A method, device and storage medium for repairing signal interruption in a digital camera.

Abstract

This invention discloses a method for repairing signal interruptions in a digital camera, comprising the following steps: S1, pre-installing a first network connection module and a second network connection module in the digital camera; S2, sending an interruption signal to the network settings module for detection after disconnection; S3, the network settings module detecting the network settings parameters of the digital camera; S4, disabling the first network connection module and enabling the second network connection module; S5, obtaining the abnormal parameters of the camera disconnection and repairing them in the router's logs and network settings; S6, sending the faulty camera's ID and network parameters to the main control backend system and sending a maintenance and repair request to the user; This invention solves the problem that existing technologies cannot determine whether network faults are due to hardware or software reasons, and cannot automatically repair network faults in digital cameras.

Description

Technical Field

[0001] This invention relates to the technical field of digital cameras, and in particular to a method, device, and storage medium for repairing signal interruptions in digital cameras. Background Technology

[0002] In today's digital age, digital cameras are widely used in many fields, such as security monitoring, video conferencing, and distance learning. Among these, digital cameras that connect via wireless networks offer users a convenient experience. However, in practical applications, the stability of their wireless network connections is constrained by various factors. In existing technologies, the network settings of the camera software are crucial for ensuring a correct connection to the wireless network, but improper settings often lead to connection problems. These factors can be broadly categorized into two parts: one is the network settings of the digital camera itself, and the other is the router's network and security settings that may block the digital camera's connection signal. Current technology cannot determine whether the network failure is due to hardware or software issues, nor can it automatically repair digital camera network failures; manual repair and maintenance are the only options. Therefore, there is an urgent need for a method to repair signal interruptions in digital cameras. Summary of the Invention

[0003] In order to solve the above-mentioned technical problems, the present invention provides a method, device and storage medium for repairing signal interruption of a digital camera.

[0004] The technical solution of this invention is implemented as follows:

[0005] A method for repairing signal interruption in a digital camera includes the following steps:

[0006] S1, a first network connection module and a second network connection module are pre-built into the digital camera;

[0007] S2 monitors the received wireless network signal connection in real time through the wireless network card, and sends an interrupt signal to the network setting module for detection after disconnection.

[0008] S3, the network setting module detects the network parameters of the digital camera, resets the network parameters, and then detects again whether the network connection signal has been reconnected;

[0009] S4, according to step S3, if the network is still disconnected after resetting, search for and reconnect to the wireless network signal again. If the network is still disconnected, disable the first network connection module and start the second network connection module.

[0010] S5, the second network connection module starts the data connection mode, establishes a connection with the main control backend system, sends the network parameters of the faulty camera to the main control backend system, the main control backend system connects to the router before the camera disconnected and obtains the network log of the router, obtains the abnormal parameters of the camera disconnection, and repairs it in the router's log and network settings;

[0011] S6, enable the first network connection module and disable the second network connection module, obtain the network connection status, if the connection with the router is still disconnected, disable the first network connection module and enable the second network connection module, determine that it is a hardware failure, send the number and network parameters of the faulty camera to the main control backend system, and send a repair request to the user.

[0012] Preferably, the first network connection module and the second network connection module in step S1 are a wireless WIFI network and a data roaming network of the SIM card sub-card, respectively.

[0013] Preferably, in step S2, the wireless network card monitors the real-time connection signal strength, sends detection data packets to the router, and the router sends received signals to the wireless network card as a complete signal judgment loop. The time threshold for sending and receiving detection data packets is set to i. If sending or receiving fails within i time, it is judged as an abnormal network state, and a network repair request signal is sent to the network setting module.

[0014] Preferably, the network setting module processes both the network repair request signal and the interrupt signal received in step S3.

[0015] Preferably, in step S1, the first network connection module and the second network connection module of the digital camera record network logs, which record network parameters, connection status and signal strength, as well as the total amount of data received and sent. The network parameters in the normal connection state are set as Table Y, the network abnormal state is set as Table T, and the signal interruption state is set as Table N.

[0016] A device includes at least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the above-described method for repairing signal interruption of a digital camera.

[0017] A storage medium storing computer program instructions, characterized in that: when the computer program instructions are executed by a processor, the above-mentioned signal interruption repair method for a digital camera is implemented.

[0018] This invention solves the problem that existing technologies cannot determine whether network failures are caused by hardware or software issues, nor can they automatically repair network failures in digital cameras. Furthermore, this invention increases the methods for identifying and repairing faults through a dual network connection module, improving the likelihood of connection restoration. Simultaneously, real-time monitoring and setting time-end thresholds to identify network anomalies enable timely detection of problems and initiation of repair requests, reducing the impact of interruptions. This invention also utilizes network logs to record various information, facilitating the resetting of settings parameters based on historical normal states, improving repair accuracy. Fourthly, it provides targeted handling procedures for different connection anomalies, enabling rapid location and attempt to resolve problems. Fifthly, it can feed back fault information to the main control backend system, facilitating unified management and scheduling of maintenance and repairs, ensuring the stability and continuity of camera network connections, and improving the user experience. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of a signal interruption repair method for a digital camera according to the present invention. Detailed Implementation

[0020] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, exemplary embodiments will be described in detail below, examples of which are illustrated in the accompanying drawings. In the following description relating to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of methods and systems consistent with some aspects of this application as detailed in the appended claims.

