Smart home camera and rapid adjustment and maintenance method thereof

Abstract

The application discloses an intelligent home camera and a quick adjustment and maintenance method thereof, relates to the technical field of camera maintenance, and inputs the state of key components into a fault diagnosis and grading model; when the fault diagnosis and grading model judges that the performance of any key component state detection reaches a set threshold or a fault, a maintenance suggestion action is triggered; a grading maintenance suggestion is pushed to a user through a mobile phone APP; the grading maintenance suggestion comprises problem description, confidence and predicted repair time; the application realizes the automation, intelligence and high efficiency of intelligent home camera maintenance, and effectively solves the defects of the existing maintenance mode. Through a dynamically adjusted self-checking period, different use frequency scenes can be adapted, the state of key components can be detected in time, potential faults can be found in advance, and the expansion of faults is avoided to affect use.

Description

Technical Field

[0001] This invention relates to the field of camera maintenance technology, and in particular to a smart home camera and its rapid adjustment and maintenance method. Background Technology

[0002] With the popularization of smart home technology, smart home cameras have been widely used in scenarios such as home security and elderly and child care. Their operational stability and functional reliability directly affect the user experience and security effectiveness. At present, the maintenance of smart home cameras on the market mostly relies on manual triggering by users or passive handling after a failure occurs. They lack proactive and regular self-inspection mechanisms, making it impossible to detect component performance degradation or potential faults in a timely manner. Often, the fault is only noticed by users after it has escalated, affecting the continuity of monitoring.

[0003] The self-test cycle of existing cameras is mostly set to a fixed value, which cannot be dynamically adjusted according to the actual usage frequency. In scenarios with frequent use, the self-test is not timely, while in scenarios with less frequent use, it results in wasted resources. At the same time, the fault diagnosis lacks a hierarchical mechanism, making it difficult for users to quickly determine the severity of the fault and the priority of handling. Moreover, the maintenance process often requires the intervention of professional personnel or users with certain professional knowledge, which is complicated and time-consuming.

[0004] Furthermore, existing cameras lack effective self-maintenance capabilities. They cannot automatically repair common issues such as lens smudges and minor software malfunctions, requiring users to manually disassemble and clean them or restore them to factory settings. This not only increases the user's workload but may also damage the device due to improper operation. At the same time, maintenance data cannot be effectively reused, and fault diagnosis and repair algorithms are difficult to continuously optimize, making it difficult to improve maintenance efficiency and accuracy over long-term use. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a smart home camera and a method for its rapid adjustment and maintenance. The technical solution is as follows:

[0006] A quick adjustment and maintenance method for smart home cameras includes the following steps:

[0007] Step 1: The camera's built-in self-test program checks the status of key components at set intervals. The status of key components includes image sensor clarity, infrared night vision function, audio input quality, and network connection stability.

[0008] Step 2: Input the status of key components into the fault diagnosis and classification model. When the fault diagnosis and classification model determines that the status of any key component has detected a performance degradation that reaches a set threshold or a fault, it triggers a maintenance recommendation action and pushes a graded maintenance recommendation to the user through the mobile APP. The graded maintenance recommendation includes a problem description, confidence level, and estimated repair time.

[0009] Step 3: After the user selects the self-maintenance mode, the camera automatically enters the self-maintenance program; the self-maintenance program includes a strong vibration step and a software repair step executed sequentially.

[0010] The strong vibration step involves activating the strong vibration module of the smart home camera, which then runs a set number of times at set vibration parameter intervals.

[0011] The software repair steps involve starting the corresponding software repair program to complete the self-repair process based on the specific performance degradation or malfunction issue identified in step 2.

[0012] Step 4: After the self-maintenance procedure is completed, repeat steps 1 and 2. If the status detection results of all key components do not trigger maintenance suggestion actions, the automatic maintenance is considered complete, and the self-maintenance results are pushed to the user via the mobile APP.

[0013] Optionally, in step 1, the self-test cycle is dynamically adjusted based on the camera's historical usage frequency; the higher the usage frequency, the shorter the self-test cycle. The detection of key component status specifically includes image sensor clarity detection, infrared night vision function detection, audio input quality detection, and network connection stability detection.

[0014] Image sensor sharpness detection involves analyzing the image of a fixed reference object at a preset distance in front of the camera and calculating the modulation transfer function value of its edges.

[0015] Infrared night vision function detection is performed in low-light environments by comparing the brightness histogram and noise level of images before and after the infrared lights are turned on to determine the working status of the infrared light group and the photosensitive element.

[0016] Audio input quality testing involves playing a built-in standard frequency test tone and analyzing the signal-to-noise ratio, total harmonic distortion, and frequency response curve of the microphone received signal.

[0017] Network connectivity stability testing involves continuously measuring the round-trip latency, bandwidth utilization, and packet loss rate of data packets between the network and a cloud server or a specified device within a local area network.

[0018] Optionally, in step 2, the tiered maintenance recommendations are divided into three levels:

[0019] The first level is advisory maintenance, corresponding to performance degradation not exceeding the first threshold and confidence level below 70%. Users are advised to observe or schedule non-emergency maintenance.

[0020] The second level is recommended maintenance, which corresponds to performance degradation exceeding the first threshold but not exceeding the second threshold, or a confidence level between 70% and 90%. Users are advised to perform self-maintenance under the guidance of the APP.

[0021] The third level is a mandatory maintenance or fault alarm, corresponding to the detection of a functional failure, performance degradation exceeding the second threshold, or a confidence level higher than 90%. Immediate action is strongly recommended, and it indicates that core monitoring functions are affected.

[0022] The push interface of the mobile APP uses color codes to distinguish levels and displays simulation diagrams and performance curves of specific faulty components.

[0023] Optionally, in step 3, the strong vibration step specifically includes:

[0024] The high-intensity vibration module includes multiple miniature eccentric wheel vibration motors;

[0025] The set vibration parameters include amplitude, frequency, and duration, where the amplitude is 0.5 mm to 1 mm and the frequency is 80 Hz to 120 Hz.

