Data processing method and device, storage medium and electronic equipment

Abstract

The invention discloses a data processing method and device, a storage medium and electronic equipment. The method comprises the following steps: acquiring abnormal data associated with image acquisition equipment; inputting the abnormal data into a target identification model to obtain a target identification code; a target processing instruction is determined based on a rule mapping relation and the target identification code, the rule mapping relation comprises a plurality of abnormal identification codes, a plurality of processing instructions and a mapping relation between the processing instructions and the abnormal identification codes, and the target identification code and the target processing instruction have the target mapping relation; and repairing the image acquisition equipment based on the target processing instruction. According to the method and the device, the technical problem of relatively low abnormity repairing efficiency of the image acquisition equipment caused by a complicated abnormity repairing process of the image acquisition equipment is solved.

Description

Technical Field

[0001] The present application relates to the field of computers, and in particular to a data processing method and device, a storage medium, and an electronic device. Background Art

[0002] At present, when an image acquisition device has an abnormality, it is often necessary to contact the technical support personnel of the image acquisition device and others to specifically collect the operation log of the image acquisition device and the network packet capture data. Then, the developer of the image acquisition device performs abnormal location analysis of the image acquisition device based on the operation log and the network packet capture data and provides a repair plan. That is, the related technology relies too much on manual labor to locate and detect the abnormality of the image acquisition device and the subsequent abnormality processing process, resulting in a long repair time. Therefore, the related technology has a technical problem that the abnormality repair process of the image acquisition device is complicated, resulting in low efficiency of the abnormality repair of the image acquisition device.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present application provide a data processing method and device, a storage medium, and an electronic device to at least solve the technical problem that the abnormality repair efficiency of the image acquisition device is low due to the complex abnormality repair process of the image acquisition device.

[0005] According to one aspect of an embodiment of the present application, a data processing method is provided, comprising: acquiring abnormal data associated with an image acquisition device, wherein the abnormal data represents operating data corresponding to an abnormality occurring in the image acquisition device; inputting the abnormal data into a target recognition model to obtain a target identification code, wherein the target recognition model is used to identify the abnormal data and determine an abnormality type of the abnormal data, and the target identification code is used to identify the abnormality type; determining a target processing instruction based on a rule mapping relationship and the target identification code, wherein the rule mapping relationship includes a plurality of abnormal identification codes, a plurality of processing instructions, and a mapping relationship between the processing instructions and the abnormal identification code, and a target mapping relationship exists between the target identification code and the target processing instruction; and repairing the image acquisition device based on the target processing instruction.

[0006] According to another aspect of an embodiment of the present application, a data processing device is also provided, including: an acquisition module, used to acquire abnormal data associated with an image acquisition device, wherein the abnormal data represents the corresponding operating data when an abnormality occurs in the image acquisition device; a first determination module, used to input the abnormal data into a target recognition model to obtain a target identification code, wherein the target recognition model is used to identify the abnormal data and determine the abnormal type of the abnormal data, and the target identification code is used to identify the abnormal type; a second determination module, used to determine a target processing instruction based on a rule mapping relationship and the target identification code, wherein the rule mapping relationship includes multiple abnormal identification codes, multiple processing instructions and a mapping relationship between the processing instructions and the abnormal identification code, and there is a target mapping relationship between the target identification code and the target processing instruction; a repair module, used to repair the image acquisition device based on the target processing instruction.

[0007] Optionally, the device is used to input the abnormal data into a target recognition model to obtain a target identification code in the following manner: using a word segmentation processing module to convert characters in the abnormal data into character vectors; inputting the character vectors into a bidirectional long short-term memory network module to determine the probability of the character belonging to each identification code in an identification code set, and determining the identification code whose occurrence probability meets a preset probability condition as the initial identification code; inputting the initial identification code into a sequence labeling module to determine the target identification code, wherein the target recognition model includes the word segmentation processing module, the bidirectional long short-term memory network module, and the sequence labeling module, and the sequence labeling module is used to adjust the positions of each named entity in the initial identification code.

[0008] Optionally, the device is used to input the initial identification code into a sequence labeling module to determine the target identification code in the following manner: input the initial identification code into the sequence labeling module to determine multiple first identification codes; input the multiple first identification codes into a multilayer perceptron to determine the weight value corresponding to each of the first identification codes; perform a weighted operation on each of the first identification codes according to the weight value to obtain the target identification code.

[0009] Optionally, the device is also used to: input the initial identification code into the sequence labeling module, and after determining multiple first identification codes, obtain a second identification code sent by a target playback platform, wherein the target playback platform is used to receive images captured by the image acquisition device, and the second identification code is used to indicate a communication abnormality between the target playback platform and the image acquisition device; input the multiple first identification codes and the second identification code into a multi-layer perceptron to determine the weight value corresponding to each of the first identification code and the second identification code; perform a weighted sum operation on each of the first identification code and the second identification code according to the weight value to obtain the target identification code.

