Self-cleaning control method of camera module and related equipment

Abstract

The embodiment of the invention provides a self-cleaning control method for a camera module and related equipment, and the method comprises the steps: obtaining the image data collected by the camera module at present, and enabling the image data to be used for evaluating the pollution state of the camera module; processing the image data based on a preset image recognition algorithm, and determining the pollution degree of the camera module; determining a piezoelectric feedback signal on the current camera module; determining a cleaning control parameter of the current camera module based on the piezoelectric feedback signal and the pollution degree; and based on the cleaning control parameter, executing a corresponding cleaning action to clean the current camera module. Through the steps of the method, the pollution degree of the surface of the lens can be intelligently judged and the corresponding cleaning behavior can be adaptively controlled and adjusted in combination with the image recognition result and the piezoelectric feedback information, so that the cleaning efficiency and the control precision are effectively improved, and the imaging quality of camera equipment is ensured.

Description

Technical Field

[0001] This invention relates to the field of ultrasonic cleaning of camera modules, and more particularly to a self-cleaning control method, device, electronic device, and storage medium for camera modules. Background Technology

[0002] With the innovation of science and technology, the application fields of camera modules are becoming more diversified, and the number of installed modules is growing rapidly, from outdoor security to indoor homes, from fixed cameras to vehicle-mounted mobile cameras. Due to the increased demand for outdoor use, the impact of the outdoor environment is more pronounced. When the ambient temperature is low (such as in winter) and the air humidity is high, the lens surface temperature is below the dew point temperature. Water vapor in the air will condense into tiny water droplets on the lens surface, resulting in blurred images, reduced contrast, or even complete obstruction of the field of view. In environments with large temperature differences between indoors and outdoors (such as from an air-conditioned room to the outdoors), in high humidity environments at dawn and night, and in low-temperature environments such as rain or snow, the lens surface may be covered by snow, or rainwater may freeze directly into ice, physically obstructing the lens and damaging its optical performance. Some camera equipment used in external environments, such as highway surveillance cameras, are installed in high and inaccessible locations. Manual cleaning is difficult and maintenance costs are high. Rainwater adhering to the lens surface needs to be cleaned in a timely manner to ensure its imaging effect and thus improve the user experience. Another example is vehicle-mounted cameras, where lens fogging may occur due to the temperature difference between the inside and outside of the vehicle, requiring timely cleaning. However, it is inconvenient for the driver to clean them while driving.

[0003] While existing technologies have introduced vibration cleaning devices or image processing mechanisms, most rely solely on image occlusion judgment or a single cleaning trigger condition, lacking intelligent judgment logic that integrates image quality and device feedback, making it difficult to achieve precise and efficient cleaning control.

[0004] Therefore, existing self-cleaning control methods for camera modules have the problem of not being able to combine image recognition and piezoelectric feedback information to accurately assess the contamination status of the camera module and dynamically adjust cleaning parameters based on feedback information. Summary of the Invention

[0005] This invention provides a self-cleaning control method for camera modules to address the problem that existing self-cleaning control methods for camera modules cannot combine image recognition and piezoelectric feedback information to accurately assess the contamination status of the camera module and dynamically adjust cleaning parameters based on feedback information.

[0006] In a first aspect, embodiments of the present invention provide a self-cleaning control method for a camera module, the method comprising the following steps: The image data currently collected by the camera module is acquired, and the image data is used to assess the contamination status of the camera module; The image data is processed based on a preset image recognition algorithm to determine the degree of contamination of the camera module; Determine the piezoelectric feedback signal on the current camera module; Based on the piezoelectric feedback signal and the degree of contamination, the cleaning control parameters of the current camera module are determined; Based on the cleaning control parameters, the corresponding cleaning action is executed to clean the current camera module.

[0007] Optionally, acquiring the image data currently collected by the camera module includes: Based on preset keyframes, image data of the camera module is acquired to obtain image frame data; The image frame data is subjected to effective data classification preprocessing to obtain the image data currently acquired by the camera module.

[0008] Optionally, the step of processing the image data based on a preset image recognition algorithm to determine the degree of contamination of the camera module includes: The image data is processed by the preset image recognition algorithm to determine the sharpness data, occlusion area data and focus blur data corresponding to the image data; Based on the image data, including sharpness data, occlusion area data, and focus blur data, the degree of contamination of the current camera module is determined.

[0009] Optionally, determining the piezoelectric feedback signal on the current camera module includes: Based on preset keyframes, the changes in electrical signals on the current camera module are collected to obtain a continuous electrical signal change dataset, which contains multiple continuously changing electrical signals. The continuous electrical signal variation dataset is subjected to electrical signal variation feature recognition processing to obtain multiple electrical signal variation feature data; Based on the multiple electrical signal change characteristic data, the piezoelectric feedback signal on the current camera module is determined.

[0010] Optionally, determining the cleaning control parameters of the current camera module based on the piezoelectric feedback signal and the degree of contamination includes: Based on the degree of contamination, it is determined whether the cleaning triggering conditions are met, including the cleaning type and cleaning intensity. If the cleaning type and cleaning intensity are satisfied, then based on the piezoelectric feedback signal, control parameters corresponding to the cleaning type and cleaning intensity are generated, and the control parameters include at least one of the following during cleaning: the driving frequency of the cleaning component, the vibration amplitude, and the cleaning duration.

[0011] Optionally, the step of performing a corresponding cleaning action on the current camera module based on the cleaning control parameters includes: Based on the cleaning control parameters, the corresponding cleaning components are driven to perform vibration dewatering or heating defogging cleaning actions on the camera module during the cleaning duration with adjusted driving frequency and vibration amplitude.