[0026] The interval operation is set as follows: one cycle consists of 5-10 seconds of vibration followed by 5-10 seconds of rest, and this cycle is repeated 3 to 5 times.

[0027] During the strong vibration step, the camera image acquisition and uploading function is paused.

[0028] Optional, the software repair steps specifically include:

[0029] If the problem is a decrease in image sharpness, the following actions will be initiated and executed sequentially: a deep learning-based image deblurring algorithm will be used to process the current video stream; an adaptive image denoising and sharpening algorithm will be optimized; and edge enhancement parameters in the image signal processor will be adjusted.

[0030] If the problem is an unstable network connection, then the following actions will be initiated and executed in sequence: automatically switch to a backup wireless access point or frequency band, dynamically adjust the video encoding bitrate and keyframe interval, and optimize the buffer and retransmission mechanism parameters of the network protocol stack.

[0031] Optionally, in step 3, between the strong vibration step and the software repair step, a surface cleanliness detection step is also included:

[0032] After the strong vibration step is completed, images of the current environment are acquired; the acquired images are analyzed using a trained convolutional neural network model to determine whether there are stubborn stains or condensation on the lens surface.

[0033] If the presence of foreign objects is detected, an enhanced cleaning reminder will be sent to the user via the mobile app, and sound waves of a specific frequency will be played to drive away the foreign objects adhering to the lens surface.

[0034] Optionally, the removal of deposits from the lens surface by playing sound waves of a specific frequency is achieved based on a sound wave emitting module. The sound wave emitting module includes a piezoelectric ceramic sheet and a core processing unit integrated into the smart home camera, which is installed close to the outer ring of the camera lens or the inside of the protective cover. The core processing unit has a built-in sound wave ion driving program, which controls the piezoelectric ceramic sheet to remove deposits from the lens surface using sound waves.

[0035] The specific method for using sound waves to remove contaminants from the lens surface is as follows: The core processing unit initiates the sound wave ion-removing program, sending a control signal to the drive circuit of the sound wave emission module; the drive circuit generates an AC voltage with a set frequency and waveform according to the control signal, driving the piezoelectric ceramic sheet to produce mechanical vibration, radiating sound waves into the air directly in front of the lens; one or more specific frequencies within the range of 35kHz to 45kHz have a repelling or interference effect on small insects, and can also be effectively absorbed by the surface of water droplets and induce micro-vibrations. Each sound wave emission lasts for 1 to 3 seconds, and is repeated 2 to 4 times depending on the degree of lens contamination or ambient temperature; image acquisition is paused during sound wave emission.

[0036] After the sound wave emission action ends, images are collected again. By comparing the changes in image features before and after the removal, the removal effect is evaluated. If the characteristics of the attached substances are not eliminated, clear physical cleaning instructions are pushed to the user via mobile APP.

[0037] Optionally, in step 4, the condition for determining whether automatic maintenance is complete is:

[0038] After repeating step 1, the detected values ​​of all critical component statuses returned to the normal range; and the AI ​​model assessment in step 2 determined that this would not affect subsequent continuous and reliable operation.

[0039] The self-maintenance results pushed to users via mobile app include charts comparing key performance parameters before and after maintenance, a list of specific steps used in the maintenance, and the overall health score of the current device.

[0040] Optionally, step 5 is also included, where each complete self-maintenance process, including the triggering cause, the strong vibration and software repair steps performed, and the final verification results, is anonymized and uploaded to the cloud maintenance knowledge base;

[0041] The cloud server utilizes massive amounts of maintenance case data to regularly optimize the parameters of the fault diagnosis and grading model in step 2 and the software repair algorithm in step 3, and pushes them to all smart home cameras via online firmware upgrades.

[0042] A smart home camera is disclosed, which enables a rapid adjustment and maintenance method for the smart home camera. The camera includes a camera body, a core processing unit, a strong vibration module, and a sound wave emission module. The camera body incorporates an image sensor, an infrared fill light, a microphone, and a network communication module. The core processing unit is communicatively connected to the camera body and is used to run self-test programs, trigger logic, and software repair programs.

[0043] The strong vibration module is fixed inside the camera body and includes at least one miniature eccentric wheel vibration motor for performing the strong vibration step;

[0044] The acoustic wave emitting module consists of at least one piezoelectric ceramic transducer. The transducer is mounted in the annular groove between the outer ring of the camera lens and the protective cover via a ring-shaped flexible circuit board, or directly attached to the inner surface of the protective cover. The vibration plane is parallel to the lens surface and is used to radiate sound waves in front of the lens. The core processing unit controls the execution actions of the strong vibration module and the acoustic wave emitting module respectively.

[0045] In summary, the present invention has at least one of the following beneficial technical effects:

[0046] This invention provides a smart home camera and its rapid adjustment and maintenance method, achieving automation, intelligence, and efficiency in smart home camera maintenance, effectively addressing the shortcomings of existing maintenance methods. Through a dynamically adjusted self-inspection cycle, it can adapt to scenarios with different usage frequencies, ensuring that the status of key components is detected in a timely manner, identifying potential faults early, and preventing the escalation of faults from affecting usability.

[0047] The fault diagnosis and grading model, combined with tiered maintenance recommendations, allows users to clearly understand the severity of the fault, the priority of handling, and the estimated time required. No specialized knowledge is needed to determine the appropriate course of action, lowering the barrier to entry for users. The self-maintenance program, through steps such as strong vibration, software repair, and acoustic wave removal, can autonomously resolve common problems such as lens contamination, minor software malfunctions, and network instability, reducing the need for manual maintenance and improving maintenance efficiency.

[0048] Uploading maintenance data to the cloud and using it for model and algorithm optimization can continuously improve fault diagnosis accuracy and self-repair success rate, enabling iterative upgrades to equipment maintenance capabilities.

[0049] The overall solution can improve the stability and lifespan of cameras, reduce downtime due to malfunctions, optimize the user experience, and reduce equipment maintenance costs and manual intervention costs. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating a method for quickly adjusting and maintaining a smart home camera according to the present invention;

[0051] Figure 2This is a schematic diagram illustrating the electrical component connection principle of a smart home camera according to the present invention;

[0052] Figure 3 This is a schematic diagram illustrating the structural principle of a smart home camera according to the present invention.