[0010] Optionally, the device is used to determine the target processing instruction based on the rule mapping relationship and the target identification code in the following manner: obtaining a rule mapping file, wherein the rule mapping file is used to store the rule mapping relationship; loading the rule mapping file into a rule engine container, wherein the rule engine container is used to perform rule matching according to the rule mapping relationship; inputting the target identification code into the rule engine container, and determining the target processing instruction corresponding to the target identification code from a plurality of the processing instructions based on the target mapping relationship.

[0011] Optionally, the apparatus is used to obtain abnormal data associated with the image acquisition device in the following manner: when the communication connection between the image acquisition device and the target playback platform is disconnected and the communication connection between the image acquisition device and the target detection platform is not disconnected, obtaining a first trigger request issued by the image acquisition device, wherein the first trigger request indicates that the image acquisition device requests to stop sending images to the target playback platform and to send the abnormal data to the target detection platform; and obtaining the abnormal data in response to the first trigger request.

[0012] Optionally, the apparatus is used to obtain abnormal data associated with the image acquisition device in the following manner: in response to a target interaction operation, sending a second trigger request to the image acquisition device, wherein the second trigger request is used to request the image acquisition device to stop sending images to a target playback platform and send the abnormal data to a target detection platform; in response to receiving a target confirmation message sent by the image acquisition device, obtaining the abnormal data.

[0013] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned data processing method when running.

[0014] According to another aspect of the embodiments of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the above data processing method.

[0015] According to another aspect of the embodiments of the present application, there is also provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the data processing method through the computer program.

[0016] In an embodiment of the present application, an abnormal data associated with an image acquisition device is obtained, wherein the abnormal data represents the corresponding operating data when an abnormality occurs in the image acquisition device; the abnormal data is input into a target recognition model to obtain a target identification code, wherein the target recognition model is used to identify the abnormal data and determine the abnormal type of the abnormal data, and the target identification code is used to identify the abnormal type; a target processing instruction is determined based on a rule mapping relationship and a target identification code, wherein the rule mapping relationship includes multiple abnormal identification codes, multiple processing instructions, and a mapping relationship between processing instructions and abnormal identification codes, and there is a target mapping relationship between the target identification code and the target processing instruction; a method of repairing the image acquisition device based on the target processing instruction achieves the purpose of improving the abnormality repair efficiency through an automated abnormality detection and processing process, thereby achieving the technical effect of quickly and accurately identifying and processing the abnormality of the image acquisition device, and further solving the technical problem that the abnormality repair process of the image acquisition device is complicated, resulting in low abnormality repair efficiency of the image acquisition device. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 is a schematic diagram of an application environment of an optional data processing method according to an embodiment of the present application;

[0019] Figure 2 is a flowchart of an optional data processing method according to an embodiment of the present application;

[0020] Figure 3 is a schematic diagram of an optional data processing method according to an embodiment of the present application;

[0021] Figure 4 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0022] Figure 5 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0023] Figure 6 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0024] Figure 7 is a schematic diagram of another optional data processing method according to an embodiment of the present application;

[0025] Figure 8 is a schematic structural diagram of an optional data processing device according to an embodiment of the present application;

[0026] Fig. 9 is a schematic diagram of the structure of an optional data processing product according to an embodiment of the present application;

[0027] Fig.10 It is a schematic diagram of the structure of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] The present application is described below in conjunction with embodiments:

[0031] According to one aspect of an embodiment of the present application, a data processing method is provided. Optionally, in this embodiment, the data processing method can be applied to Figure 1 In the hardware environment composed of the server 101 and the terminal device 103 shown in FIG. Figure 1As shown, the server 101 is connected to the terminal 103 via a network, and can be used to provide services for the terminal device or an application 107 installed on the terminal device. The application can be a video application, an instant messaging application, a browser application, an educational application, a game application, etc. A database 105 may be set up on the server or independently of the server to provide data storage services for the server 101, for example, a game data storage server. The above-mentioned network may include, but is not limited to, a wired network and a wireless network, wherein the wired network includes, for example, a local area network, a metropolitan area network, and a wide area network; the wireless network includes, for example, Bluetooth, WIFI, and other networks that implement wireless communication; the terminal device 103 may be a terminal configured with an application program, and may include, but is not limited to, at least one of the following: a mobile phone (such as an Android phone, an iOS phone, etc.), a laptop computer, a tablet computer, a PDA, a MID (Mobile Internet Devices), a PAD, a desktop computer, a smart TV, an intelligent voice interaction device, a smart home appliance, a vehicle-mounted terminal, an aircraft, a virtual reality (VR) terminal, an augmented reality (AR) terminal, a mixed reality (MR) terminal, and other computer devices; the above-mentioned server may be a single server, a server cluster consisting of multiple servers, or a cloud server.