[0012] Optionally, after performing the corresponding cleaning action on the current camera module based on the cleaning control parameters, the method further includes: Acquire updated image data and real-time piezoelectric feedback signal of the camera module after it has completed the cleaning action; The image data of the camera module is re-evaluated based on the updated image data, and the cleaning effect of the current camera module after the cleaning action is determined by combining the real-time piezoelectric feedback signal. If the preset cleaning threshold is not reached, the cleaning control parameters corresponding to the cleaning component are adaptively adjusted, and the corresponding cleaning action is re-executed until the cleaning effect of the current camera module after the cleaning action is completed reaches the preset cleaning threshold. If the preset cleaning threshold is reached, the cleaning process will stop.

[0013] Secondly, embodiments of the present invention also provide an online interactive device, the online interactive device comprising: The first acquisition module is used to acquire the image data currently collected by the camera module, and the image data is used to assess the contamination status of the camera module; The first determining module is used to process the image data based on a preset image recognition algorithm to determine the degree of contamination of the camera module; The second determining module is used to determine the piezoelectric feedback signal on the current camera module; The third determining module is used to determine the cleaning control parameters of the current camera module based on the piezoelectric feedback signal and the degree of contamination. The first control module is used to perform a corresponding cleaning action to clean the current camera module based on the cleaning control parameters.

[0014] Thirdly, embodiments of the present invention provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the self-cleaning control method for a camera module provided in embodiments of the present invention.

[0015] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the steps in the self-cleaning control method for a camera module provided in the embodiments of the present invention.

[0016] In this embodiment of the invention, image data currently collected by the camera module is acquired, and the image data is used to assess the contamination status of the camera module. The image data is processed based on a preset image recognition algorithm to determine the degree of contamination of the camera module. The piezoelectric feedback signal on the current camera module is determined. Based on the piezoelectric feedback signal and the degree of contamination, cleaning control parameters for the current camera module are determined. Based on the cleaning control parameters, corresponding cleaning actions are executed to clean the current camera module. Through the above method steps, the degree of contamination on the lens surface can be intelligently judged and the corresponding cleaning behavior can be adaptively controlled and adjusted by combining image recognition results and piezoelectric feedback information, effectively improving cleaning efficiency and control accuracy, and ensuring the imaging quality of the camera equipment. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a self-cleaning control method for a camera module provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a camera module structure provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of another online interactive device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] like Figure 1 As shown, Figure 1This is a flowchart of a self-cleaning control method for a camera module provided in an embodiment of the present invention. The self-cleaning control method for the camera module includes the following steps: 101. Obtain the image data currently collected by the camera module.

[0021] In this embodiment of the invention, the self-cleaning control method of the camera module can be applied to a camera self-cleaning processing platform. The camera self-cleaning processing platform has functions such as camera self-cleaning processing data processing, camera self-cleaning processing data transmission and reception, and camera self-cleaning processing data memory storage. It can be built based on a server or server cluster. The server or server cluster can be an electronic device with camera self-cleaning data processing capability.

[0022] The aforementioned camera module can be integrated into a security monitoring camera to acquire images of the monitored area and perform image quality judgment and self-cleaning control. It can be composed of the following components, including but not limited to an image acquisition unit (such as a CMOS image sensor), a piezoelectric cleaning component (such as a piezoelectric ceramic ring made of PZT-4 material), a main control chip (SoC), and an image processing module.

[0023] It is understandable that the aforementioned camera modules can be integrated not only into security monitoring cameras, but also into any type of integrated camera, such as vehicle cameras, ATM monitoring cameras, and street warning cameras. In other words, they can not only be integrated into security monitoring to improve security, but also into any type of camera module to improve the image clarity of current camera modules in complex environments.

[0024] Specifically, it can be widely used in other scenarios equipped with imaging devices, such as automotive front-mounted cameras, environmental perception cameras in advanced driver assistance systems (ADAS), detection modules in industrial vision systems, drone vision modules, as well as smart homes, robot vision, outdoor work terminals and other application scenarios.

[0025] These types of camera modules typically include structures such as lenses, image sensors, and processing chips. Their core requirement is to maintain the stability and clarity of image acquisition quality, especially in outdoor or dynamic environments where they are susceptible to pollutants such as dust, water mist, and oil, and these pollutants are not easily removed by manual cleaning.

[0026] More specifically, the aforementioned image acquisition unit can periodically capture images of external monitoring screens to obtain continuous video frame images. The SoC main control chip can extract representative key frame images from the continuous video frame images according to the key frame acquisition cycle set by the aforementioned camera self-cleaning processing platform, and use them as image data for pollution status assessment.

[0027] The aforementioned image data not only includes the image itself, but may also include extractable image sharpness information, brightness distribution, occlusion morphology, and focus area features, among other multi-dimensional features, for comprehensive judgment of pollution status.

[0028] 102. Based on a preset image recognition algorithm, process the image data to determine the degree of contamination of the camera module.

[0029] In this embodiment of the invention, the preset image recognition algorithm may include, but is not limited to, multiple processing steps such as image sharpness analysis, edge sharpness calculation, occlusion area detection and focus shift evaluation. It should be noted that the preset image is implemented by the algorithm through an embedded vision processing framework or the built-in neural network engine of the main control chip (SoC) to achieve the purpose of feature extraction and pollution level determination of the original image data.

