] S490. Based on the frequency and duration of the event, determine the severity level of tire wear on one side, and generate and output the corresponding level of warning information based on the severity level.

[0144] In this step, the electronic device determines the severity level of tire wear on one side based on the frequency and duration of the event, and generates and outputs a corresponding level of warning information based on the severity level.

[0145] In some embodiments, the severity levels can be divided into: Level 0 (Normal): The incidence of wear events is extremely low (<5% of the time), and the aggregated characteristic value is consistently below a preset second threshold. Level 1 (Slight Wear): Wear events are stably triggered in the target scenario (>30% of the target scenario time), but not in normal scenarios; the confidence levels of event frequency and event duration accumulate to the preset Level 1 threshold. Level 2 (Moderate Wear): Wear events also begin to be triggered intermittently in normal scenarios (>20% of the normal scenario time). The confidence levels of event frequency and event duration accumulate to the Level 2 threshold. Level 3 (Severe Wear): Wear events are triggered continuously and intensely in normal scenarios (>60% of the time). The aggregated characteristic value consistently approaches or exceeds the preset second threshold.

[0146] By using quantitative analysis to achieve multi-level early warning and matching it with different levels of handling strategies, we can avoid false alarms that disturb the public, provide strong intervention in emergency situations, and provide a better user experience.

[0147] In some embodiments, S490 generates and outputs warning information corresponding to the severity level, including S4901, S4902, and S4903: S4901. When the severity level is Level 1, a tire check reminder icon is displayed on the instrument panel.

[0148] In this step, when the severity level is Level 1 (which can be Minor), a tire check reminder icon can be displayed on the dashboard to prompt the user to check the tires.

[0149] S4902. When the severity level is Level 2, a voice prompt will be output, and tire inspection suggestions will be displayed on the central control screen.

[0150] In this step, when the severity level is Level 2 (which can be classified as moderate), a voice prompt can be issued, and tire inspection suggestions can be displayed on the central control screen. For example, the voice prompt might say, "Please note that abnormal tire wear has been detected. It is recommended to schedule an inspection as soon as possible." The central control screen would then display detailed information such as, "Abnormal wear on the inner side of the left front tire. Possible cause: misalignment. Recommendation: Tire rotation and four-wheel alignment." S4903 When the severity level is level 3, output a hazard warning signal and recommend speed limit.

[0151] In this step, when the severity level is Level 3 (which can be classified as severe), a hazard warning signal is output, and a speed limit is recommended. For example, the output might be: "Warning! Severe tire wear poses a safety risk. Please slow down immediately and drive with caution!" Specific action guidelines are displayed: "Severe Safety Warning: Tire wear has affected safety. Please avoid high-speed driving and proceed to a service center immediately." A speed limit is also recommended: a request is sent to the vehicle control system, and a message appears on the instrument panel: "System recommended maximum speed: 80 km / h."

[0152] This disclosure clarifies the specific strategy for outputting warning information, employing a tiered response strategy ranging from "mild prompts" to "strong warnings + recommended speed limits." This significantly optimizes the user experience, minimizing unnecessary driving interference while ensuring the effectiveness of safety warnings.

[0153] This embodiment of the disclosure collects audio signals from a moving vehicle, filters the audio signals to obtain a target audio signal, extracts features from the target audio signal to obtain energy feature values ​​for a preset feature frequency band. Then, it determines whether the vehicle is currently in the target scene, aggregates the energy feature values ​​to obtain aggregated feature values. Further, if the vehicle is currently in the target scene, the aggregated feature value is compared with a preset first threshold. If the aggregated feature value is greater than or equal to the preset first threshold, it is determined that unilateral tire wear has occurred. If the vehicle is not currently in the target scene, the aggregated feature value is compared with a preset second threshold. If the aggregated feature value is greater than or equal to the preset second threshold, it is determined that unilateral tire wear has occurred. Then, within a preset time window, the frequency and duration of unilateral tire wear events are determined. Based on the event frequency and duration, the severity level of unilateral tire wear is determined, and a corresponding level of warning information is generated and output based on the severity level. Therefore, a scene-based judgment and dynamic threshold mechanism was introduced. By identifying the target scene and applying a lower detection threshold, the system cleverly utilizes the physical characteristics of the environment to amplify the effective signal. This allows the system to detect problems when wear is less severe and the signal is weaker, achieving early warning. The dual-threshold mechanism prevents false alarms in noisy, ordinary environments. The persistence and severity of wear are assessed by accumulating event frequency and duration, and a graded response strategy is adopted for different levels. This significantly optimizes the user experience, ensuring the effectiveness of safety warnings while minimizing unnecessary driving interference.