[0032] Combination Figure 1 As shown, the above data processing method can be executed by an electronic device, which can be a terminal device or a server. The above data processing method can be implemented by the terminal device or the server respectively, or by the terminal device and the server together.

[0033] The above is only an example and is not specifically limited in this embodiment.

[0034] Optionally, as an optional implementation, as Figure 2 As shown, the method for processing the above data includes:

[0035] S202, acquiring abnormal data associated with the image acquisition device, wherein the abnormal data represents the corresponding operation data when the image acquisition device is abnormal;

[0036] Optionally, in an embodiment of the present application, the above-mentioned image acquisition device may include but is not limited to a camera, a mobile phone, a notebook, a smart wearable device, etc., and the above-mentioned abnormal data may include but is not limited to a device operation log, network packet capture data, and a system error report.

[0037] For example, when the image acquisition device cannot capture video normally, the device operation log and network packet capture data of the image acquisition device can be obtained (for example, if the image acquisition device communicates with the target playback platform via the https protocol, the TCPDUMP file can be captured).

[0038] S204, inputting the abnormal data into a target recognition model to obtain a target identification code, wherein the target recognition model is used to identify the abnormal data and determine the abnormal type of the abnormal data, and the target identification code is used to identify the abnormal type;

[0039] Optionally, in an embodiment of the present application, the above-mentioned target recognition model may include but is not limited to a deep learning model based on BERT, LSTM, CRF named entity recognition and multi-layer perceptron, and the above-mentioned target identification code may include the error information code of the above-mentioned image acquisition device, for example, using "ERROR1" to indicate an authorization permission error, using "ERROR2" to indicate a login error, etc.; the type of the above-mentioned abnormal data may include but is not limited to a device firmware error type or a network communication error type, etc.

[0040] For example, Figure 3 is a schematic diagram of an optional data processing method according to an embodiment of the present application. The target recognition model is as follows: Figure 3 As shown, including BERT, LSTM, CRF, etc., the abnormal data is input into the target recognition model. The model will identify the cause of the abnormal data based on its training data and algorithm. For example, if the model identifies that the abnormal data is caused by a device firmware error, the model will output a target identification code, such as "ERROR3". If the model identifies that the abnormal data is caused by a network communication error, it will output another target identification code, such as "ERROR4".

[0041] It should be noted that BERT is a pre-trained language model based on the Transformer architecture that can capture bidirectional contextual information. The BERT model is pre-trained on a large-scale corpus and can then be fine-tuned to adapt to specific tasks. BERT can be used to process input text and extract rich semantic information. It can be used for encoding and representation learning of text to better understand questions and answers in question answering. LSTM is a variant of recurrent neural network (RNN) that is specifically designed to process sequence data. It uses a gating mechanism to capture long-term dependencies and is suitable for processing sequence data such as time series and natural language. LSTM can be used to process sequence information in information text, such as error messages, special keywords, etc. It helps to retain contextual information in the text and improve the model's ability to understand long texts. CRF is a statistical modeling method for annotating sequence data. In sequence annotation tasks, CRF can consider the dependencies between tags to better capture sequence structure. In medical texts, CRF can be used for named entity recognition (NER) tasks, such as identifying proper nouns, error codes, fixed hexadecimal / binary combination protocols, etc. By introducing the CRF layer, the consistency between labels and the accuracy of overall sequence labeling can be improved).

[0042] In an exemplary embodiment, abnormal data is input into the target recognition model, and three identification codes are obtained after being processed by the CRF direction matrix, namely: identification code A, identification code B, and identification code C, wherein different identification codes represent different types of abnormal data, that is, after training the key information classification model extracted by NER, the abnormal data can be identified and divided into useless data, various types of error codes (identification code A, identification code B, identification code C), and various types of protocols. Then, the multi-layer perceptron is entered for error classification, and a new single error code is determined based on identification code A, identification code B, and identification code C, that is, the above-mentioned target identification code.

[0043] For example, when the platform login fails, the device fails and the cloud service cannot be used. Through data NER sorting, the HTTPS hexadecimal error code of the platform login error is found in the abnormal data. Then, the HTTPS hexadecimal error code is input into the perceptron, and the classification result is that the target identification code is 0x10040004, indicating that the current user authority check fails.