[0030] Specifically, image recognition algorithms can perform Sobel edge detection on acquired images and statistically analyze gradient magnitude changes to derive image sharpness indices. If edges are blurry or the image is out of focus, it indicates that the image is not sharp enough. They can also analyze the deviation between the image brightness histogram distribution and the color mean to detect large areas of obstruction, such as water stains, fog, or foreign objects obstructing the lens. Furthermore, they can use frequency domain Fourier transform to assess the concentration of spectral energy in the image, determining whether focus blur is caused by lens smudges. Finally, the analytical data obtained from the above three steps are comprehensively calculated to determine the corresponding pollution level, such as clean, lightly polluted, moderately polluted, and heavily polluted.

[0031] For example, in one image recognition process, the above-mentioned camera self-cleaning processing platform detected an image clarity score of 43 (below the threshold of 60), the occluded area reached 38% of the total image area, the focus was seriously off, and the image recognition module output the pollution level as "moderate pollution".

[0032] 103. Determine the piezoelectric feedback signal on the current camera module.

[0033] In this embodiment of the invention, the piezoelectric feedback signal can refer to the response of the piezoelectric ceramic component in the camera module to changes in electrical parameters such as voltage, current or impedance due to external loads, such as water stains, dust or other contaminants attached to the lens, when it is under excited vibration or working conditions. It can be understood that the piezoelectric feedback signal can be used to reflect the degree of physical interference of the adhering substances on the lens surface, i.e. the degree of contamination.

[0034] For example, when a water droplet falls onto the lens, the droplet has a certain mass applied to the piezoelectric ceramic component, which generates a change in electrical signal, which is received and detected by the aforementioned camera self-cleaning platform.

[0035] In one possible embodiment, the aforementioned camera self-cleaning platform can assess image clarity based on image recognition algorithms, and can also combine the working response signal of the piezoelectric ceramic component as an auxiliary judgment basis at the physical level. For example, a PZT-4 type piezoelectric ceramic ring is provided in the aforementioned camera module. The ceramic ring is attached to the bottom of the lens. When it is subjected to a small reverse mechanical load generated by the surface deposits of the lens (such as water droplets, fog, dust, etc.), the PZT-4 type piezoelectric ceramic ring will generate a weak electrical signal change (such as voltage, current or impedance change), and the aforementioned camera self-cleaning platform receives the change, which is the piezoelectric feedback signal.

[0036] 104. Based on the piezoelectric feedback signal and the degree of contamination, determine the cleaning control parameters of the current camera module.

[0037] In this embodiment of the invention, the cleaning control parameters may include, but are not limited to, control parameters used to control the lens cleaning action of the cleaning component of the current camera module, such as drive frequency, vibration amplitude, cleaning duration, cleaning mode, number of repetitions, and temperature / power protection limits.

[0038] Specifically, the aforementioned driving frequency can be used to set the operating frequency of the piezoelectric ceramic vibration to match the optimal cleaning efficiency and energy transfer characteristics; the aforementioned vibration amplitude can be used to adjust the vibration intensity of the piezoelectric ceramic, thereby changing its ability to remove deposits (such as water droplets and dust); the aforementioned cleaning duration can be used to control the execution time of the cleaning action, ensuring that the cleaning action achieves the desired effect while avoiding excessive energy consumption; the aforementioned cleaning mode can be used to select a specific cleaning method (such as vibration, heating, or a combination of both) to adapt to different types of mirror contaminants; the aforementioned number of repetitions can be used to enhance the cleaning effect, performing multiple rounds of cleaning operations when a single cleaning is insufficient; the aforementioned temperature / power protection limit can be used to protect the camera module and piezoelectric components, preventing overheating of devices or circuit overload due to continuous high-intensity cleaning.

[0039] In one possible embodiment, the aforementioned camera self-cleaning platform calculates a final contamination score by weighting the contamination level in the determined contamination degree and the degree of change in the piezoelectric feedback signal. Based on the contamination score, multiple cleaning response levels are defined. For example, when the comprehensive score is below a first threshold, the lens is determined to be clean and no cleaning is required; when the score is in the middle range, it is determined to be slightly contaminated and short-term low-frequency vibration cleaning is performed; when the score is high, it is determined to be moderate or heavy contamination, and strategies such as high-frequency long-term vibration, heating cleaning, or multiple rounds of repeated cleaning are adopted.

[0040] 105. Based on the cleaning control parameters, execute the corresponding cleaning action to clean the current camera module.

[0041] In this embodiment of the invention, after generating cleaning control parameters, the camera module self-cleaning platform starts the corresponding cleaning components according to the corresponding cleaning control parameters. For example, it drives the piezoelectric ceramic ring to operate at a set vibration frequency and voltage amplitude, and performs vibration dewatering or heating defogging operations on the lens surface within a set duration, thereby effectively removing water droplets, fog or stains attached to the lens surface.

[0042] More specifically, during the cleaning process, the aforementioned camera module self-cleaning platform can also simultaneously monitor the piezoelectric feedback signal to ensure that the cleaning action is executed precisely according to the predetermined intensity and safety conditions. For example, during the cleaning process, as the stains decrease, the piezoelectric feedback signal will become weaker and weaker, and the cleaning intensity of the corresponding cleaning component can be appropriately reduced. At this time, the aforementioned camera module self-cleaning platform will adjust the cleaning control parameters, such as reducing the drive frequency.

[0043] Through the above methods, the platform achieves closed-loop response control of "identification-judgment-execution", which improves the imaging stability and self-maintenance capability of the camera module in outdoor or high humidity and high pollution environments, reduces the frequency of manual maintenance, and enhances the intelligence and application adaptability of the camera module.