[0154] Figure 4 This is a schematic diagram of the overall architecture of a tire single-sided wear detection system provided in this embodiment of the disclosure, as shown below. Figure 4 As shown, the tire single-sided wear detection system includes a data acquisition layer, a signal processing and analysis layer, and a decision and output layer.

[0155] Among them, 1) the data acquisition layer includes: In-vehicle microphones: used to collect broadband sound signals generated by the friction between the tires and the ground while the vehicle is in motion. They are preferably placed in the vehicle chassis or inside the passenger compartment near each wheel to form an array to assist in sound source localization.

[0156] Vehicle-mounted camera: Used to capture environmental images from the side of the vehicle.

[0157] 2) Signal processing and analysis decision layer, including: Audio preprocessing module: performs noise reduction and bandpass filtering on the original audio, filters out irrelevant interference such as wind noise and engine noise, and retains the mid-to-high frequency range (e.g., 80Hz-1000Hz) related to tire friction.

[0158] Visual analysis module: Through image recognition technology, it analyzes camera footage in real time to determine whether the vehicle is in a "sound reflection enhancement scene" (such as identifying bridge piers, guardrails, tunnel walls, etc.).

[0159] Feature extraction module: Performs Fast Fourier Transform (FFT) on the preprocessed audio signal to transform it from the time domain to the frequency domain. Focuses on monitoring the energy amplitude of specific frequency bands (e.g., 150Hz±50Hz) associated with unilateral wear.

[0160] Scene fusion module: When the visual analysis module recognizes a scene with enhanced reflection, or when GPS / map data indicates that a vehicle has entered a tunnel or bridge, this module will trigger a sign to notify the feature extraction module to assign higher weight or credibility to the sound data within this time period.

[0161] Wear classification module: Classifies wear based on the energy amplitude (A) of the characteristic frequency band.

[0162] Strategy execution module: Executes the corresponding strategy based on the hierarchical results. Level 1: Displays an icon on the dashboard that says "Recommend checking your tires".

[0163] Level 2: Issues a voice prompt: "Abnormal tire wear detected. Please go to the service center for inspection as soon as possible."

[0164] Level 3: Issues continuous and strong warnings and displays specific suggestions on the central control screen (such as "Please perform a four-wheel alignment and tire check immediately"). It can also link with the vehicle system to suggest limiting the maximum speed.

[0165] 3) Output layer, including: Human-machine interface: responsible for informing the driver of all warning information and strategy execution results through the instrument panel, central control screen or voice system.

[0166] Figure 5This is a schematic diagram of the structure of a tire unilateral wear detection device provided in an embodiment of this disclosure.

[0167] In this embodiment, the tire unilateral wear detection device can be housed within an electronic device, and is understood as a functional module within the aforementioned electronic device. Specifically, the electronic device can be a server or a terminal, wherein the terminal specifically includes an in-vehicle terminal, a computer, or a tablet computer, etc., without limitation.

[0168] like Figure 5 As shown, the tire single-sided wear detection device 700 may include a data acquisition module 710, a data acquisition module 720, an extraction module 730, and a detection module 740.

[0169] The acquisition module 710 is used to acquire audio signals while the vehicle is in motion; The module 720 is used to filter the audio signal to obtain a target audio signal, wherein the target audio signal is an audio signal related to tire wear. The extraction module 730 is used to extract features from the target audio signal to obtain energy feature values ​​of a preset feature frequency band, which is used to analyze the wear condition of one side of the tire. The detection module 740 is used to perform detection based on the energy characteristic value, obtain the detection result of tire wear on one side, and output warning information.

[0170] In some embodiments of this disclosure, when the obtaining module 720 filters the audio signal to obtain the target audio signal, it is specifically used for: The audio signal is preprocessed to obtain a preprocessed audio signal; The preprocessed audio signal is subjected to high-pass filtering to obtain the first audio signal; The first audio signal is denoised using spectral subtraction to obtain the second audio signal. The second audio signal is subjected to bandpass filtering to obtain the target audio signal.

[0171] In some embodiments of this disclosure, when the extraction module 730 extracts features from the target audio signal to obtain energy feature values ​​of a preset feature frequency band, it is specifically used for: The target audio signal is transformed in the frequency domain to obtain a frequency domain signal; The average amplitude of each frequency component within a preset characteristic frequency band is calculated based on the frequency domain signal, and the average amplitude is determined as the energy characteristic value.