[0044] For another example, if the target playback platform also sends an identification code D, the classification results can be balanced by the weight of each error ratio pre-set in the multi-layer perceptron, and a new single error code, that is, the above-mentioned target identification code, is determined based on identification code A, identification code B, identification code C, and identification code D, wherein the method of balancing the classification results by the weight of each error ratio pre-set in the multi-layer perceptron can include but is not limited to being achieved by setting the weight of each error code, and the weight of each error code satisfies the following conditions:

[0045] The balance degree δ=W1f+W2μ+W3c…+WnΦ, where w1+w2+..Wn=1, f, μ, c, etc. represent error codes, and w1 to Wn represent the weights of each error code.

[0046] Specifically, Figure 4 is a schematic diagram of an optional data processing method according to an embodiment of the present application. The neural network structure of the multilayer perceptron can be shown as follows Figure 4 As shown in the figure, it consists of three layers: input layer, hidden layer and output layer.

[0047] S206, determining a target processing instruction based on a rule mapping relationship and a target identification code, wherein the rule mapping relationship includes a plurality of exception identification codes, a plurality of processing instructions, and a mapping relationship between the processing instructions and the exception identification codes, and a target mapping relationship between the target identification code and the target processing instruction;

[0048] S208, repairing the image acquisition device based on the target processing instruction.

[0049] Optionally, in an embodiment of the present application, the above-mentioned target processing instruction is for execution by the above-mentioned image acquisition device. For example, assuming that the target identification code is "0x10040005", indicating that the image acquisition device needs to be restarted, the corresponding target processing instruction is "equipment restart", which enables the image acquisition device to automatically restart the device.

[0050] Exemplarily, a rule engine is used to automatically determine whether the target identification code meets the target screening conditions. If the target identification code is queried in the rule mapping relationship in the rule engine container, the processing instruction corresponding to the target identification code can be used as the above-mentioned target processing instruction. At this time, the target identification code meets the target screening conditions.

[0051] Specifically, the data processing method proposed in this application can be applied to various equipment self-detection systems. Figure 5 is a schematic diagram of an optional data processing method according to an embodiment of the present application, and its system structure can be as follows Figure 5 As shown, it includes an analysis module, a detection module, a collection module, a repair module, a communication module, and a camera (the above-mentioned image acquisition device). Specifically, Figure 6 is a schematic diagram of an optional data processing method according to an embodiment of the present application, and the specific implementation steps may include but are not limited to the following: Figure 6 As shown:

[0052] S602, collecting error log information of the image acquisition device, etc.;

[0053] S604, inputting the error log information of the image acquisition device into the target recognition model to obtain a plurality of error identification codes, and then inputting the plurality of error identification codes into a multi-layer perceptron to obtain a unique error identification code as the target identification code;

[0054] S606, input the target identification code into the rule engine, obtain the target processing instruction corresponding to the target identification code, and send it to the image acquisition device for the image acquisition device to execute the target processing instruction to complete self-repair.

[0055] Through the embodiments of the present application, an abnormal data associated with an image acquisition device is obtained, wherein the abnormal data represents the corresponding operating data when an abnormality occurs in the image acquisition device; the abnormal data is input into a target recognition model to obtain a target identification code, wherein the target recognition model is used to identify the abnormal data and determine the abnormal type of the abnormal data, and the target identification code is used to identify the abnormal type; a target processing instruction is determined based on a rule mapping relationship and a target identification code, wherein the rule mapping relationship includes multiple abnormal identification codes, multiple processing instructions, and a mapping relationship between processing instructions and abnormal identification codes, and there is a target mapping relationship between the target identification code and the target processing instruction; a method of repairing the image acquisition device based on the target processing instruction is adopted, and the purpose of improving the abnormality repair efficiency is achieved through an automated abnormality detection and processing process, thereby achieving the technical effect of quickly and accurately identifying and processing the abnormality of the image acquisition device, and further solving the technical problem that the abnormality repair process of the image acquisition device is complicated, resulting in low abnormality repair efficiency of the image acquisition device.

[0056] As an optional solution, the above-mentioned abnormal data is input into the target recognition model to obtain the target identification code, including: using the word segmentation processing module to convert the characters in the above-mentioned abnormal data into character vectors; inputting the above-mentioned character vectors into the bidirectional long short-term memory network module to determine the occurrence probability of the above-mentioned characters belonging to each identification code in the identification code set, and determining the identification code whose occurrence probability meets the preset probability condition as the initial identification code; inputting the above-mentioned initial identification code into the sequence labeling module to determine the above-mentioned target identification code, wherein the above-mentioned target recognition model includes the above-mentioned word segmentation processing module, the above-mentioned bidirectional long short-term memory network module, and the above-mentioned sequence labeling module, and the above-mentioned sequence labeling module is used to adjust the position of each named entity in the above-mentioned initial identification code.