[0044] In this embodiment of the invention, image data currently collected by the camera module is acquired, and the image data is used to assess the contamination status of the camera module. The image data is processed based on a preset image recognition algorithm to determine the degree of contamination of the camera module. The piezoelectric feedback signal on the current camera module is determined. Based on the piezoelectric feedback signal and the degree of contamination, cleaning control parameters for the current camera module are determined. Based on the cleaning control parameters, corresponding cleaning actions are executed to clean the current camera module. Through the above method steps, the degree of contamination on the lens surface can be intelligently judged and the corresponding cleaning behavior can be adaptively controlled and adjusted by combining image recognition results and piezoelectric feedback information, effectively improving cleaning efficiency and control accuracy, and ensuring the imaging quality of the camera equipment.

[0045] Optionally, in the step of acquiring the image data currently collected by the camera module, the image data of the camera module can also be acquired based on preset keyframes to obtain image frame data; and the image frame data can be preprocessed by effective data classification to obtain the image data currently collected by the camera module.

[0046] In this embodiment of the invention, the aforementioned preset keyframe may refer to the image acquisition points predefined by the camera module self-cleaning processing platform according to time intervals, changes in image features, or other conditions, used to extract representative frame images from a continuous image stream as a basis for pollution identification.

[0047] In this embodiment, the aforementioned preset keyframes can be generated in the following two ways: Time-driven type: The above-mentioned camera module self-cleaning platform can extract an image frame as a key frame according to a preset time interval. For example, it can capture the current image every 5 seconds to determine whether the lens is contaminated. The preset time interval can be adaptively adjusted according to environmental factors. When the environmental factors are severe and the camera module is contaminated quickly, the preset time interval can be shortened adaptively.

[0048] Event-driven: The self-cleaning processing platform of the above-mentioned camera module can extract a frame as a key frame based on the instant of change in image features. For example, when the camera detects a sudden change in image brightness (such as encountering car headlights at night) or a decrease in clarity exceeding a threshold through the light sensor or other components, it will automatically mark the frame as a key frame. Or, if a stain suddenly blocks the light, it will be marked as a key frame.

[0049] The aforementioned image frame data can refer to the information of a single frame of image obtained from a single image acquisition. It is usually in matrix format and includes pixel brightness, contrast, and edge details.

[0050] In this embodiment, data cleaning, denoising, normalization, and region segmentation can be performed on the acquired raw image frame data to extract structured data that is beneficial to image contamination analysis, thereby achieving the purpose of effective data classification preprocessing and improving recognition accuracy and response efficiency.

[0051] In one possible embodiment, the aforementioned camera module self-cleaning processing platform extracts image frame data from the image data acquired by the camera module based on preset keyframes. It then removes abnormal and interfering images from the extracted image frame data through operations such as boundary cropping, grayscale normalization, noise reduction, sharpness histogram generation, and brightness feature extraction. This process preserves data regions and parameter features related to contamination characteristics, allowing the self-cleaning processing platform to input the processed image data into a preset image recognition algorithm for subsequent contamination level determination.

[0052] By following the above steps, we can not only reduce the data processing load and improve the response efficiency of pollution identification, but also effectively remove low-quality or abnormal frames, ensuring that the data used for pollution judgment is representative and stable, thereby improving the overall accuracy and intelligence of the camera module self-cleaning control platform.

[0053] Optionally, in the step of processing image data based on a preset image recognition algorithm to determine the degree of contamination of the camera module, the method further includes processing the image data using the preset image recognition algorithm to determine the corresponding sharpness data, occlusion area data, and focus blur data; and determining the current degree of contamination of the camera module based on the corresponding sharpness data, occlusion area data, and focus blur data.

[0054] In this embodiment of the invention, the aforementioned camera module self-cleaning processing platform can process the edge details, brightness changes, structural textures, and other features of the image after acquiring keyframe images using image analysis algorithms to extract quantifiable pollution-related indicators. Generally, this can be accomplished by the image processing module integrated into the SoC or an embedded algorithm module (such as the OpenCV library or a lightweight CNN model). Specifically, it can extract and identify sharpness data, occlusion area data, and focus blur data. For example, the image sharpness can be evaluated by calculating the average amplitude of the image grayscale gradient using the Sobel operator. The more blurred the image, the smoother the edges, and the lower the gradient amplitude. Histogram analysis and connected component segmentation algorithms can also be used to determine whether there are large areas of abnormal brightness in the image (such as water droplet occlusion, stains, foreign objects, etc.). Frequency domain energy analysis (such as Fourier transform) can also be used to determine the proportion of high-frequency components in the image. The less high-frequency content, the more blurred the image.

[0055] This implementation describes the identification of focus blur data. Image frequency domain analysis can be used to extract focus blur data to determine if the image is blurry due to lens contamination, water vapor, or focus failure. Specifically, after acquiring keyframe images, the aforementioned camera module self-cleaning platform converts the image from the spatial domain to the frequency domain and uses the Fast Fourier Transform (FFT) algorithm to calculate the energy distribution of the image at each frequency component. Generally, a normal, clear image has rich high-frequency energy distribution and obvious spectral edge components; while when the image is blurry, out of focus, or has fog on the lens, its high-frequency components will be significantly attenuated, and the frequency domain energy will be concentrated in the low-frequency part.

[0056] For example, after the aforementioned camera module self-cleaning processing platform acquires an image frame, the high-frequency energy ratio of the image in the frequency domain is 18.7%. The aforementioned camera module self-cleaning processing platform compares it with a set threshold (25%) and finds that it is significantly lower, so it marks the image as "focus blur".