[0021] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0022] Example 1

[0023] A method for repairing signal interruption in a digital camera includes the following steps:

[0024] S1, a first network connection module and a second network connection module are pre-built into the digital camera;

[0025] Preferably, the first network connection module and the second network connection module in step S1 are a wireless WIFI network and a data roaming network of the SIM card sub-card, respectively.

[0026] Preferably, in step S1, the first network connection module and the second network connection module of the digital camera record network logs, which record network parameters, connection status and signal strength, as well as the total amount of data received and sent. The network parameters in the normal connection state are set as Table Y, the network abnormal state is set as Table T, and the signal interruption state is set as Table N.

[0027] The digital camera has a pre-installed first network connection module and a second network connection module. The first network connection module is a wireless Wi-Fi network, used to establish a network connection through a nearby wireless router in normal environments to meet daily video transmission needs; the second network connection module is a data roaming network of the secondary SIM card, serving as a backup network connection method. It provides an alternative network access method when the Wi-Fi network fails and cannot connect normally, ensuring that the camera can still maintain data transmission capabilities to a certain extent.

[0028] Both network connection modules are equipped with network log recording capabilities, providing detailed information on network parameters (such as IP address, subnet mask, gateway, network name, password, etc.), connection status (connected, connecting, disconnected, etc.), signal strength, and total data received and transmitted. To facilitate rapid identification and processing of different states, network parameters for normal connection states are organized into Table Y, network anomaly states into Table T, and signal interruption states into Table N.

[0029] S2 monitors the received wireless network signal connection in real time through the wireless network card, and sends an interrupt signal to the network setting module after disconnection.

[0030] Preferably, in step S2, the wireless network card monitors the real-time connection signal strength, sends detection data packets to the router, and the router sends received signals to the wireless network card as a complete signal judgment loop. The time threshold for sending and receiving detection data packets is set to i. If sending or receiving fails within i time, it is judged as an abnormal network state, and a network repair request signal is sent to the network setting module.

[0031] Preferably, the network setting module processes both the network repair request signal and the interrupt signal received in step S3.

[0032] The wireless network card in a digital camera plays a crucial role in real-time monitoring of the received wireless network signal connection. It employs a comprehensive monitoring method that goes beyond simply detecting signal strength. Instead, it actively sends detection data packets to the router and waits for the router to relay the received signal back to the wireless network card, thus creating a complete and dynamic signal assessment cycle.

[0033] This method allows for a more comprehensive and accurate assessment of the stability and quality of wireless network signal connections. In each signal assessment cycle, the wireless network card analyzes the transmitted and received signals according to preset rules and algorithms to obtain information on signal strength, signal integrity, transmission delay, and other aspects, thereby gaining a deeper understanding of the current wireless network signal connection status.

[0034] To promptly detect network anomalies, a time threshold of 'i' is set for detecting data packet transmission and reception signals. The setting of this time threshold needs to consider various factors, such as the complexity of the network environment, the processing capacity of the devices, and the normal latency range of data transmission. For example, in a typical home or office network environment, 'i' can be initially set to 5 seconds. However, in more complex network environments (such as large public places or industrial plants), this threshold may need to be appropriately extended or shortened based on the actual situation.

[0035] Within this set time threshold i, if a failure occurs in sending a test data packet—that is, the wireless network card fails to successfully send the data packet within the specified time or fails to receive the received signal from the router—it will be determined as a network abnormality. Once a network abnormality is determined, the wireless network card will immediately trigger the corresponding mechanism to generate and send a network repair request signal to the network settings module, so as to promptly initiate the subsequent repair process and avoid prolonged interruption of camera functionality due to network problems.

[0036] S3, the network setting module detects the network parameters of the digital camera, resets the network parameters, and then detects again whether the network connection signal has been reconnected;

[0037] Preferably, in step S3, the network settings module resets the real-time network settings to the network parameters of the most recent normal connection state through the network log;

[0038] When the network settings module receives a network repair request signal or interruption signal from the wireless network card, it will immediately initiate a comprehensive detection process of the digital camera's network parameters. First, it will delve into the previously recorded network logs to accurately locate the network parameter record corresponding to the current moment, and perform a detailed comparative analysis with the network parameters of the most recent normal connection state (i.e., the parameters recorded in Table Y).