[0172] In some embodiments of this disclosure, before the detection based on the energy characteristic value is performed to obtain the detection result of tire unilateral wear, the device 700 further includes: The judgment module 750 is used to determine whether the vehicle is currently in the target scene, which is a signal-to-noise ratio enhanced scene.

[0173] In some embodiments of this disclosure, when the detection module 740 performs detection based on the energy characteristic value to obtain the detection result of tire wear on one side, it is specifically used for: The energy characteristic values ​​are aggregated to obtain aggregated characteristic values; When the vehicle is currently in the target scenario, the aggregated feature value is compared with a preset first threshold. If the aggregated feature value is greater than or equal to the preset first threshold, it is determined that unilateral tire wear has occurred; or, If the vehicle is not currently in the target scenario, the aggregated feature value is compared with a preset second threshold. If the aggregated feature value is greater than or equal to the preset second threshold, it is determined that tire wear has occurred on one side. Wherein, the preset first threshold is less than the preset second threshold.

[0174] In some embodiments of this disclosure, when the determination module 750 determines whether the vehicle is currently in the target scene, it is specifically used for: Scene recognition is performed based on the current image data collected by the vehicle to obtain a first recognition result and a first confidence level; or Based on the vehicle's current location information and a high-precision map, scene matching is performed to obtain a second recognition result and a second confidence level.

[0175] In some embodiments of this disclosure, when the determination module 750 determines whether the vehicle is currently in the target scene, it is specifically used for: Obtain the first identification result, the first confidence level, the second identification result, and the second confidence level; Based on the current external environment information, the weights of the first identification result and the second identification result are dynamically allocated; A comprehensive confidence level is obtained by weighting and fusing the first confidence level, the second confidence level, and their respective assigned weights. If the overall confidence level exceeds a preset threshold, it is determined that the current location is the target scene.

[0176] In some embodiments of this disclosure, when the detection module 740 outputs warning information, it is specifically used for: Within a preset time window, determine the frequency and duration of events that cause unilateral tire wear. Based on the frequency and duration of the events, the severity level of tire wear on one side is determined, and a corresponding warning message is generated and output based on the severity level.

[0177] In some embodiments of this disclosure, when the detection module 740 generates and outputs warning information corresponding to the severity level, it is specifically used for: When the severity level is Level 1, a tire check reminder icon will be displayed on the dashboard; When the severity level is Level 2, a voice prompt will be output, and tire inspection suggestions will be displayed on the central control screen; When the severity level is Level 3, a hazard warning signal is output, and a speed limit is recommended.

[0178] It should be noted that, Figure 5 The tire single-sided wear detection device 700 shown can perform the various steps in the above method embodiments and achieve the various processes and effects in the above method embodiments, which will not be elaborated here.

[0179] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure.

[0180] In this embodiment of the disclosure, Figure 6 The electronic device shown can be a server or a terminal. Specifically, the terminal includes in-vehicle terminals, computers, or tablets, etc., without limitation.

[0181] like Figure 6 As shown, the electronic device may include a processor 810 and a memory 820 storing computer program instructions.

[0182] Specifically, the processor 810 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this disclosure.

[0183] Memory 820 may include mass storage for information or instructions. For example, and not limitingly, memory 820 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 820 may include removable or non-removable (or fixed) media. Where appropriate, memory 820 may be internal or external to the integrated gateway device. In a particular embodiment, memory 820 is non-volatile solid-state memory. In a particular embodiment, memory 820 includes read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (Electrically Programmable ROM, EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0184] The processor 810 reads and executes computer program instructions stored in the memory 820 to perform the steps of the tire unilateral wear detection method provided in this embodiment of the present disclosure.

[0185] In one example, the electronic device may also include a transceiver 830 and a bus 840. Wherein, as... Figure 6 As shown, the processor 810, memory 820 and transceiver 830 are connected via bus 840 and communicate with each other.

[0186] Bus 840 may include hardware, software, or both. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 840 may include one or more buses.

[0187] This disclosure also provides a computer-readable storage medium that can store a computer program. When the computer program is executed by a processor, the processor enables the processor to implement the tire unilateral wear detection method provided in this disclosure.

[0188] When the computer program is executed by the processor, it can perform the following steps: acquire audio signals during vehicle operation; filter the audio signals to obtain a target audio signal, which is an audio signal related to tire wear; extract features from the target audio signal to obtain energy feature values ​​of a preset feature frequency band, which is used to analyze the wear of one side of the tire; perform detection based on the energy feature values ​​to obtain the detection result of tire wear on one side, and output warning information.