[0057] In one possible embodiment, the above-mentioned camera module self-cleaning platform uses a preset image recognition algorithm to identify and process the image data during the process of identifying and judging the degree of contamination of the image data, so as to extract the image's corresponding sharpness data, occlusion area data and focus blur data, and comprehensively evaluate the lens contamination level accordingly.

[0058] Optionally, the step of determining the piezoelectric feedback signal on the current camera module further includes: acquiring the electrical signal change amount on the current camera module based on preset key frames to obtain a continuous electrical signal change dataset; performing electrical signal change feature recognition processing on the continuous electrical signal change dataset to obtain multiple electrical signal change feature data; and determining the piezoelectric feedback signal on the current camera module based on the multiple electrical signal change feature data.

[0059] In this embodiment of the invention, the aforementioned change in electrical signal can refer to the difference between the response voltage, current, or impedance of the piezoelectric ceramic during the excitation process and the previous sampling time. For example, the voltage changes from 134V to 132.5V, with a change of −1.5V.

[0060] The aforementioned continuous electrical signal change dataset can contain multiple continuously changing electrical signals. Specifically, it can refer to an ordered data set consisting of electrical signal change records collected continuously at multiple time points within the excitation period, often a voltage time series or a current waveform series, such as [134V, 132.5V, 131V, …].

[0061] In this embodiment, the camera module self-cleaning processing platform can perform normalization, fitting, spectrum analysis, statistical extraction and other operations on the above continuous electrical signal change dataset to extract key indicators that can characterize the pollution impact characteristics, including but not limited to electrical signal change characteristic data with representative physical characteristics such as "amplitude decrease rate", "average impedance fluctuation" and "number of current pulse responses".

[0062] In another possible embodiment, the above-mentioned camera module self-cleaning platform can also determine the corresponding contamination state by matching the extracted electrical signal change feature data with the feature values ​​in the preset cleaning judgment. For example, if the above-mentioned electrical signal change feature data meets the conditions such as "amplitude attenuation rate > 20% and impedance fluctuation > 12Ω", it is considered that there are attachments or contamination layers on the lens, and a "high load piezoelectric feedback signal" state is generated as an important reference for cleaning trigger judgment.

[0063] Optionally, in the step of determining the cleaning control parameters of the current camera module based on the piezoelectric feedback signal and the degree of contamination, the method further includes determining whether the cleaning triggering conditions are met based on the piezoelectric feedback signal and the degree of contamination; if the cleaning type and cleaning intensity are met, then corresponding control parameters that can perform cleaning treatment on the cleaning type and cleaning intensity are generated.

[0064] In this embodiment of the invention, the cleaning triggering conditions may include, but are not limited to, the physical properties of the stains used to determine whether cleaning should be performed, such as cleaning type and cleaning intensity. These conditions are typically based on a combination of image contamination levels and piezoelectric feedback characteristics. For example, if a preset threshold is reached (e.g., a score greater than 0.6), the triggering condition is met. Specifically, the threshold determination may include image contamination indicators such as sharpness, occlusion area, and image contrast; piezoelectric feedback indicators such as piezoelectric element operating frequency deviation, abnormal capacitance value, and excitation response attenuation; and time factors such as continuous running time exceeding a set threshold. For instance, if the image sharpness is below 0.5 (indicating blurriness), the occlusion area is greater than 20%, and the piezoelectric feedback signal exhibits response hysteresis (delay greater than 300ms), then the cleaning triggering condition can be determined to be met, and the next cleaning action can be initiated.

[0065] The aforementioned cleaning types can refer to the physical properties of pollutants, such as water mist, water droplets, oil film, dust, etc. Generally, they can be classified and judged by image blur features, occlusion patterns, and feedback modes. For example, a large area of ​​low contrast with slight impedance changes can be judged as "fog".

[0066] The cleaning intensity mentioned above can refer to the required cleaning energy or action level, which depends on the firmness of the contaminant adhesion and the size of the contaminant area. Generally, it can be divided into three levels: light, medium and heavy, corresponding to different vibration intensities, heating times and other cleaning actions.

[0067] For example, when the identification result is moderate occlusion, persistent presence, and feedback vibration attenuation, the above-mentioned camera module self-cleaning platform determines that moderate cleaning is required. It then sets the vibration amplitude to 120Vpp, the vibration frequency to 120kHz, and the duration to 1000ms to meet the current cleaning intensity and clean the corresponding stains.

[0068] The aforementioned control parameters may include, but are not limited to, at least one of the following when starting cleaning: the driving frequency of the cleaning component, the vibration amplitude, and the cleaning duration. It should be noted that the control parameters can be adaptively adjusted based on feedback from the cleaning results after cleaning is completed.

[0069] The aforementioned cleaning components can be functional parts that perform cleaning actions, including but not limited to piezoelectric vibration elements, heating elements, film conditioners, etc.

[0070] In one possible embodiment, before generating cleaning control parameters, the camera module self-cleaning platform performs a fusion analysis of the current image contamination status and piezoelectric feedback status to determine whether the cleaning triggering conditions are met, and determines the cleaning type and intensity to be performed accordingly, and further generates matching cleaning control parameters.

[0071] By using the above methods and steps, we can dynamically decide on cleaning strategies for different types of contamination, such as fog, water droplets, dust, or oil stains. This not only improves cleaning efficiency and imaging quality but also reduces energy consumption and the number of times cleaning devices work ineffectively.

[0072] Optionally, in the step of performing corresponding cleaning actions to clean the current camera module based on cleaning control parameters, the corresponding cleaning component can also be driven based on the cleaning control parameters to perform vibration dewatering or heating defogging cleaning actions on the camera module during the cleaning duration with adjusted drive frequency and vibration amplitude.