[0039] During the comparison process, every key network parameter is checked one by one, including whether the IP address has changed, whether the subnet mask matches, whether the gateway setting is correct, and whether the network name (SSID) and network password are accurate. Through this comprehensive and meticulous comparison, any possible abnormal parameter settings can be quickly identified, which may be the direct cause of signal interruption or network anomalies.

[0040] If any discrepancies or anomalies are detected during the comparative analysis, the network settings module will automatically activate the parameter reset function to precisely reset the real-time network settings to the network parameters of the most recent normal connection state (i.e., reset according to the parameter values ​​in Table Y). This step aims to eliminate signal interruption issues caused by incorrect or abnormal parameter settings by restoring the previous normal connection configuration, so that the camera's network connection can be restored to normal as soon as possible.

[0041] When resetting parameters, the network settings module strictly follows preset procedures and rules to ensure that each parameter is accurately reset to the correct value. Furthermore, to guarantee the stability and reliability of the reset process, relevant parameters are verified multiple times before and after the reset to prevent new problems from arising due to reset errors.

[0042] After resetting the network parameters, the network settings module does not stop there. Instead, it immediately checks again to see if the network connection signal can be re-established. It works in conjunction with related hardware components such as the wireless network card to attempt to re-establish the connection with the external network (such as a router), and monitors various signal feedbacks and status changes during the connection establishment process.

[0043] This method allows for a preliminary assessment of whether a simple parameter reset can resolve the signal interruption issue. If a connection is successfully established and the network signal is stable during the reconnection test, it indicates that the signal interruption was likely caused by abnormal network parameters and has been successfully fixed by resetting the parameters. Conversely, if a connection still cannot be established, further investigation is needed to identify other possible causes of the signal interruption before proceeding to subsequent repair steps.

[0044] S4, according to step S3, if the network is still disconnected after resetting, search for and reconnect to the wireless network signal again. If the network is still disconnected, disable the first network connection module and start the second network connection module.

[0045] Preferably, in step S4, the process of re-searching for and reconnecting to the wireless network signal involves resetting network parameters, obtaining the name of the most recently connected router from the network log, reconnecting to the router with that name, and setting a reconnection threshold. If the router name cannot be found, it is determined that the router is abnormal. If the router name is found but a connection cannot be established, it is determined that the first network connection module is in an abnormal state, the first network connection module is disabled, and the second network connection module is started.

[0046] According to step S3, if the connection remains lost after resetting the network parameters, a more thorough re-search for the wireless network signal and an attempt to reconnect are required. During this process, the network settings module will extract the name of the most recently connected router from the previously recorded network logs.

[0047] Then, targeting this router by its name, a new search and connection attempt is initiated. During this re-search process, the wireless network card will send search signals at a certain frequency (e.g., every few seconds) within its detectable range, attempting to find the wireless network signal emitted by the target router. Simultaneously, the searched wireless network signals will undergo preliminary screening and identification to ensure that the signal found is indeed the correct signal emitted by the target router, and not other interfering signals or signals from routers with similar names.

[0048] To avoid unlimited search and connection attempts, which could lead to wasted resources and excessively long repair times, a reconnection threshold is preset. This reconnection threshold can be flexibly set according to actual needs; for example, the number of reconnection attempts can be set to 3. Each time a reconnection attempt is made, the network settings module records the number of connection attempts and compares it with the reconnection threshold.

[0049] If the router name cannot be found during the search process, meaning that even after multiple searches (reaching the reconnection threshold), the target router's wireless network signal cannot be found, then the router is considered to be malfunctioning. This could be due to a faulty router, the router being turned off, or a problem with the router's wireless network signal transmission function.

[0050] If the router name is found, but a connection cannot be established after multiple attempts (reaching the reconnection threshold), it is determined that the first network connection module is in an abnormal state. This situation may be due to hardware failure, software failure, driver issues, or other reasons within the first network connection module (wireless Wi-Fi network module), preventing it from establishing a normal connection with the router.

[0051] When the first network connection module is determined to be abnormal, in order to attempt to restore the data connection through the backup network connection method and ensure that the basic functions of the camera are not greatly affected, the operation of disabling the first network connection module (i.e., the wireless WIFI network module) and activating the second network connection module (i.e., the data roaming network module of the secondary SIM card) will be performed.

[0052] During the module switching operation, the first network connection module is first shut down, stopping all network connection-related operations, including ceasing the sending and receiving of wireless network signals and disabling related hardware drivers. Then, the second network connection module is started, including initializing related hardware devices, loading drivers, and configuring network parameters, ensuring that the second network connection module can quickly enter a working state and attempt to establish a connection with the external network using the backup network connection method.

[0053] S5, the second network connection module starts the data connection mode, establishes a connection with the main control backend system, sends the network parameters of the faulty camera to the main control backend system, the main control backend system connects to the router before the camera disconnected and obtains the network log of the router, obtains the abnormal parameters of the camera disconnection, and repairs it in the router's log and network settings;

[0054] Establish connection with the main control backend system:

[0055] When a router malfunction is detected, the first network connection module is disabled and the second network connection module is activated. The second network connection module then initiates a data connection, actively establishing a connection with the main control system. During the connection establishment process, certain communication protocols and security mechanisms are followed to ensure connection stability and data transmission security.