[0053] Figure 4 This is a side-view structural principle diagram of a smart home camera according to the present invention.

[0054] Explanation of reference numerals in the attached diagram: 1. Camera body; 2. Core processing unit; 3. Strong vibration module; 4. Sound wave emission module. Detailed Implementation

[0055] The present invention will be further described in detail below with reference to the accompanying drawings.

[0056] This invention discloses a smart home camera and a method for its rapid adjustment and maintenance.

[0057] Reference Figures 1-4 Example 1: A method for quick adjustment and maintenance of a smart home camera, comprising the following steps:

[0058] Step 1: The camera's built-in self-test program checks the status of key components at set intervals. The status of key components includes image sensor clarity, infrared night vision function, audio input quality, and network connection stability.

[0059] Step 2: Input the status of key components into the fault diagnosis and classification model. When the fault diagnosis and classification model determines that the status of any key component has detected a performance degradation that reaches a set threshold or a fault, it triggers a maintenance recommendation action and pushes a graded maintenance recommendation to the user through the mobile APP. The graded maintenance recommendation includes a problem description, confidence level, and estimated repair time.

[0060] Step 3: After the user selects the self-maintenance mode, the camera automatically enters the self-maintenance program; the self-maintenance program includes a strong vibration step and a software repair step executed sequentially.

[0061] The strong vibration step involves activating the strong vibration module 3 of the smart home camera, which then runs a set number of times at set vibration parameter intervals.

[0062] The software repair steps involve starting the corresponding software repair program to complete the self-repair process based on the specific performance degradation or malfunction issue identified in step 2.

[0063] Step 4: After the self-maintenance procedure is completed, repeat steps 1 and 2. If the status detection results of all key components do not trigger maintenance suggestion actions, the automatic maintenance is considered complete, and the self-maintenance results are pushed to the user via the mobile APP.

[0064] In Example 2, step 1, the set cycle is dynamically adjusted based on the camera's historical usage frequency; the higher the usage frequency, the shorter the self-test cycle. The detection of key component status specifically includes image sensor clarity detection, infrared night vision function detection, audio input quality detection, and network connection stability detection.

[0065] Image sensor sharpness detection involves analyzing the image of a fixed reference object at a preset distance in front of the camera and calculating the modulation transfer function value of its edges.

[0066] Infrared night vision function detection is performed in low-light environments by comparing the brightness histogram and noise level of images before and after the infrared lights are turned on to determine the working status of the infrared light group and the photosensitive element.

[0067] Audio input quality testing involves playing a built-in standard frequency test tone and analyzing the signal-to-noise ratio, total harmonic distortion, and frequency response curve of the microphone received signal.

[0068] Network connectivity stability testing involves continuously measuring the round-trip latency, bandwidth utilization, and packet loss rate of data packets between the network and a cloud server or a specified device within a local area network.

[0069] By adopting the above technical solution, the self-test logic in step 1 is based on the correlation between component wear and usage frequency. The set cycle is dynamically adjusted according to the historical usage frequency. The more frequently the components are used, the higher the risk of component wear. Shortening the self-test cycle can capture performance changes in a timely manner. Key component detection adopts targeted quantitative judgment methods. Image sensor sharpness is quantitatively evaluated by analyzing the edge modulation transfer function value of a fixed reference image. Infrared night vision function judges the cooperative working status of infrared lamp group and photosensitive element by comparing the image brightness histogram and noise level before and after the infrared lamp is turned on and off in low light environment. Audio input quality is analyzed by the received signal of the built-in standard frequency test tone, and the sound acquisition effect is quantified by the signal-to-noise ratio, total harmonic distortion, and frequency response curve. Network connection stability is judged by continuously monitoring the round-trip latency of data packets, bandwidth utilization, and packet loss rate to determine the smoothness and reliability of network transmission.

[0070] Step 2 uses a fault diagnosis and grading model to take the component status data obtained from self-inspection as input and compare it with preset performance thresholds to determine whether the component has performance degradation or faults. At the same time, it quantifies the confidence of the problem and the repair time, and generates graded maintenance suggestions based on this and pushes them through a mobile APP so that users can intuitively understand the problem situation and the direction of handling.

[0071] The core of the autonomous maintenance mode in step 3 is to design corresponding repair logic for common self-healing problems. The strong vibration step runs at intervals according to set parameters through a miniature strong vibration module, using vibration to remove the slight dust and impurities attached to the lens surface, laying the foundation for image quality restoration. The software repair step matches the corresponding repair program according to the fault diagnosis results, and specifically solves problems such as image clarity and network connection caused by abnormal software parameters, achieving precise repair.

[0072] The loop verification logic in step 4 involves repeating the self-check and diagnosis process after self-maintenance is completed to confirm whether the component performance has returned to the normal range, ensuring that the repair effect meets the standards, and pushing the maintenance results to the user to form a complete maintenance closed loop and ensure the stable operation of the equipment in the future.

[0073] Example 3, in step 2, the tiered maintenance recommendations are divided into three levels:

[0074] The first level is advisory maintenance, corresponding to performance degradation not exceeding the first threshold and confidence level below 70%. Users are advised to observe or schedule non-emergency maintenance.

[0075] The second level is recommended maintenance, which corresponds to performance degradation exceeding the first threshold but not exceeding the second threshold, or a confidence level between 70% and 90%. Users are advised to perform self-maintenance under the guidance of the APP.

[0076] The third level is a mandatory maintenance or fault alarm, corresponding to the detection of a functional failure, performance degradation exceeding the second threshold, or a confidence level higher than 90%. Immediate action is strongly recommended, and it indicates that core monitoring functions are affected.

[0077] The push interface of the mobile APP uses color codes to distinguish levels and displays simulation diagrams and performance curves of specific faulty components.

[0078] By adopting the above technical solution, based on the quantitative differentiation of fault severity and processing priority, and combined with the dual indicators of performance degradation threshold and confidence level, precise adaptation of maintenance recommendations is achieved, balancing user ease of operation and equipment operational safety. The hierarchical logic aligns with the actual impact range of component faults, classifying them into three levels according to problem severity to ensure that problems of different degrees are addressed accordingly.