[0073] In this embodiment of the invention, after the camera module self-cleaning platform determines that cleaning is triggered based on cleaning control parameters, it can drive the cleaning components on the camera module to perform cleaning actions. Specifically, it can be based on, for example... Figure 2 The diagram illustrates the structure of a camera module. An image sensor chip captures the current image from the lens, generating image data, which is then transmitted to a SOC chip for processing via a communication interface. The SOC chip incorporates an image recognition algorithm that identifies blurred areas, occluded areas, and decreased sharpness in the image data. It also combines this with historical image trends to determine the degree of lens contamination. A piezoelectric ceramic sheet is incorporated into the camera module, adhering to the back of a lens (e.g., a G1 lens). The SOC chip applies a driving signal to the piezoelectric sheet via a power control circuit, causing it to vibrate. Simultaneously, it acquires the reverse piezoelectric signal (feedback voltage) in real time to assess the damping or adhesion between the piezoelectric structure and the lens caused by contaminants such as water droplets and dust, thereby obtaining dynamic contamination feedback. Then, cleaning control parameters are generated. The SOC chip matches the image recognition result with the piezoelectric feedback signal as input to determine whether to trigger the cleaning operation. If the cleaning triggering conditions are met (such as severe occlusion + decrease in piezoelectric response amplitude), the required cleaning control parameters are automatically calculated, such as vibration frequency (e.g., 38kHz), vibration amplitude (e.g., 6μm), and cleaning duration (e.g., 1.5s), and output as control signals. Finally, the aforementioned SOC chip controls the piezoelectric ceramic sheet to vibrate at high frequency, and through resonance, the droplets or dust adhering to the lens surface are quickly detached. Combined with the epoxy adhesive bonding structure of the G1 lens, a good energy transfer path can be formed, improving the cleaning effect.

[0074] By using the above methods and steps, the state of contamination can be accurately determined, and a matching cleaning plan can be generated adaptively to avoid ineffective cleaning or false triggering. At the same time, the cleaning efficiency can be improved by leveraging structural resonance, which significantly improves the reliability and image quality of the camera in outdoor, humid or contaminated environments.

[0075] Optionally, in the step after performing the corresponding cleaning action on the current camera module based on the cleaning control parameters, the updated image data and real-time piezoelectric feedback signal of the current camera module after the cleaning action are completed can be obtained; the image data of the camera module is re-evaluated based on the updated image data, and the cleaning effect of the current camera module after the cleaning action is completed is determined in combination with the real-time piezoelectric feedback signal; if the preset cleaning threshold is not reached, the cleaning control parameters corresponding to the cleaning component are adaptively adjusted, and the corresponding cleaning action is re-executed until the cleaning effect of the current camera module after the cleaning action is completed reaches the preset cleaning threshold; if the preset cleaning threshold is reached, the cleaning action is stopped.

[0076] In this embodiment of the invention, the aforementioned updated image data can be acquired immediately after the cleaning action is completed, re-collecting image data from the current lens field of view. This image data reflects the visual state of the lens module after cleaning and can be used to determine whether the image has become clear and clean. For example, if the original image contains water mist or stains, whether these interfering areas are eliminated in the updated image after cleaning is an important criterion for judging the cleaning effect.

[0077] The aforementioned real-time piezoelectric feedback signal can be the feedback voltage of the piezoelectric ceramic sheet after the cleaning action is synchronously acquired. This voltage reflects the resonance response intensity of the ceramic sheet after excitation. If the droplets or dirt on the mirror surface are not removed, it will hinder the normal vibration of the ceramic sheet, resulting in abnormal waveform or low amplitude of the feedback signal. Conversely, when the mirror surface is clean, the vibration response of the piezoelectric sheet is clear and stable.

[0078] In this embodiment, the aforementioned camera module self-cleaning platform will re-analyze the updated image data in terms of image clarity, edge sharpness, focus quality, etc., and compare the results with the image data before cleaning to determine whether cleaning has restored the image to a normal visual effect. At the same time, the image evaluation result is combined with the real-time piezoelectric feedback signal to comprehensively determine whether the set standard has been met.

[0079] The aforementioned preset cleaning threshold can be a preset quantitative threshold for judging the cleaning effect, which may include, but is not limited to, image sharpness indicators (such as edge gradient greater than a certain value), focus offset less than a certain tolerance range, and stable amplitude of feedback piezoelectric signal higher than a certain voltage threshold. Generally speaking, when multiple indicators meet the cleaning threshold at the same time, the cleaning is considered complete.

[0080] In another possible embodiment, if the preset cleaning threshold is not reached, the control parameters will be automatically adjusted (such as increasing the vibration amplitude, extending the cleaning time, etc.), and the cleaning action will be re-executed. The adjustment can be based on a rule base or dynamically optimized through a feedback closed-loop control algorithm to ensure efficient cleaning can be completed under various pollution conditions.

[0081] By using the above methods and steps, "updated image data" and "real-time piezoelectric feedback signals" can be obtained and "re-evaluated" accordingly, achieving closed-loop verification and multiple rounds of optimization of the cleaning effect. This ensures that cleaning is only performed under necessary conditions, improving stability, image quality, and energy efficiency.