[0056] For example, encrypted communication protocols, such as SSL / TLS, might be used to encrypt transmitted data, preventing it from being stolen or tampered with during transmission. Simultaneously, authentication procedures would be performed to ensure that both parties in the connection (the camera and the main control system) can accurately identify each other, preventing unauthorized access.

[0057] Data transmission and analysis:

[0058] Once a connection is established with the main control system, the camera will send its complete network parameters to the main control system. These network parameters include all relevant information previously recorded in the network logs, such as IP address, subnet mask, gateway, network name (SSID), and network password, so that the main control system can have a comprehensive understanding of the camera's network settings.

[0059] Upon receiving these parameters, the main control system immediately establishes a connection with the router from which the camera disconnected. During the connection process, it adheres to the corresponding communication protocols and security mechanisms to ensure the connection's legitimacy and stability. Then, the main control system retrieves the router's network logs and network parameters. Through detailed analysis of this information, it can accurately identify the abnormal parameters causing the camera's disconnection, and subsequently perform targeted repairs on the router's logs and network settings.

[0060] Router repair procedures:

[0061] When repairing router logs and network settings, the main control system will adjust the relevant router settings based on the abnormal parameters obtained. For example, if an incorrect IP address is found, it will be corrected; if the network name (SSID) or network password does not match, it will be adjusted or reset accordingly.

[0062] Meanwhile, for other issues that may affect network connectivity reflected in the router's network logs, such as network congestion and signal interference, the main control system will also take corresponding measures to handle them according to the specific circumstances. For example, for network congestion issues, it may alleviate the problem by adjusting the router's bandwidth allocation strategy; for signal interference issues, it may suggest that the user adjust the router's location or change the frequency band.

[0063] Error handling:

[0064] If a network connection failure occurs during the connection process between the main control system and the router, it is determined that the router's network connection is abnormal, and a repair request will be promptly sent to the user. When sending the request, the user will be clearly informed that the router may have a serious network problem requiring further inspection and repair. The request will also provide possible cause analyses and suggestions to help the user better understand the problem and how to resolve it.

[0065] S6, enable the first network connection module and disable the second network connection module, obtain the network connection status, if the connection with the router is still disconnected, disable the first network connection module and enable the second network connection module, determine that it is a hardware failure, send the number and network parameters of the faulty camera to the main control backend system, and send a repair request to the user.

[0066] Preferably, when the router is determined to be abnormal, the first network connection module is disabled, the second network connection module is enabled, and a connection is established with the main control backend system through the second network connection module. The main control backend system connects with the router and obtains the router's network logs and network parameters, and performs abnormal repair on the router. Preferably, when the main control backend system cannot establish a network connection with the router, it is determined that the router's network connection is abnormal, and a repair request is sent to the user.

[0067] A device includes at least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the above-described method for repairing signal interruption of a digital camera.

[0068] A storage medium storing computer program instructions, characterized in that: when the computer program instructions are executed by a processor, the above-mentioned signal interruption repair method for a digital camera is implemented.

[0069] Example 2

[0070] This embodiment provides a method for repairing signal interruption in a digital camera, including the following steps:

[0071] S1, define the scope of data collection, comprehensively collect data related to digital camera signal interruption from multiple sources, and ensure the diversity and integrity of the data in order to provide rich information for neural network training.

[0072] Camera's own data: In-depth analysis of the camera's own operational logs, covering the dynamic changes of various key parameters over time. For example, detailed records of real-time signal strength values, frame rate changes, resolution settings, and fluctuations in internal device temperature. Simultaneously, precise capture of the specific timestamps of each signal interruption, as well as the overall device status of the camera just moments before the interruption, such as whether it was in different operating modes like busy recording, live streaming, or relatively static shooting.

[0073] Network Environment Data: Comprehensive collection of information regarding the network environment in which the camera is located. This includes detailed router logs, with a focus on real-time changes in network traffic, the number of devices currently connected to the router, and the usage of various frequency bands. Furthermore, it is necessary to investigate potential sources of wireless network signal interference in the area, such as the frequency bands and transmission power used by other nearby wireless devices, as these factors can potentially affect the camera's signal transmission.

[0074] Manual data recording: Designated personnel are responsible for recording the surrounding environmental conditions and the specific operation of the camera each time a signal interruption occurs. Detailed records should be kept of environmental factors such as light intensity, ambient temperature, and humidity, as well as the specific operations the camera was performing (e.g., recording high-definition video, live streaming, simple still photography). Simultaneously, all subsequent remedial measures taken in response to the signal interruption should be recorded, along with their actual effects, such as whether the signal was successfully restored and the stability of the restored signal.