[0057] For example, Figure 5As shown, the above target recognition model can be deployed in Figure 5 The analysis module in processes the abnormal data from the acquisition module.

[0058] Specifically, the bidirectional long short-term memory network module Bi-LSTM is a deep learning model that can capture the forward and backward dependencies in a sequence. In the bidirectional long short-term memory network module, the input character vector will be processed by Bi-LSTM to obtain the occurrence probability of each character, that is, the possibility that the character is the codeword in the first identification code.

[0059] Optionally, in an embodiment of the present application, the above-mentioned preset probability condition is used to determine which characters are considered to be target characters, and only when the probability of occurrence of a character exceeds a preset probability threshold, the character is considered to be part of the first identification code, and the above-mentioned sequence labeling module is used to include a direction matrix, which is used to label each element in the sequence (the above-mentioned characters) with a corresponding label (the above-mentioned characters belong to the identification code in the identification code set).

[0060] In an exemplary embodiment, assuming that the abnormal data is a log containing error information, first, the word segmentation processing module converts each character in the text into a character vector, which can be based on character embedding to represent the position of each character in a certain vector space. Next, the character vector is input into a bidirectional long short-term memory network (BiLSTM) module to output the probability of each character belonging to each identification code in a specific identification code set. Furthermore, according to a preset probability condition (such as a probability greater than 0.5), it is possible to determine from the output of the BiLSTM which characters are part of the initial identification code. Finally, the initial identification code is input into a sequence labeling module to further analyze the character sequence and adjust and confirm the boundary of the initial identification code.

[0061] As an optional scheme, the above-mentioned determination of the target identification code based on the above-mentioned first identification code and the above-mentioned second identification code includes: inputting the above-mentioned initial identification code into the above-mentioned sequence labeling module to determine multiple first identification codes; inputting the above-mentioned multiple first identification codes into a multi-layer perceptron to determine the weight value corresponding to each of the above-mentioned first identification codes; performing a weighted operation on each of the above-mentioned first identification codes according to the above-mentioned weight value to obtain the above-mentioned target identification code.

[0062] As an optional scheme, after inputting the above-mentioned initial identification code into the above-mentioned sequence labeling module and determining multiple first identification codes, the above-mentioned method also includes: obtaining a second identification code sent by the target playback platform, wherein the above-mentioned target playback platform is used to receive the image captured by the above-mentioned image acquisition device, and the above-mentioned second identification code is used to indicate that the communication between the above-mentioned target playback platform and the above-mentioned image acquisition device is abnormal; inputting the above-mentioned multiple first identification codes and the above-mentioned second identification code into a multi-layer perceptron to determine the weight value corresponding to each of the above-mentioned first identification code and the above-mentioned second identification code; performing a weighted sum operation on each of the above-mentioned first identification code and the above-mentioned second identification code according to the above-mentioned weight value to obtain the above-mentioned target identification code.

[0063] Optionally, in an embodiment of the present application, the first identification code and the second identification code can be composed of hexadecimal, binary or other character strings, and the target playback platform can be a device such as a television, computer, mobile phone, tablet, or a dedicated image display or processing device.

[0064] Exemplarily, by obtaining the first identification code and the second identification code, the communication status between the image acquisition device and the target playback platform can be detected. If the first identification code indicates that the device is abnormal, the image acquisition device may need to be inspected and repaired; if the second identification code indicates that the communication is abnormal, it may be necessary to check the network connection or communication protocol, etc.

[0065] Through the embodiments of the present application, the communication status between the image acquisition device and the target playback platform can be accurately and conveniently obtained, abnormalities can be discovered and resolved in a timely manner, and the operating stability and reliability of the image acquisition device can be improved.

[0066] In an exemplary embodiment, the initial identification code is input into the sequence labeling module, and a plurality of first identification codes are determined, wherein the initial identification codes are: "123456789" and "987654321". Then, the plurality of first identification codes are input into the multilayer perceptron to determine the weight value corresponding to each first identification code, the first identification code is: "123456789" and "987654321". Assume that the multilayer perceptron assigns a weight value to each identification code, for example: "123456789" and "987654321". The weight value of "56789" is 0.8, and the weight value of "987654321" is 0.3. A weighted sum operation is performed on each first identification code to obtain: "123456789"×0.8=987654.32; "987654321"×0.3=296296.96. According to the result of the weighted operation, the target identification code is determined to be 1283950.605. An integer can be taken as the target identification code 1283950, etc. This application does not impose any limitation on it.

[0067] In another exemplary embodiment, after the target playback platform detects a communication anomaly, it generates a second identification code as shown below:

[0068] The second identification code: "NET_ERR_001". This identification code "NET_ERR_001" is an error code defined by the target playback platform and is used to indicate the specific type of communication anomaly. "NET_ERR_001" means "network connection error".