[0079] Level 1 advisory maintenance addresses minor performance fluctuations. At this stage, the risk of component failure is low, and immediate action is not required. Users are advised to observe or schedule non-emergency maintenance to avoid unnecessary operations and reduce resource waste. Level 2 recommended maintenance addresses moderate performance degradation or medium-confidence faults. These issues can be resolved by the user themselves. It is recommended to perform self-maintenance guided by the app, which reduces user difficulty and resolves problems promptly, preventing further performance degradation. Level 3 mandatory maintenance or fault alarms address severe faults or high-confidence issues. These problems have already affected or are about to affect the core monitoring functions of the camera and must be addressed immediately. Strong alerts ensure user attention and prevent security vulnerabilities or equipment damage due to delayed handling.

[0080] The mobile app push interface uses color codes to distinguish levels, allowing users to quickly and intuitively identify maintenance priorities and judge the severity of problems without having to read the text carefully. It displays simulated diagrams of faulty components and performance curves to visualize abstract performance data and fault locations, helping users clearly understand the specific situation of the fault and providing an intuitive reference for whether to choose self-maintenance or seek professional help, further reducing the cognitive threshold for users to maintain equipment.

[0081] In Example 4, step 3, the strong vibration step specifically includes:

[0082] The high-intensity vibration module 3 includes multiple miniature eccentric wheel vibration motors;

[0083] The set vibration parameters include amplitude, frequency, and duration, where the amplitude is 0.5 mm to 1 mm and the frequency is 80 Hz to 120 Hz.

[0084] The interval operation is set as follows: one cycle consists of 5-10 seconds of vibration followed by 5-10 seconds of rest, and this cycle is repeated 3 to 5 times.

[0085] During the strong vibration step, the camera image acquisition and uploading function is paused.

[0086] Example 5, the software repair steps specifically include:

[0087] If the problem is a decrease in image sharpness, the following actions will be initiated and executed sequentially: a deep learning-based image deblurring algorithm will be used to process the current video stream; an adaptive image denoising and sharpening algorithm will be optimized; and edge enhancement parameters in the image signal processor will be adjusted.

[0088] If the problem is an unstable network connection, then the following actions will be initiated and executed in sequence: automatically switch to a backup wireless access point or frequency band, dynamically adjust the video encoding bitrate and keyframe interval, and optimize the buffer and retransmission mechanism parameters of the network protocol stack.

[0089] Example 6, in step 3, between the strong vibration step and the software repair step, a surface cleanliness detection step is also included:

[0090] After the strong vibration step is completed, images of the current environment are acquired; the acquired images are analyzed using a trained convolutional neural network model to determine whether there are stubborn stains or condensation on the lens surface.

[0091] If the presence of foreign objects is detected, an enhanced cleaning reminder will be sent to the user via the mobile app, and sound waves of a specific frequency will be played to drive away the foreign objects adhering to the lens surface.

[0092] Example 7: Playing sound waves of a specific frequency to remove deposits from the lens surface is achieved based on the sound wave emitting module 4. The sound wave emitting module 4 includes a piezoelectric ceramic sheet and a core processing unit 2, which are integrated into the smart home camera and installed close to the outer ring of the camera lens or the inner side of the protective cover. The core processing unit 2 has a built-in sound wave ion driving program, which controls the piezoelectric ceramic sheet to remove deposits from the lens surface using sound waves.

[0093] The specific method for using sound waves to remove deposits from the lens surface is as follows: The core processing unit 2 initiates the sound wave ion-removing program and sends a control signal to the drive circuit of the sound wave emission module 4; the drive circuit generates an AC voltage with a set frequency and waveform according to the control signal, driving the piezoelectric ceramic sheet to produce mechanical vibration, radiating sound waves in the air directly in front of the lens; one or more specific frequencies within the range of 35kHz to 45kHz have a repelling or interference effect on small insects, and can also be effectively absorbed by the surface of water droplets and induce micro-vibrations. Each sound wave emission lasts for 1 to 3 seconds, and is repeated 2 to 4 times depending on the degree of lens contamination or ambient temperature; image acquisition is paused during sound wave emission.

[0094] After the sound wave emission action ends, images are collected again. By comparing the changes in image features before and after the removal, the removal effect is evaluated. If the characteristics of the attached substances are not eliminated, clear physical cleaning instructions are pushed to the user via mobile APP.

[0095] By adopting the above technical solution, the core design principle of the strong vibration step is to efficiently remove minor deposits from the lens surface and component gaps through targeted vibration parameters and operating logic, while avoiding damage to the precision components inside the camera. The strong vibration module uses multiple miniature eccentric wheel vibration motors to achieve uniform vibration, ensuring that the vibration force covers the key areas of the camera and improving the removal effect of deposits.

[0096] Setting the amplitude to a range of 0.5 mm to 1 mm and the frequency to a range of 80 Hz to 120 Hz is a comprehensive consideration of both the efficiency of removing deposits and the need for component protection. This parameter range generates sufficient vibration force to remove light dust and impurities without causing internal wiring to loosen or precision components to be damaged due to excessive vibration amplitude or frequency. An interval operation mode of vibrating for 5 to 10 seconds, stopping for 5 to 10 seconds as one cycle, and repeating 3 to 5 cycles, avoids the heat accumulation generated by continuous vibration and provides time for the removed deposits to fall off, further improving the cleaning effect.

[0097] The image acquisition and upload functions are paused during the strong vibration process because the vibration will cause the images to shake violently, and the acquired images will have no effective monitoring value. At the same time, pausing the upload can reduce the waste of network resources and prevent temporary signal abnormalities caused by vibration from being misjudged as faults.

[0098] The software repair steps are based on targeted repair logic according to the fault type. Based on the specific performance degradation or fault problem diagnosed in step 2, the corresponding software repair program is matched to achieve accurate repair and efficient resolution of software-level anomalies, restoring the normal function of components without manual intervention.