[0075] S2, Data Cleaning and Preliminary Processing: This involves meticulous cleaning and preliminary processing of the collected raw data to improve data quality, ensure that the data accurately reflects the actual situation, and lay a good foundation for subsequent neural network training.

[0076] Handling Missing Values: For missing values ​​in the dataset, appropriate imputation methods are employed based on the characteristics of different parameters. For example, for a key parameter like signal strength, if missing values ​​occur, the trend of signal strength changes over consecutive time periods can be analyzed first. Then, the average imputation method can be used, which calculates the average signal strength over adjacent time periods and fills the missing value's location with this average. Similar appropriate imputation methods can be used for other similar numerical parameters, such as equipment temperature, depending on the specific circumstances.

[0077] Identifying and Handling Outliers: Outliers in the dataset are identified by setting reasonable threshold ranges. For example, based on past experience and industry standards, the normal signal strength range is determined to be within a certain interval (e.g., -30dBm to -10dBm). If a signal strength value exceeds this range, it is considered an outlier. For identified outliers, their causes are further analyzed. If the outlier is due to temporary equipment failure or brief interference, but subsequently returns to normal, the data can be retained and labeled. If the data is clearly erroneous (e.g., unreasonable values ​​such as extremely high or low signal strength), it should be deleted directly to avoid interfering with subsequent analysis and training.

[0078] Remove duplicate data: Carefully examine the dataset and remove duplicate data records to ensure that each data point has unique information value, thus avoiding inaccurate patterns learned by the model during neural network training due to duplicate data.

[0079] S3, Data Labeling and Classification, involves accurately labeling and classifying the cleaned and preliminarily processed data, enabling the neural network to clearly identify different states and types of signal interruptions, thereby improving the model's prediction accuracy.

[0080] Signal state labeling: Each data record in the dataset is labeled according to the actual signal condition. Periods of normal signal transmission are labeled "normal," and periods of signal interruption are explicitly labeled "interrupted." This simple and clear labeling method allows the neural network to first distinguish between the two basic signal states: normal and interrupted.

[0081] Detailed Interruption Cause Labeling: Building upon the initial labeling of signal interruptions as "interruptions," the data is further subdivided based on the specific cause of the interruption. For example, an interruption caused by insufficient signal strength is labeled "weak signal interruption"; an interruption caused by network congestion due to excessive simultaneous data transmission is labeled "network congestion interruption"; an interruption caused by overheating of the camera equipment due to prolonged operation, affecting signal transmission, is labeled "device overheating interruption"; an interruption caused by hardware failure of the camera itself (such as a damaged wireless network card or antenna failure) is labeled "other hardware failure interruption"; and an interruption caused by software problems (such as camera driver failure or conflicts with related applications) is labeled "software problem interruption," etc. Through such detailed labeling, the neural network can more accurately learn the characteristic patterns of signal interruptions caused by different reasons during training, thereby accurately identifying the cause of the interruption and taking targeted remedial measures in subsequent fault diagnosis.

[0082] S4, Data Formatting and Set Partitioning: The cleaned and labeled dataset is organized in a format suitable for neural network input and reasonably divided into different subsets for model training, validation, and testing, ensuring the scientific nature and effectiveness of model training.

[0083] S41, Data Formatting: This process organizes each data record into a format acceptable to the neural network. Typically, each data record is formatted as a vector, where each element corresponds to a parameter value related to a signal interruption and a signal interruption label. For example, a vector might contain elements such as signal strength, network traffic, device temperature, camera operation mode encoding (e.g., 1 for recording, 2 for live streaming, 3 for still photography), and signal interruption labels (e.g., 0 for normal, 1 for interruption). This formatting process allows the data to be smoothly input into the neural network for processing.

[0084] S42, Dataset Partitioning: Divide the prepared dataset into training, validation, and test sets according to a certain ratio. A common ratio is 70% for training, 20% for validation, and 10% for test. In practice, the partitioning can be flexible depending on the actual size of the dataset. For example, if 1000 cleaned and labeled valid data records are collected, then 700 records can be used as the training set, 200 as the validation set, and 100 as the test set. The training set is used to allow the neural network to learn patterns and rules in the data; the validation set is used to evaluate and adjust the model's performance during training to prevent overfitting; the test set is used to evaluate the model's final performance after training to determine whether the model can accurately predict actual situations.

[0085] S5, Neural Network Model Selection and Construction;

[0086] S51. Analyze the characteristics and needs of the data, select a suitable neural network architecture. Based on the characteristics of the data collected, organized and analyzed in the early stage, as well as the specific needs for digital camera signal interruption repair, carefully select the most suitable neural network architecture to ensure that the model can effectively process the input data and accurately predict the signal interruption situation.