[0069] For example, assuming that the first identification code is A and the second identification code is 1, the initial identification codes are A and 1. Then, the initial identification codes "A" and "1" are converted into hexadecimal and input into the pre-trained multi-layer perceptron. This multi-layer perceptron has learned how to determine the mapping weight value of the input identification code through a large amount of training data, for example, the weight value of A is 0.7, the weight value of 1 is 0.3, and the initial identification code is weighted according to the weight value corresponding to each initial identification code.

[0070] For example, the two identification codes are regarded as a combination, and then the weighted value is calculated to obtain the target identification code represented as: weight value of A*weight value of A+1*1=0.7*A+0.3*1.

[0071] For example, assume that there is a monitoring system including an image acquisition device and a target playback platform. The image acquisition device is responsible for acquiring images and sending them to the target playback platform. If an abnormality occurs during the communication process, the target playback platform will send a second identification code to indicate the abnormality. The image acquisition device attempts to send image data to the target playback platform, but due to network problems, the data fails to be successfully transmitted. The target playback platform detects the communication abnormality and generates a second identification code as shown below:

[0072] The second identification code: “COMM_ERR_001”, this identification code “COMM_ERR_001” is used to indicate a communication abnormality.

[0073] Meanwhile, the image acquisition device may have generated multiple first identification codes for identifying different image data batches or states, for example:

[0074] First identification code: "IMG_0001", "IMG_0002", "IMG_0003";

[0075] Next, the first identification code and the second identification code are input into the multi-layer perceptron to determine the weight value corresponding to each identification code. Assume that the weight value of "IMG_0001" is 0.7; the weight value of "IMG_0002" is 0.5; the weight value of "IMG_0003" is 0.6; the weight value of "COMM_ERR_001" is 0.9. Finally, according to the weight value determined by the multi-layer perceptron, a weighted sum operation is performed on each first identification code and the second identification code, and the target identification code is "0x2".

[0076] Through the embodiments of the present application, weights can be assigned to different identification codes through a multi-layer perceptron, and the most important identification code can be determined as the target identification code through weighted summation to ensure the accuracy of the target identification code.

[0077] As an optional solution, the above-mentioned determination of the target processing instruction based on the rule mapping relationship and the above-mentioned target identification code includes: obtaining a rule mapping file, wherein the above-mentioned rule mapping file is used to store the above-mentioned rule mapping relationship; loading the above-mentioned rule mapping file into a rule engine container, wherein the above-mentioned rule engine container is used to perform rule matching according to the above-mentioned rule mapping relationship; inputting the above-mentioned target identification code into the above-mentioned rule engine container, and determining the above-mentioned target processing instruction corresponding to the above-mentioned target identification code from the multiple above-mentioned processing instructions based on the above-mentioned target mapping relationship.

[0078] Optionally, in an embodiment of the present application, the above-mentioned rule engine container is a software component that can parse and store the rule mapping relationship in the rule mapping file and make decisions based on the rule mapping relationship.

[0079] Specifically, the rule engine container can automatically query the above-mentioned target identification code from the rule mapping relationship, and obtain a processing instruction corresponding to the target identification code, and determine the processing instruction as the above-mentioned target processing instruction, and the above-mentioned rule mapping file can be stored in a database, XML file, JSON file or other external storage medium.

[0080] For example, the target identification code is represented as "AAAA", and there is a rule in the rule mapping relationship that shows "the identification code is AAAA, indicating that the device has been running for too long and needs to be restarted. At this time, the target processing instruction obtained is a restart device instruction.