[0099] To address the issue of degraded image sharpness, a deep learning-based image deblurring algorithm and an adaptive image denoising and sharpening algorithm are sequentially activated, and the edge enhancement parameters of the image signal processor are adjusted. This is because such problems are often caused by software factors such as algorithm parameter offsets and image transmission interference. Through algorithm optimization and parameter adjustment, the imaging quality can be gradually improved, restoring the normal imaging effect of the image sensor without the need for hardware operations.

[0100] To address network connectivity instability, the system sequentially performs actions such as automatically switching to backup wireless access points or frequency bands, dynamically adjusting video encoding bitrate and keyframe intervals, and optimizing network protocol stack buffer and retransmission mechanism parameters. This design incorporates key factors affecting network transmission. Switching to backup access points and frequency bands can resolve the issue of weak network signals, adjusting encoding parameters can reduce network transmission pressure, and optimizing protocol stack parameters can reduce packet loss and latency. These multiple steps work together to achieve rapid recovery of network connectivity stability.

[0101] A surface cleanliness testing step is added between the strong vibration step and the software repair step to verify the cleaning effect of the strong vibration step. This avoids the subsequent software repair from failing to achieve the expected results due to stubborn stains not being removed, while also reducing ineffective software repair operations and improving self-maintenance efficiency.

[0102] After the strong vibration step is completed, images are acquired and analyzed using a trained convolutional neural network model. This model can accurately identify whether there are stubborn stains or condensation on the lens surface. Trained on a large amount of data from various contamination scenarios, the model can effectively distinguish between minor deposits and stubborn stains, avoiding misjudgments. If stubborn stains or condensation are detected, an enhanced cleaning prompt is sent and sonic repellent is initiated. This is because such deposits cannot be removed by strong vibration and require further repair using sonic repellent. The user is also informed in advance of the current maintenance progress and potential subsequent operational needs.

[0103] The principle of using sound waves to remove deposits from the lens surface is to utilize the mechanical vibration effect of sound waves at a specific frequency to target the deposits on the lens surface, achieving contactless cleaning and avoiding scratches caused by physical cleaning. At the same time, the coordinated control of the sound wave emitting module and the core processing unit ensures that the removal process is precise and controllable.

[0104] The acoustic emission module uses a piezoelectric ceramic sheet. Combined with the acoustic ion-repelling program of the core processing unit, it can achieve precise control of the acoustic frequency and waveform. The piezoelectric ceramic sheet is installed on the outer ring of the lens or the inside of the protective cover, ensuring that the acoustic waves act directly on the area in front of the lens, improving the repelling effect. The core processing unit sends control signals to the drive circuit, which generates a corresponding AC voltage to drive the piezoelectric ceramic sheet to vibrate and radiate acoustic waves, forming a complete control closed loop and realizing the automated operation of acoustic ion repelling.

[0105] The specific frequency is set in the 35kHz to 45kHz range because this frequency band of sound waves can both repel or interfere with small insects, preventing them from attaching and affecting imaging, and can be effectively absorbed by the surface of water droplets, inducing micro-vibrations that cause the water droplets to detach from the lens surface. At the same time, this frequency band of sound waves will not affect the human body or interfere with the operation of other camera components. Each transmission lasts 1 to 3 seconds and is repeated 2 to 4 times, which can be flexibly adjusted according to the degree of pollution and ambient temperature to ensure effective repellency while avoiding energy waste.

[0106] Image acquisition is paused during acoustic wave emission because acoustic vibrations can briefly affect the stability of image acquisition, thus avoiding the acquisition of invalid images. After the emission ends, images are acquired again and the features before and after the removal are compared, which can accurately assess the removal effect. If the attached material is not removed, physical cleaning instructions are sent, which clearly informs users of the current maintenance limits and provides specific operational directions, taking into account both the autonomy of self-maintenance and the convenience of user operation.

[0107] In Example 8, step 4, the condition for determining that automatic maintenance is complete is:

[0108] After repeating step 1, the detected values ​​of all critical component statuses returned to the normal range; and the AI ​​model assessment in step 2 determined that this would not affect subsequent continuous and reliable operation.

[0109] The self-maintenance results pushed to users via mobile app include charts comparing key performance parameters before and after maintenance, a list of specific steps used in the maintenance, and the overall health score of the current device.

[0110] By adopting the above technical solution, dual verification ensures that the self-maintenance effect meets the standards, avoiding misjudgments caused by single detection errors or short-term recovery, ensuring the continued stable operation of the equipment, and improving the self-maintenance closed-loop logic. Repeating step 1 to detect the status of key components is to intuitively and quantitatively verify the repair effect of the self-maintenance program on faults or performance degradation issues, ensuring that the detection values ​​of all key components return to the normal range, and confirming from the hardware performance level that the fault has been resolved.

[0111] The requirement that the AI ​​model in step 2 comprehensively assesses that it will not affect the subsequent continuous and reliable operation is because a simple short-term recovery of the component status may have potential hidden dangers. The AI ​​model can combine the component's historical operating data, current environmental parameters, and data from this maintenance process to comprehensively judge the stability of the equipment's subsequent operation, avoid situations where there is a short-term recovery but the risk of long-term fault recurrence, further improve the effectiveness and reliability of self-maintenance, and reduce the probability of subsequent faults recurring.

[0112] The self-maintenance results, pushed to users via a mobile app, aim to provide a comprehensive and intuitive understanding of the entire self-maintenance process and its final outcome, thus satisfying users' right to know about the device's maintenance status. Before-and-after performance parameter comparison charts clearly demonstrate the performance changes brought about by the maintenance, allowing users to intuitively perceive the value of the repair. A list of specific maintenance steps clarifies the maintenance actions performed, enhancing user trust in the self-maintenance process. The current overall device health score simplifies users' assessment of the device's current status, eliminating the need to review complex parameters one by one, allowing for quick understanding of the device's overall operating condition and further optimizing the user experience.

[0113] Example 9 also includes step 5, where each complete self-maintenance process, including the triggering cause, the strong vibration and software repair steps performed, and the final verification result, is anonymized and uploaded to the cloud maintenance knowledge base;

[0114] The cloud server utilizes massive amounts of maintenance case data to regularly optimize the parameters of the fault diagnosis and grading model in step 2 and the software repair algorithm in step 3, and pushes them to all smart home cameras via online firmware upgrades.