[0087] S52, Multilayer Perceptron (MLP) Considerations: If the collected data mainly consists of discrete parameter values, and the relationships between these parameters are relatively simple and direct—for example, primarily single numerical parameters such as signal strength, device temperature, and network traffic—and their relationship with signal interruption is relatively clear, then a multilayer perceptron (MLP) might be a suitable choice. MLPs can effectively handle this feature vector-based input data. Through their multilayered neuron structure, they perform nonlinear transformations on the input data, thereby enabling the classification or prediction of signal interruption conditions.

[0088] S53, Convolutional Neural Networks (CNNs): When data has a certain spatial or temporal correlation, CNNs may be more advantageous. For example, if we consider that video frame data from a camera may have some potential correlation with signal interruption (e.g., image features in the video frame may reflect some anomalies before the signal interruption), or that signal strength changes over a continuous time period have certain spatial distribution characteristics (e.g., the trend of signal strength changes in a certain area over different time periods), then CNNs can use their unique structures such as convolutional layers and pooling layers to mine the relationship between potential features in video frame images and signal interruption, or capture the changing patterns of parameters such as signal strength at different time points and their correlation with interruption.

[0089] S54, Recurrent Neural Networks (RNNs): If the data exhibits clear time-series characteristics, meaning that the changes in various parameters (such as signal strength, device temperature, etc.) are continuous and dependent over time, then Recurrent Neural Networks (RCCNs) and their variants (such as Long Short-Term Memory Networks (LSTM), Gated Recurrent Units (GRUs), etc.) may be a better choice. RNNs can handle this kind of time-series data very well. Through their internal recurrent structure, they remember the state information from previous moments, thereby capturing the changing patterns of parameters such as signal strength at different points in time and their correlation with interruptions.

[0090] S55. Determine the specific parameter settings of the neural network. After selecting a suitable neural network architecture, it is necessary to further determine the specific parameter settings of the network, including the number of nodes in the input layer, hidden layer and output layer, to ensure that the network can accurately process the input data and output the expected prediction results.

[0091] S551, Input Layer Settings: The number of nodes in the input layer is determined based on the dimension of the input data vector, i.e., the number of parameters related to signal interruption considered. For example, if 10 relevant parameters such as signal strength, network traffic, device temperature, and camera operating mode are considered during the data collection phase, then the input layer is set to 10 nodes, so that each node corresponds to one input parameter, ensuring that the data is accurately input into the neural network for processing.

[0092] S552, Hidden Layer Settings: The number of hidden layers and the number of nodes per layer need to be determined through experimentation and adjustment. Generally, start with a smaller number of hidden layers and nodes, and then gradually increase or adjust based on performance during training. For example, start with 2 hidden layers, each with 20 nodes. During training, closely monitor evaluation metrics such as accuracy and recall on the validation set. If performance is unsatisfactory, try increasing the number of hidden layers or nodes. However, when increasing the number of hidden layers or nodes, be careful to avoid overfitting, i.e., the model becoming too fitted to the training data, leading to a decrease in performance on the test set.

[0093] S553 Output Layer Settings: The number of nodes in the output layer is determined based on the target to be predicted. In a digital camera signal interruption repair scenario, the output layer can be set to 2 nodes, representing "normal" and "interrupted" states respectively, so that the neural network can directly determine whether the signal is interrupted; or multiple nodes can be set according to different interruption reasons, such as 5 nodes, corresponding to different types of interruption situations such as "weak signal interruption", "network congestion interruption", "device overheating interruption", "other hardware failure interruption", and "software problem interruption", so as to more accurately determine the cause of the interruption and take targeted repair measures.

[0094] S6, Model Training: Setting Training Parameters. Properly setting the training parameters of the neural network is crucial for the model's training effectiveness and performance. It directly relates to whether the model can accurately learn patterns and rules in the data and play a positive role in subsequent fault diagnosis.

[0095] S61, Learning Rate Setting: The learning rate determines the step size at which the model updates weights in each iteration. An excessively large learning rate may prevent the model from converging, meaning it cannot find a stable optimal solution during training, preventing the error between the predicted and actual results from continuously decreasing. An excessively small learning rate may make the training process too slow, requiring a significant amount of time to achieve good model performance. Generally, a learning rate of 0.01 can be set initially, and then adjusted during training based on the model's performance on the validation set. If the model's accuracy, recall, or other metrics on the validation set do not show significant improvement, or exhibit large fluctuations, it may be necessary to adjust the learning rate appropriately, such as increasing or decreasing it by a certain percentage (e.g., a step size of 0.005), and then retrain.

[0096] S62, Iteration Count Setting: The iteration count refers to the number of training rounds of the model. Generally, as the number of iterations increases, the model's performance gradually improves. However, after a certain point, overfitting may occur, meaning the model fits the training data too closely, leading to a decrease in performance on the test set. You can start by setting the iteration count to 1000. During training, observe the performance on the validation set. If signs of overfitting appear, such as a decline in accuracy and recall on the validation set after a certain period, you can appropriately reduce the iteration count, for example, to 800 or another suitable value, and then retrain.