[0082] like Figure 3 As shown, this embodiment of the invention also provides a self-cleaning control device 300 for a camera module. The online interactive device 300 includes: The first acquisition module 301 is used to acquire the image data currently collected by the camera module, and the image data is used to evaluate the pollution status of the camera module; The first determining module 302 is used to process the image data based on a preset image recognition algorithm to determine the degree of contamination of the camera module; The second determining module 303 is used to determine the piezoelectric feedback signal on the current camera module; The third determining module 304 is used to determine the cleaning control parameters of the current camera module based on the piezoelectric feedback signal and the degree of contamination. The first control module 305 is used to perform a corresponding cleaning action to clean the current camera module based on the cleaning control parameters.

[0083] Optionally, the first acquisition module 301 mentioned above includes: The first acquisition submodule is used to acquire image data from the camera module based on preset key frames to obtain image frame data; The first processing submodule is used to perform effective data classification preprocessing on the image frame data to obtain the image data currently acquired by the camera module.

[0084] Optionally, the first determining module 302 mentioned above includes: The first determining submodule is used to identify and process the image data through the preset image recognition algorithm to determine the sharpness data, occlusion area data and focus blur data corresponding to the image data; The second determining submodule is used to determine the degree of contamination of the current camera module based on the sharpness data, occlusion area data, and focus blur data corresponding to the image data.

[0085] Optionally, the second determining module 303 mentioned above includes: The second acquisition submodule is used to collect the electrical signal change on the current camera module based on a preset key frame to obtain a continuous electrical signal change dataset, which contains multiple continuously changing electrical signals. The second processing submodule is used to perform electrical signal change feature recognition processing on the continuous electrical signal change dataset to obtain multiple electrical signal change feature data. The third determining submodule is used to determine the piezoelectric feedback signal on the current camera module based on the multiple electrical signal change characteristic data.

[0086] Optionally, the third determining module 304 mentioned above includes: The fourth determining submodule is used to determine whether the cleaning triggering conditions are met based on the piezoelectric feedback signal and the degree of contamination. The cleaning triggering conditions include cleaning type and cleaning intensity. The first generation submodule is used to generate corresponding control parameters that can perform cleaning treatment on the cleaning type and cleaning intensity if the cleaning type and cleaning intensity are satisfied. The control parameters include at least one of the driving frequency of the cleaning component, the vibration amplitude and the cleaning duration during cleaning.

[0087] Optionally, the first control module 305 mentioned above includes: The adjustment submodule is used to drive the corresponding cleaning component based on the cleaning control parameters, and to perform vibration dewatering or heating defogging cleaning actions on the camera module during the cleaning duration with the adjusted drive frequency and vibration amplitude.

[0088] Optionally, the above-mentioned device further includes: The second acquisition module is used to acquire the updated image data and real-time piezoelectric feedback signal of the camera module after it has completed the cleaning action. The evaluation module is used to re-evaluate the image data of the camera module based on the updated image data, and determine whether the cleaning effect of the current camera module after the cleaning action has reached the preset cleaning threshold by combining the real-time piezoelectric feedback signal. The cleaning module is used to adaptively adjust the cleaning control parameters corresponding to the cleaning component and re-execute the corresponding cleaning action if the preset cleaning threshold is not reached, until the cleaning effect of the current camera module after the cleaning action is completed reaches the preset cleaning threshold. The stop module is used to stop the cleaning action if the preset cleaning threshold is reached.

[0089] like Figure 4 As shown, this embodiment of the invention also provides an electronic device 400, including a processor, which can execute any of the above-described self-cleaning control methods for camera modules.

[0090] Specifically, it includes a processor 401 and a memory 402, as well as a computer program stored in the memory 402 and capable of running on the processor 401, which executes the self-cleaning control method for the camera module, wherein: The processor 401 executes the calculator program for the self-cleaning control method of the camera module stored in the memory 402, and performs the following steps: Obtain the image data currently acquired by the camera module; The image data is processed based on a preset image recognition algorithm to determine the degree of contamination of the camera module; Determine the piezoelectric feedback signal on the current camera module; Based on the piezoelectric feedback signal and the degree of contamination, the cleaning control parameters of the current camera module are determined; Based on the cleaning control parameters, the corresponding cleaning action is executed to clean the current camera module.

[0091] Optionally, the processor 401 performs the step of acquiring the image data currently collected by the camera module, including: Based on preset keyframes, image data of the camera module is acquired to obtain image frame data; The image frame data is subjected to effective data classification preprocessing to obtain the image data currently acquired by the camera module.

[0092] Optionally, the processor 401 executes the processing of the image data based on a preset image recognition algorithm to determine the degree of contamination of the camera module, including: The image data is processed by the preset image recognition algorithm to determine the sharpness data, occlusion area data and focus blur data corresponding to the image data; Based on the image data, including sharpness data, occlusion area data, and focus blur data, the degree of contamination of the current camera module is determined.

[0093] Optionally, the processor 401 performs the process of determining the piezoelectric feedback signal on the current camera module, including: Based on preset keyframes, the changes in electrical signals on the current camera module are collected to obtain a continuous electrical signal change dataset, which contains multiple continuously changing electrical signals. The continuous electrical signal variation dataset is subjected to electrical signal variation feature recognition processing to obtain multiple electrical signal variation feature data; Based on the multiple electrical signal change characteristic data, the piezoelectric feedback signal on the current camera module is determined.

[0094] Optionally, the processor 401 further performs the step of determining the cleaning control parameters of the current camera module based on the piezoelectric feedback signal and the degree of contamination, including: Based on the piezoelectric feedback signal and the degree of contamination, it is determined whether the cleaning triggering conditions are met, wherein the cleaning triggering conditions include cleaning type and cleaning intensity; If the cleaning type and cleaning intensity are satisfied, corresponding control parameters are generated to perform cleaning treatment on the cleaning type and cleaning intensity. The control parameters include at least one of the following during cleaning: the driving frequency of the cleaning component, the vibration amplitude, and the cleaning duration.