[0115] By adopting the above technical solution, the data from each complete self-maintenance process is anonymized and uploaded to the cloud-based maintenance knowledge base. The principle is to aggregate maintenance instances from different devices and scenarios, accumulating diverse fault triggering causes, maintenance step execution effects, and verification results data, all while protecting user privacy. Anonymization avoids the leakage of user privacy information and ensures the universality of the uploaded data, providing effective support for subsequent model and algorithm optimization and solving the problems of limited maintenance data and insufficient optimization space for single devices.

[0116] The cloud server regularly optimizes the fault diagnosis and classification model in step 2 and the software repair algorithm parameters in step 3 using massive maintenance case data. The principle is based on the iterative logic of big data analysis and machine learning. The massive case data can cover more complex fault scenarios and performance degradation. Through in-depth data mining and analysis, the deviation parameters in the model and algorithm are corrected to improve the accuracy of fault diagnosis, the rationality of classification, and the success rate of software repair.

[0117] The optimized model and algorithm parameters are pushed to all cameras through online firmware upgrades. The principle is to achieve full coverage of the optimization results. Without manual operation by the user, all devices can obtain better maintenance capabilities simultaneously, avoiding uneven maintenance levels due to individual device differences. At the same time, it reduces the operation cost of manual upgrades and ensures that the maintenance performance of the entire product system can be continuously iterated and adapted to changes in different usage scenarios and differences in component wear over the long term.

[0118] Example 10: A smart home camera, used to implement a method for rapid adjustment and maintenance of a smart home camera, includes a camera body 1, a core processing unit 2, a strong vibration module 3, and a sound wave emission module 4. The camera body 1 has a built-in image sensor, an infrared fill light, a microphone, and a network communication module. The core processing unit 2 is communicatively connected to the camera body 1 and is used to run a self-test program, trigger logic, and a software repair program.

[0119] The strong vibration module 3 is fixed inside the camera body 1 and includes at least one miniature eccentric wheel vibration motor for performing the strong vibration step;

[0120] The acoustic wave emitting module 4 is at least one piezoelectric ceramic transducer. The transducer is attached to the annular groove between the outer ring of the camera lens and the protective cover via a ring flexible circuit board, or is directly attached to the inner surface of the protective cover. The vibration plane is parallel to the lens surface and is used to radiate sound waves in front of the lens. The core processing unit 2 controls the execution actions of the strong vibration module 3 and the acoustic wave emitting module 4 respectively.

[0121] The following specific embodiments illustrate the implementation principle of the present invention:

[0122] Scenario 1: A smart home camera in the living room, which has been in use for 8 months and is used frequently. It is turned on for an average of 16 hours a day, and the self-test cycle is dynamically adjusted to once every 2 hours based on the historical usage frequency, which is in line with the setting in Example 2 that "the higher the usage frequency, the shorter the self-test cycle".

[0123] Fault Trigger: The camera's built-in self-test program detects the image sensor's clarity. By analyzing the image of a fixed reference object (a painting hanging in the living room) 1.5 meters in front of the lens, its edge modulation transfer function value is calculated. The value is found to be below the normal threshold, indicating a performance degradation exceeding the first threshold but not exceeding the second threshold. Simultaneously, the fault diagnosis and grading model assesses a confidence level of 82%, triggering a level-two recommended maintenance. Maintenance suggestions are pushed via the mobile app, clearly describing the problem as a decrease in image sensor clarity, suspected to be slight dust accumulation on the lens and software parameter deviation, with a confidence level of 82% and an estimated repair time of 5 minutes.

[0124] Maintenance execution: When the user selects the self-maintenance mode in the mobile APP, the camera automatically enters the self-maintenance program and executes the strong vibration step and the software repair step in sequence.

[0125] Strong vibration procedure: Turn on strong vibration module 3 and run according to the set vibration parameters: amplitude 0.8 mm, frequency 100 Hz, with a cycle of 8 seconds of vibration followed by 8 seconds of rest, repeating for 4 cycles; during strong vibration operation, the camera image acquisition and uploading function is paused.

[0126] Software repair steps: To address the issue of decreased image sharpness, the current video stream is processed sequentially using a deep learning-based image deblurring algorithm, followed by adaptive image denoising and sharpening algorithm optimization. The edge enhancement parameters in the image signal processor are then adjusted to complete the software-level repair.

[0127] Maintenance Verification and Results: After the self-maintenance procedure is completed, the camera repeatedly executes the self-test procedure and fault diagnosis process. The edge modulation transfer function value of the image sensor is found to have returned to the normal range. Furthermore, the fault diagnosis and grading model (AI model) confirms that it will not affect subsequent continuous and reliable operation, thus completing the automatic maintenance. The mobile app pushes the self-maintenance results to the user, including a comparison chart of image clarity parameters before and after maintenance, a list of strong vibration parameters executed, and a list of software repair steps. The current overall health score of the device is 92 points.

[0128] Scene 2: Stubborn stains on the lens:

[0129] The outdoor courtyard smart home camera is exposed to dusty environments daily. Its self-test cycle is dynamically adjusted to once every hour, which is consistent with the self-test cycle setting in Example 2. After a rainfall, condensation and slight dust adhered to the lens surface.

[0130] Fault Trigger: The self-test program detected a severe decrease in image sensor clarity. The fault diagnosis and grading model assessed the confidence level at 78%, indicating that the performance degradation exceeded the first threshold. A secondary recommended maintenance suggestion was pushed out, indicating that the image was blurry and there was suspected deposits on the lens surface. The estimated repair time was 8 minutes.

[0131] Maintenance execution: When the user selects the self-maintenance mode, the camera enters the self-maintenance program and sequentially executes the strong vibration step, the surface cleanliness detection step, the acoustic wave removal step, and the software repair step.