[0097] S63, Batch Size Setting: Batch size refers to the amount of data input into the model during each training session. An appropriate batch size can improve training efficiency. Common batch sizes include 32 and 64. You can start with 32 as the batch size and adjust it based on the training results. If you find that the model converges slowly or encounters memory shortages during training, you can consider increasing the batch size, such as adjusting it to 64. If you find that the model is overfitting during training, you can also try decreasing the batch size, such as adjusting it to 16, and then retraining.

[0098] S64, Start Training: Input the training set data into the built neural network model and start training according to the set training parameters. The model will continuously learn the patterns and rules in the data and adjust its own weights and biases to minimize the error between the prediction results and the actual results.

[0099] S65, For the Multilayer Perceptron (MLP) model: The training process involves continuously updating the network's weights and biases using the backpropagation algorithm to minimize the error between the network's predicted output and the actual result. Specifically, in each iteration, a batch of training data (according to a set batch size) is first input into the network. After forward propagation, the network's output is obtained. Then, based on the error between the output and the actual result, the gradient of each node is calculated using the backpropagation algorithm, and the network's weights and biases are updated accordingly. By repeatedly performing this process until the set number of iterations is reached, the error between the network's output and the actual result gradually decreases, eventually reaching a relatively stable minimum value.

[0100] S66, For Convolutional Neural Network (CNN) models: While also trained using the backpropagation algorithm, the training process is relatively more complex due to the unique structure of CNNs, including convolutional and pooling layers. In each iteration, a batch of training data is first input into the network. After a series of operations such as convolution in the convolutional layers and pooling in the pooling layers, the network output is obtained. Then, based on the error between the output and the actual result, the gradient of each node is calculated using the backpropagation algorithm, and the network weights and biases are updated accordingly. This process is repeated continuously until a set number of iterations is reached, gradually reducing the error between the network's output and the actual result until a relatively stable minimum value is reached.

[0101] S67, For Recurrent Neural Network (RNN) models: Their training process is also based on the backpropagation algorithm, but due to the internal recurrent structure of RNNs, their training process has some unique aspects. In each iteration, a batch of training data is first input into the network. After processing by the RNN's recurrent structure, the network's output is obtained. Then, based on the error between the output and the actual result, the gradient of each node is calculated using the backpropagation algorithm, and the network's weights and biases are updated accordingly. By continuously repeating this process until the set number of iterations is reached, the error between the network's output and the actual result gradually decreases, eventually reaching a relatively stable minimum value.

[0102] S7, Model Evaluation and Adjustment: Evaluate model performance on the validation set. During model training, periodically evaluate the model's performance on the validation set to understand the training effect in a timely manner, identify potential problems, and adjust the model based on the evaluation results to ensure that the model can accurately predict the actual situation.

[0103] S71. Selecting Evaluation Metrics: Choose appropriate evaluation metrics to measure model performance based on specific circumstances. Common evaluation metrics include accuracy, recall, and F1 score. Accuracy refers to the proportion of correctly predicted samples out of the total number of predicted samples; recall refers to the proportion of correctly predicted samples out of the actual number of samples with interrupted signals; the F1 score is a metric that comprehensively considers accuracy and recall, calculated as: F1 = 2 * (Accuracy * Recall) / (Accuracy + Recall). These metrics reflect the model's predictive ability from different perspectives. By comprehensively analyzing these metrics, a complete understanding of the model's performance can be achieved.

[0104] S72, Calculate Evaluation Metrics: On the validation set, calculate the corresponding evaluation metric values ​​based on the model's predictions and actual results. For example, assuming the number of correctly predicted samples on the validation set is 80 and the total number of predicted samples is 100, then the accuracy = 80 / 100 = 0.8; assuming the number of samples with actual signal interruptions is 60, then the recall = 80 / 60 = 1.33 (here, the recall is greater than 1 because the number of correctly predicted samples is greater than the number of samples with actual signal interruptions, which may occur in reality, such as predicting some cases that were originally thought to be normal but were actually interruptions); the F1 score = 2 * (0.8 * 1.33) / (0.8 + 1.33) = 1.02 (this is just for demonstrating the calculation process; in reality, the F1 score should be between 0 and 1). By regularly calculating these evaluation metrics, we can understand the performance changes of the model on the validation set in a timely manner, so as to take timely measures to adjust it.

[0105] S73. Adjust the model based on the evaluation results. If the model's performance is found to be unsatisfactory based on the evaluation results on the validation set, such as low accuracy, recall, or F1 score, the model needs to be adjusted according to the specific situation to improve its performance and enable it to more accurately predict the actual situation.

[0106] S74, Addressing Overfitting: If poor performance is due to model overfitting, the following adjustments may be necessary. First, try reducing the number of hidden layers or nodes per layer. This reduces the model's fitting ability, preventing it from overfitting the training data and improving performance on the test set. For example, if the model's accuracy on the validation set is low and analysis reveals overfitting, try reducing the number of hidden layers from 2 to 1, or add an L2 regularization term to the loss function to improve performance. Adding a regularization term can limit the model's fitting ability to some extent, making it focus more on the general patterns of the data during training rather than the details.