[0095] Optionally, the processor 401 further performs the cleaning action based on the cleaning control parameters to clean the current camera module, including: Based on the cleaning control parameters, the corresponding cleaning components are driven to perform vibration dewatering or heating defogging cleaning actions on the camera module during the cleaning duration with adjusted driving frequency and vibration amplitude.

[0096] Optionally, after the processor 401 executes the cleaning action based on the cleaning control parameters to perform the corresponding cleaning action on the current camera module, the method further includes: Acquire updated image data and real-time piezoelectric feedback signal of the camera module after it has completed the cleaning action; The image data of the camera module is re-evaluated based on the updated image data, and the cleaning effect of the current camera module after the cleaning action is determined by combining the real-time piezoelectric feedback signal. If the preset cleaning threshold is not reached, the cleaning control parameters corresponding to the cleaning component are adaptively adjusted, and the corresponding cleaning action is re-executed until the cleaning effect of the current camera module after the cleaning action is completed reaches the preset cleaning threshold. If the preset cleaning threshold is reached, the cleaning process will stop.

[0097] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the self-cleaning control method for the camera module or the self-cleaning control method for the application-side camera module provided in this invention, and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0098] Those skilled in the art will understand that implementing all or part of the processes in the above embodiments can be done by a computer program instructing related hardware, and can be stored in a computer-readable storage medium. When executed, the program can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0099] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A self-cleaning control method for a camera module, characterized in that, include: Obtain the image data currently acquired by the camera module; The image data is processed based on a preset image recognition algorithm to determine the degree of contamination of the camera module; Determine the piezoelectric feedback signal on the current camera module; Based on the piezoelectric feedback signal and the degree of contamination, the cleaning control parameters of the current camera module are determined; Based on the cleaning control parameters, the corresponding cleaning action is executed to clean the current camera module.

2. The self-cleaning control method for a camera module as described in claim 1, characterized in that, The step of acquiring the image data currently collected by the camera module includes: Based on preset keyframes, image data of the camera module is acquired to obtain image frame data; The image frame data is subjected to effective data classification preprocessing to obtain the image data currently acquired by the camera module.

3. The self-cleaning control method for a camera module as described in claim 1, characterized in that, The step of processing the image data based on a preset image recognition algorithm to determine the degree of contamination of the camera module includes: The image data is processed by the preset image recognition algorithm to determine the sharpness data, occlusion area data and focus blur data corresponding to the image data; Based on the image data, including sharpness data, occlusion area data, and focus blur data, the degree of contamination of the current camera module is determined.

4. The self-cleaning control method for a camera module as described in claim 2, characterized in that, The determination of the piezoelectric feedback signal on the current camera module includes: Based on preset keyframes, the changes in electrical signals on the current camera module are collected to obtain a continuous electrical signal change dataset, which contains multiple continuously changing electrical signals. The continuous electrical signal variation dataset is subjected to electrical signal variation feature recognition processing to obtain multiple electrical signal variation feature data; Based on the multiple electrical signal change characteristic data, the piezoelectric feedback signal on the current camera module is determined.

5. The self-cleaning control method for a camera module as described in claim 1, characterized in that, The step of determining the cleaning control parameters of the current camera module based on the piezoelectric feedback signal and the degree of contamination includes: Based on the piezoelectric feedback signal and the degree of contamination, it is determined whether the cleaning triggering conditions are met, wherein the cleaning triggering conditions include cleaning type and cleaning intensity; If the cleaning type and cleaning intensity are satisfied, corresponding control parameters are generated to perform cleaning treatment on the cleaning type and cleaning intensity. The control parameters include at least one of the following during cleaning: the driving frequency of the cleaning component, the vibration amplitude, and the cleaning duration.

6. The self-cleaning control method for a camera module as described in claim 1, characterized in that, The step of performing a cleaning action on the current camera module based on the cleaning control parameters includes: Based on the cleaning control parameters, the corresponding cleaning components are driven to perform vibration dewatering or heating defogging cleaning actions on the camera module during the cleaning duration with adjusted driving frequency and vibration amplitude.

7. The self-cleaning control method for a camera module as described in claim 6, characterized in that, After performing the corresponding cleaning action on the current camera module based on the cleaning control parameters, the method further includes: Acquire updated image data and real-time piezoelectric feedback signal of the camera module after it has completed the cleaning action; The image data of the camera module is re-evaluated based on the updated image data, and the cleaning effect of the current camera module after the cleaning action is determined by combining the real-time piezoelectric feedback signal. If the preset cleaning threshold is not reached, the cleaning control parameters corresponding to the cleaning component are adaptively adjusted, and the corresponding cleaning action is re-executed until the cleaning effect of the current camera module after the cleaning action is completed reaches the preset cleaning threshold. If the preset cleaning threshold is reached, the cleaning process will stop.

8. A self-cleaning control device for a camera module, characterized in that, include: The first acquisition module is used to acquire the image data currently collected by the camera module, and the image data is used to assess the contamination status of the camera module; The first determining module is used to process the image data based on a preset image recognition algorithm to determine the degree of contamination of the camera module; The second determining module is used to determine the piezoelectric feedback signal on the current camera module; The third determining module is used to determine the cleaning control parameters of the current camera module based on the piezoelectric feedback signal and the degree of contamination. The first control module is used to perform a corresponding cleaning action to clean the current camera module based on the cleaning control parameters.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the self-cleaning control method for a camera module as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the self-cleaning control method for a camera module as described in any one of claims 1 to 7.

Images