[0132] 1. Strong vibration step: The strong vibration module 3 operates at an amplitude of 1.0 mm and a frequency of 120 Hz. It vibrates for 10 seconds and stops for 10 seconds as one cycle. It repeats for 5 cycles, pauses the image acquisition and upload function, and removes loose dust from the lens surface.

[0133] 2. Surface cleanliness detection steps: After the strong vibration step, the camera captures images of the current environment. The images are analyzed by a trained convolutional neural network model to determine if there is still condensation on the lens surface. Then, an enhanced cleaning reminder is pushed to the user via a mobile APP.

[0134] 3. Acoustic wave ion removal procedure: The core processing unit 2 initiates the acoustic wave ion removal program, controlling the acoustic wave emission module 4 to emit acoustic waves at a frequency of 38kHz, each lasting 2 seconds, repeating for 3 cycles; during the acoustic wave emission, the image acquisition function is paused. After the acoustic wave emission ends, the camera re-acquires images, comparing the image features before and after the ion removal to confirm that the condensed water mist has been completely eliminated.

[0135] 4. Software repair steps: Activate image deblurring, noise reduction and sharpening algorithms, fine-tune image signal processor parameters, and further optimize image clarity.

[0136] Maintenance verification results: After repeating the self-test and fault diagnosis process, the image sensor clarity returned to normal, the AI ​​model assessment showed no potential hidden dangers, and automatic maintenance was completed; the mobile APP pushed the maintenance results, including the specific parameters of strong vibration and acoustic wave removal, the comparison of lens surface images before and after removal, and the current equipment health score of 89 points.

[0137] Scenario 3: Unstable network connection:

[0138] The smart home camera in the bedroom is connected to the home wireless network and is used at a moderate frequency. Its self-test cycle is once every 4 hours. Recently, the home network frequency band has been congested, causing abnormal network connection of the camera, which is consistent with the scenario setting of network connection stability detection in Example 2.

[0139] Fault Trigger: The self-test program continuously measures the round-trip latency, bandwidth utilization, and packet loss rate of data packets with the cloud server. It finds that the packet loss rate has reached 15%, exceeding the normal threshold, and the network connection stability has decreased. The fault diagnosis and grading model assesses the confidence level at 85% and pushes a level-two recommended maintenance suggestion, indicating "unstable network connection, suspected wireless signal interference or frequency band congestion", with an estimated repair time of 3 minutes.

[0140] Maintenance execution: When the user selects the self-maintenance mode, the camera enters the self-maintenance program, skips the strong vibration step, and directly executes the software repair step.

[0141] Software repair steps: In response to the unstable network connection issue, the core processing unit 2 sequentially initiates the following actions: automatically switches to the home backup wireless access point (switching from the 2.4G band to the 5G band), dynamically adjusts the video encoding bitrate (from 1080P 2Mbps to 1080P 1.5Mbps), and optimizes the buffer and retransmission mechanism parameters of the network protocol stack to reduce data packet loss.

[0142] Maintenance Verification and Results: After the software repair was completed, the self-test process was repeated. It was found that the packet loss rate dropped to 2%, the round-trip latency returned to normal, the AI ​​model assessed that it could continue to operate stably, and the automatic maintenance was completed. The mobile APP pushed the maintenance results, including a comparison chart of network parameters before and after the maintenance, the specific operation steps of this software repair, and the device health score of 90 points.

[0143] Subsequent optimizations: This complete self-maintenance process was triggered by network frequency band congestion; the execution steps included three software repair actions; the verification result was successful; the data was anonymized and uploaded to the cloud maintenance knowledge base; the cloud server, combining a large number of similar maintenance cases, optimized the frequency band congestion identification parameters of the fault diagnosis and grading model, as well as the bitrate adjustment logic in the software repair steps. Subsequently, through online firmware upgrades, the optimized parameters will be pushed to the camera to improve the accuracy of subsequent network anomaly diagnosis and repair efficiency.

[0144] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for quick adjustment and maintenance of a smart home camera, characterized in that... This includes the following steps: Step 1: The camera's built-in self-test program checks the status of key components at set intervals. The status of key components includes image sensor clarity, infrared night vision function, audio input quality, and network connection stability. Step 2: Input the status of key components into the fault diagnosis and classification model. When the fault diagnosis and classification model determines that the status of any key component has detected a performance degradation that reaches a set threshold or a fault, it triggers a maintenance recommendation action and pushes a graded maintenance recommendation to the user through the mobile APP. The graded maintenance recommendation includes a problem description, confidence level, and estimated repair time. Step 3: After the user selects the self-maintenance mode, the camera automatically enters the self-maintenance program; the self-maintenance program includes a strong vibration step and a software repair step executed sequentially. The strong vibration step involves activating the strong vibration module (3) of the smart home camera, which then runs a set number of times according to the set vibration parameters. The software repair steps involve starting the corresponding software repair program to complete the self-repair process based on the specific performance degradation or malfunction issue identified in step 2. Step 4: After the self-maintenance procedure is completed, repeat steps 1 and 2. If the status detection results of all key components do not trigger maintenance suggestion actions, the automatic maintenance is considered complete, and the self-maintenance results are pushed to the user via the mobile APP.

2. The method for quick adjustment and maintenance of a smart home camera according to claim 1, characterized in that... In step 1, the self-test cycle is dynamically adjusted based on the camera's historical usage frequency; the higher the usage frequency, the shorter the self-test cycle. The detection of key component status specifically includes image sensor clarity detection, infrared night vision function detection, audio input quality detection, and network connection stability detection. Image sensor sharpness detection involves analyzing the image of a fixed reference object at a preset distance in front of the camera and calculating the modulation transfer function value of its edges. Infrared night vision function detection is performed in low-light environments by comparing the brightness histogram and noise level of images before and after the infrared lights are turned on to determine the working status of the infrared light group and the photosensitive element. Audio input quality testing involves playing a built-in standard frequency test tone and analyzing the signal-to-noise ratio, total harmonic distortion, and frequency response curve of the microphone received signal. Network connectivity stability testing involves continuously measuring the round-trip latency, bandwidth utilization, and packet loss rate of data packets between the network and a cloud server or a specified device within a local area network.