[0107] S75, Addressing Underfitting: If underfitting is causing poor performance, the following adjustments may be necessary. First, try increasing the number of hidden layers or nodes per layer. This can improve the model's fitting ability, allowing it to better learn patterns and rules in the data. For example, if the model's accuracy on the validation set is low, and analysis reveals underfitting, try increasing the number of hidden layers from 2 to 3, or increasing the number of nodes per layer from 20 to 30 (the specific increase should be determined based on actual conditions and multiple trials). Simultaneously, adjust training parameters such as the learning rate. If the learning rate is set too low, the model may update weights with too small a step size during training, making it difficult for the model to quickly learn effective information from the data, resulting in underfitting. In this case, try increasing the learning rate appropriately, for example, from 0.001 to 0.01 (again, the specific adjustment should be determined based on actual training results), so that the model can update weights with a more suitable step size during training, accelerating the learning speed and better fitting the data. Additionally, extending the number of iterations can be considered, giving the model more training rounds to fully learn the patterns and rules in the data. For example, if the original iteration count was set at 500, it could be increased to 800 or even more. However, during the process of increasing the number of iterations, it is crucial to continuously monitor the model's performance on the validation set to prevent overfitting. Through these comprehensive adjustments, it is hoped that the unsatisfactory model performance caused by underfitting can be improved, enabling the model to more accurately predict and judge signal interruptions from digital cameras.

[0108] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for repairing signal interruption in a digital camera, characterized in that, Includes the following steps: S1, a first network connection module and a second network connection module are pre-built into the digital camera, wherein the second network connection module is a backup network module, which provides an alternative network access method when the first network connection module fails and cannot connect normally, ensuring that the camera can still maintain data transmission capability. S2, the wireless network signal connection between the wireless network card and the router is monitored in real time through the wireless network card of the digital camera, and an interrupt signal is sent to the network setting module after a disconnection is detected. S3, the network settings module detects and resets the network parameters of the digital camera, and then checks again whether to reconnect; S4, according to step S3, when in the disconnected state, re-search for and reconnect to the wireless network signal; when still in the disconnected state, disable the first network connection module and start the second network connection module. S5, the second network connection module starts the data connection mode, establishes a connection with the main control backend system, sends the network parameters of the faulty camera to the main control backend system, the main control backend system connects to the router before the camera disconnected and obtains the network log of the router, obtains the abnormal parameters of the camera disconnected based on the network log, and repairs the network settings of the router. S6, enable the first network connection module and disable the second network connection module, obtain the network connection status, if the connection with the router is still disconnected, disable the first network connection module and enable the second network connection module, determine that it is a hardware failure, send the number and network parameters of the faulty camera to the main control backend system, and send a repair request to the user.

2. The method for repairing signal interruption in a digital camera according to claim 1, characterized in that, The first network connection module and the second network connection module in step S1 are respectively the wireless WIFI network module and the data roaming network module of the SIM card sub-card.

3. The method for repairing signal interruption in a digital camera according to claim 1, characterized in that, In step S1, the first network connection module and the second network connection module of the digital camera record network logs, which include network parameters, connection status and signal strength, as well as the total amount of data received and sent.

4. The method for repairing signal interruption in a digital camera according to claim 1, characterized in that, In step S3, the reset specifically refers to resetting the real-time network to the network parameters of the most recent normal connection state.

5. A method for repairing signal interruption in a digital camera according to claim 1, characterized in that, In step S4, if the wireless network signal is re-searched and reconnected, and the connection is still lost, the first network connection module is disabled and the second network connection module is started. Specifically, this includes re-searching and reconnecting to the wireless network signal, obtaining the name of the most recently connected router, reconnecting to the router with that name, and setting a reconnection threshold. If the router name cannot be found, it is determined that the router is abnormal, the first network connection module is disabled, and the second network connection module is started. If the router name is found but a connection cannot be established, it is determined that the first network connection module is abnormal, the first network connection module is disabled, and the second network connection module is started.

6. A method for repairing signal interruption in a digital camera according to claim 5, characterized in that, If, after determining that the router is malfunctioning, disabling the first network connection module, and enabling the second network connection module to repair the router, the main control backend system cannot establish a network connection with the router, then the router is deemed to have a network connection malfunction, and a repair request is sent to the user.

7. A signal interruption repair device for a digital camera, comprising at least one processor, at least one memory, and computer program instructions stored in the memory, wherein when the computer program instructions are executed by the processor, the method described in any one of claims 1-6 is implemented.

8. A storage medium storing computer program instructions thereon, characterized in that: The method described in any one of claims 1-6 is implemented when the computer program instructions are executed by the processor.

Images