[0189] The aforementioned storage medium may, for example, include a memory 820 containing computer program instructions, which can be executed by a processor 810 of an electronic device to complete the tire unilateral wear detection method provided in this embodiment. Optionally, the storage medium may be a non-transitory computer-readable storage medium, such as a read-only memory (ROM), random access memory (RAM), external cache memory, compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, flash memory, and optical data storage device. By way of illustration and not limitation, RAM is available in various forms, such as static random access memory (SRAM) and dynamic random access memory (DRAM).

[0190] This disclosure also provides a vehicle that includes electronic devices that can implement the various processes and effects described in the above embodiments of this disclosure, which will not be elaborated here.

[0191] This disclosure also provides a computer program product, which includes a computer program or instructions. When the computer program or instructions are executed by a processor, they implement the tire unilateral wear detection method provided in this disclosure and can achieve the various processes and effects in the above embodiments of this disclosure, which will not be elaborated here.

[0192] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method of detecting single side wear of a tire, characterized by, The method comprises: Collecting an audio signal in vehicle driving; Filtering the audio signal to obtain a target audio signal, the target audio signal being an audio signal related to tire wear; Extracting features from the target audio signal to obtain an energy feature value of a preset feature frequency band, the preset feature frequency band being used to analyze tire single-side wear condition; Based on the energy feature value, detecting a tire single-side wear detection result and outputting a warning information.

2. The method of claim 1, wherein, The filtering of the audio signal to obtain a target audio signal comprises: Pretreating the audio signal to obtain a pretreated audio signal; High-pass filtering the pretreated audio signal to obtain a first audio signal; Based on the spectrum subtraction method, denoising the first audio signal to obtain a second audio signal; Band-pass filtering the second audio signal to obtain the target audio signal.

3. The method of claim 1, wherein, The feature extraction from the target audio signal to obtain an energy feature value of a preset feature frequency band comprises: Frequency domain transformation of the target audio signal to obtain a frequency domain signal; Based on the frequency domain signal, calculating the average amplitude of each frequency component in the preset feature frequency band, and determining the average amplitude as the energy feature value.

4. The method of claim 1, wherein, Before the detection based on the energy feature value to obtain a tire single-side wear detection result, the method further comprises: Judging whether the vehicle is currently in a target scene, the target scene being a signal-to-noise ratio enhanced scene.

5. The method of claim 4, wherein, The detection based on the energy feature value to obtain a tire single-side wear detection result comprises: Aggregating the energy feature value to obtain an aggregated feature value; In the case that the vehicle is currently in the target scene, comparing the aggregated feature value with a preset first threshold value, if the aggregated feature value is greater than or equal to the preset first threshold value, it is determined that the tire single-side wear occurs currently; or, In the case that the vehicle is not currently in the target scene, comparing the aggregated feature value with a preset second threshold value, if the aggregated feature value is greater than or equal to the preset second threshold value, it is determined that the tire single-side wear occurs currently; Wherein, the preset first threshold value is less than the preset second threshold value.

6. The method of claim 4, wherein, The judgment of whether the vehicle is currently in a target scene comprises: Based on the current image data collected by the vehicle, scene recognition is performed to obtain a first recognition result and a first confidence; or Based on the current position information of the vehicle and the high-precision map, scene matching is performed to obtain a second recognition result and a second confidence.

7. The method of claim 6, wherein, The judgment of whether the vehicle is currently in a target scene comprises: Obtaining a first recognition result, a first confidence, a second recognition result and a second confidence; Based on the current external environment information, dynamically allocating the weights of the first recognition result and the second recognition result; Based on the first confidence, the second confidence and the respective allocated weights, weighted fusion is performed to obtain a comprehensive confidence; If the comprehensive confidence exceeds a preset threshold, it is determined that the vehicle is currently in the target scene.

8. The method of claim 1, wherein, The output of the warning information comprises: In a preset time window, determine the event frequency and event duration of the occurrence of tire single-side wear; According to the event frequency and the event duration, determine the severity level of the tire single-side wear, and generate and output the corresponding level of warning information according to the severity level.

9. The method of claim 8, wherein, The generation and output of the corresponding level of warning information according to the severity level comprises: When the severity level is a first level, display a tire inspection prompt icon on the instrument panel; When the severity level is a second level, output a voice prompt and display a tire inspection suggestion on the center control screen; When the severity level is a third level, output a danger warning signal and suggest a speed limit.

10. A vehicle characterized by comprising: Comprise: a memory; a processor; and a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method as claimed in any one of claims 1-9.

Images