[0081] As an optional solution, the above-mentioned input of the above-mentioned abnormal data into a pre-trained target recognition model to obtain a target identification code includes: obtaining sample abnormal data, wherein the above-mentioned sample abnormal data includes sample abnormal log information and sample abnormal debugging information of the above-mentioned image acquisition device, and the above-mentioned sample abnormal data is pre-labeled with a sample identification code label through a natural language processing module; inputting the above-mentioned abnormal data into an initial recognition model, and using a word segmentation processing module to convert each sample character in the above-mentioned sample abnormal data into a corresponding sample character vector; inputting the above-mentioned sample character vector into a bidirectional long short-term memory network module to determine the predicted occurrence probability of each of the above-mentioned sample characters ... character vector is input into a bidirectional long short-term memory network module to determine the predicted occurrence probability of each of the above-mentioned sample characters; wherein the above-mentioned sample character vector is input into a bidirectional long short-term memory network module to determine the predicted occurrence probability of each of the above-mentioned sample characters; wherein the above-mentioned sample character vector is input into a bidirectional long short-term memory network module to determine the predicted occurrence probability of each of the above-mentioned sample characters; wherein the above-mentioned sample character vector is input into a bidirectional long short-term memory network module to determine the predicted occurrence probability of each of the above-mentioned sample characters; wherein the above-mentioned sample character vector is input into a bidirectional long short-term memory network module to determine the predicted occurrence probability of each of the above-mentioned sample characters; wherein the above-mentioned sample character vector is input into a bidirectional long short-term memory network module to determine the predicted The sample characters whose predicted occurrence probability meets the preset probability conditions belong to the sample prediction identification codes; the probability transfer operation is performed on the predicted characters according to the direction matrix to determine the sample prediction identification codes, wherein the initial recognition model includes the word segmentation processing module, the bidirectional long short-term memory network module, and the direction matrix, and the direction matrix is ​​used to adjust the positions of the sample characters in the sample prediction identification codes; the loss function of the initial recognition model is adjusted based on the sample prediction identification codes and the sample identification code labels, and when the model parameters of the initial recognition model meet the preset parameter conditions, the initial recognition model is determined as the target recognition model.

[0082] In an exemplary embodiment, first, it is necessary to collect sample abnormality log information and sample abnormality debugging information of the image acquisition device, then use natural language processing technology to preprocess the collected sample abnormality data, including word segmentation, part-of-speech tagging, etc., and assign a sample identification code label to each sample data, and then input the preprocessed sample abnormality data into an initial recognition model. The word segmentation processing module in the model converts each character in the sample data into a corresponding character vector, and uses a bidirectional long short-term memory network to determine the predicted probability of each character. Bi-LSTM can process sequence data and can capture long-term dependencies in the data.

[0083] Furthermore, when the predicted probability of occurrence meets the preset conditions, the character is considered to be part of the sample prediction identification code. Then, the direction matrix is ​​used to perform a probability transfer operation on the predicted character to determine the final sample prediction identification code to ensure the accuracy of the prediction identification code. The sample prediction identification code and the sample identification code label are used to adjust the loss function of the initial recognition model to ensure that the training level of the initial recognition model meets the business requirements. At this time, the initial recognition model can be determined as the target recognition model and put into use in real business scenarios.

[0084] As an optional scheme, the above-mentioned acquisition of abnormal data associated with the above-mentioned image acquisition device includes: when the communication connection between the above-mentioned image acquisition device and the target playback platform is disconnected, and the communication connection between the above-mentioned image acquisition device and the target detection platform is not disconnected, obtaining a first trigger request issued by the above-mentioned image acquisition device, wherein the above-mentioned first trigger request indicates that the above-mentioned image acquisition device requests to stop sending images to the above-mentioned target playback platform, and send the above-mentioned abnormal data to the target detection platform; in response to the above-mentioned first trigger request, obtaining the above-mentioned abnormal data.

[0085] Optionally, in the embodiment of the present application, the target detection platform may include but is not limited to: Figure 5 As shown, it includes an acquisition module, an analysis module, etc. The above-mentioned first trigger request indicates that the image acquisition device needs to change its behavior mode, execute preset target processing instructions, and solve abnormal problems.

[0086] In an exemplary embodiment, assuming that the image acquisition device is a camera, the data processing method proposed in this application can be applied to scenarios where users are not aware of the situation, and to handle a large number of camera outages:

[0087] S1, the camera is down and the communication module of the playback cloud platform is disconnected, the camera protection program is started and the communication module of the detection and repair system is connected (the above-mentioned first trigger request), and the signal and camera information are automatically collected to the collection module of the detection and repair system;

[0088] S2, the acquisition module receives an abnormal signal, triggers the acquisition operation, and receives the log file and tcpdump file reported by the camera;

[0089] S3, the acquisition module sends the log file and the tcpdump file to the analysis module, the analysis module analyzes and extracts the single error code, and after the analysis is completed, sends the detection signal and the single error code to the detection module;

[0090] S4, after receiving the detection signal, the detection module processes the single error code (the target identification code mentioned above), and after analysis, finds that an upgrade is required, and provides the upgrade package download path and upgrade instructions to the repair module;

[0091] S5, the repair module sends the upgrade package and upgrade instruction (the above-mentioned target processing instruction) to the camera via the communication module, and the camera completes the upgrade.

[0092] S6, the camera's log recording program crashes and automatically repairs the log.

[0093] Exemplarily, the image acquisition device should stop sending images to the target playback platform, which can be achieved by including but not limited to stopping image encoding, closing the network transmission channel, etc. The above-mentioned abnormal data may include but is not limited to device status, error log, environmental parameters, etc.