3. A method for quick adjustment and maintenance of a smart home camera according to claim 2, characterized in that... In step 2, the tiered maintenance recommendations are divided into three levels: The first level is advisory maintenance, corresponding to performance degradation not exceeding the first threshold and confidence level below 70%. Users are advised to observe or schedule non-emergency maintenance. The second level is recommended maintenance, which corresponds to performance degradation exceeding the first threshold but not exceeding the second threshold, or a confidence level between 70% and 90%. Users are advised to perform self-maintenance under the guidance of the APP. The third level is a mandatory maintenance or fault alarm, corresponding to the detection of a functional failure, performance degradation exceeding the second threshold, or a confidence level higher than 90%. Immediate action is strongly recommended, and it indicates that core monitoring functions are affected. The push interface of the mobile APP uses color codes to distinguish levels and displays simulation diagrams and performance curves of specific faulty components.

4. A method for quick adjustment and maintenance of a smart home camera according to claim 3, characterized in that... In step 3, the strong vibration step specifically includes: The strong vibration module (3) includes multiple miniature eccentric wheel vibration motors; The set vibration parameters include amplitude, frequency, and duration, where the amplitude is 0.5 mm to 1 mm and the frequency is 80 Hz to 120 Hz. The interval operation is set as follows: one cycle consists of 5-10 seconds of vibration followed by 5-10 seconds of rest, and this cycle is repeated 3 to 5 times. During the strong vibration step, the camera image acquisition and uploading function is paused.

5. A method for quick adjustment and maintenance of a smart home camera according to claim 4, characterized in that... The specific steps for software repair include: If the problem is a decrease in image sharpness, the following actions will be initiated and executed sequentially: a deep learning-based image deblurring algorithm will be used to process the current video stream; an adaptive image denoising and sharpening algorithm will be optimized; and edge enhancement parameters in the image signal processor will be adjusted. If the problem is an unstable network connection, then the following actions will be initiated and executed in sequence: automatically switch to a backup wireless access point or frequency band, dynamically adjust the video encoding bitrate and keyframe interval, and optimize the network protocol stack buffer and retransmission mechanism parameters.

6. A method for quick adjustment and maintenance of a smart home camera according to claim 5, characterized in that... In step 3, between the strong vibration step and the software repair step, a surface cleanliness detection step is also included: After the strong vibration step is completed, images of the current environment are acquired; the acquired images are analyzed using a trained convolutional neural network model to determine whether there are stubborn stains or condensation on the lens surface. If the presence of foreign objects is detected, an enhanced cleaning reminder will be sent to the user via the mobile app, and sound waves of a specific frequency will be played to drive away the foreign objects adhering to the lens surface.

7. A method for quick adjustment and maintenance of a smart home camera according to claim 6, characterized in that... The sound wave emission module (4) is used to drive away the deposits on the lens surface by playing sound waves of a specific frequency. The sound wave emission module (4) includes a piezoelectric ceramic sheet and a core processing unit (2) integrated into the smart home camera and installed close to the outer ring of the camera lens or the inner side of the protective cover. The core processing unit (2) has a built-in sound wave ion driving program, which controls the piezoelectric ceramic sheet to drive away the deposits on the lens surface by sound waves. The specific method for using sound waves to remove deposits from the lens surface is as follows: The core processing unit (2) initiates the sound wave ion-removing program and sends a control signal to the drive circuit of the sound wave emission module (4); the drive circuit generates an AC voltage with a set frequency and waveform according to the control signal, which drives the piezoelectric ceramic sheet to generate mechanical vibration and radiates sound waves in the air directly in front of the lens; one or more frequency points in the range of 35kHz to 45kHz have a repelling or interference effect on small insects, and can be effectively absorbed by the surface of water droplets and induce micro-vibrations. Each sound wave emission lasts for 1 to 3 seconds and is repeated 2 to 4 times depending on the degree of lens contamination or ambient temperature; image acquisition is paused during sound wave emission. After the sound wave emission action ends, images are collected again. By comparing the changes in image features before and after the removal, the removal effect is evaluated. If the characteristics of the attached substances are not eliminated, clear physical cleaning instructions are pushed to the user via mobile APP.

8. A method for quick adjustment and maintenance of a smart home camera according to claim 7, characterized in that... In step 4, the conditions for determining whether automatic maintenance is complete are: After repeating step 1, the detected values ​​of all critical component statuses returned to the normal range; and the AI ​​model assessment in step 2 determined that this would not affect subsequent continuous and reliable operation. The self-maintenance results pushed to users via mobile app include charts comparing key performance parameters before and after maintenance, a list of specific steps used in the maintenance, and the overall health score of the current device.

9. A method for quick adjustment and maintenance of a smart home camera according to claim 8, characterized in that... It also includes step 5, where each complete self-maintenance process, including the triggering cause, the strong vibration and software repair steps performed, and the final verification results, is anonymized and uploaded to the cloud maintenance knowledge base; The cloud server utilizes massive amounts of maintenance case data to regularly optimize the parameters of the fault diagnosis and grading model in step 2 and the software repair algorithm in step 3, and pushes them to all smart home cameras via online firmware upgrades.

10. A smart home camera, characterized in that... The method for quickly adjusting and maintaining a smart home camera as described in claim 9 includes a camera body (1), a core processing unit (2), a strong vibration module (3), and a sound wave emission module (4). The camera body (1) has a built-in image sensor, an infrared fill light, a microphone, and a network communication module. The core processing unit (2) is communicatively connected to the camera body (1) and is used to run a self-test program, trigger logic, and a software repair program. The strong vibration module (3) is fixed inside the camera body (1) and includes at least one miniature eccentric wheel vibration motor for performing the strong vibration step; The acoustic wave emitting module (4) is at least one piezoelectric ceramic transducer. The transducer is attached to the annular groove between the outer ring of the camera lens and the protective cover via a ring flexible circuit board, or is directly attached to the inner surface of the protective cover. The vibration plane is parallel to the lens surface and is used to radiate sound waves in front of the lens. The core processing unit (2) controls the execution actions of the strong vibration module (3) and the acoustic wave emitting module (4) respectively.

Images