[0094] As an optional solution, the above-mentioned acquisition of abnormal data associated with the above-mentioned image acquisition device includes: in response to a target interaction operation, sending a second trigger request to the above-mentioned image acquisition device, wherein the above-mentioned second trigger request is used to request the above-mentioned image acquisition device to stop sending images to the target playback platform, and send the above-mentioned abnormal data to the target detection platform; in response to receiving a target confirmation message sent by the above-mentioned image acquisition device, the above-mentioned abnormal data is acquired.

[0095] It should be noted that the user can also spontaneously request on the target detection platform to detect whether the image acquisition device has any abnormality, so that the image acquisition device enters a self-repair process.

[0096] In an exemplary embodiment, the target confirmation message indicates that the image acquisition device has successfully stopped sending images to the target playback platform and started sending abnormal data to the target detection platform.

[0097] Exemplarily, the data processing method proposed in this application can be applied to application scenarios where users initiate device detection on their own:

[0098] S1, the user finds that a large number of camera videos display abnormalities, clicks the repair button in the detection and repair system (target interactive operation), and then sends a repair request to the camera (second trigger request);

[0099] S2, the acquisition module receives the repair signal, triggers the acquisition operation, and obtains the log file and tcpdump file uploaded by the camera;

[0100] S3, the acquisition module sends the analysis signal and the file to be analyzed to the analysis module, the analysis module performs analysis, extracts the single error code, and sends the detection signal and the single error code to the detection module after the analysis is completed;

[0101] S4, after receiving the detection signal, the detection module processes the single error code and analyzes that a certain platform frequently triggers the camera to restart and the video abnormality is also caused by the frequent restart;

[0102] S5, the repair module sends a message to the camera via the communication module to disconnect the link of a certain platform, and informs the user of the abnormal reason through a text message;

[0103] S6, the camera records the repair log.

[0104] Through the embodiments of the present application, the entire process from detecting anomalies, terminating normal operations, analyzing data, determining instructions to repairing equipment is ensured to be automated, so that the image acquisition device can respond quickly and resume normal operation when an abnormality occurs.

[0105] In an exemplary embodiment, in the case of a large number of device logs, packet captures, and various information, through the embodiments of the present application, effective feature data can be extracted quickly and efficiently, and automated detection and maintenance can be performed. The system realizes an integrated logical closed loop, which saves labor costs while improving the on-site problem-solving efficiency of image acquisition equipment.

[0106] Specifically, first collect information from the faulty image acquisition device, including error logs of all submodules, windbg, Tcpdump files, etc. Then put this valid information into the "self-detection" module for automatic error cause analysis. Once the error cause analysis is completed, the corresponding generated solution is fed back to the front-end device for "self-repair" operation, and the faulty device can be repaired without human intervention, which may include but is not limited to:

[0107] S1, error collection module:

[0108] The error information of camera devices can include three types of files. The first is packet capture information (tcpdump), which mainly involves all network-related information, such as network latency, network rate, and http return values ​​related to platform interaction. The second is error logs, which are generally more extensive log information and prints recorded inside all programs. There are more obvious and specific errors and may carry custom fixed error codes, such as authorization permission errors, login errors, etc. The third is the dump file generated by the internal program crash. This type of error information is more difficult to handle and needs to be combined with specific program codes.

[0109] S2, self-test module:

[0110] The self-detection module includes the early deep learning training, and the error types and solutions output after the error information is imported.

[0111] S2-1, NER extracts key information: Deep learning based on BERT+LSTM+CRF named entity recognition: First, data preprocessing is performed. NER is a sequence labeling task that requires boundary determination. We need to label a string of consecutive words with the same label and convert the joint label into the original label through BIOS label conversion. For example, the platform-related return information we obtain from packet capture contains hexadecimal information A4:1A:3A:39:78:4D. First, we use the list splitting function to decompose it into individual characters ['A','4',':','1','A',':','3','A',':','3','9',':','7','8',':','4','D'], and then use the marking function to process it into the following format: ['B-name','I-name','O','B-name','I-name','O','B-name','I-name','O','B-name','I-name','O','B-name','I-name']. If it is information in the log (such as "login error"), it is also marked.

[0112] S2-2, build a data set, add a CLS in front of each set of fixed input data, all input characters will be converted into an id, and store the index of the starting position of the word with the label. Before starting model training, use id2label and label2id to create a mapping set of the desired label to its id. (The specific process is not explained, see the labe2id function in python for details); after construction, split the data set, we will collect the packet capture information and log information of the feature marks in all logs in a ratio of 9:1, and split the training data into a training set and a validation set.

[0113] S2-3, model construction, for example, assuming the classification of error code types and HTTP protocol fields, there are more other types in actual situations:

[0114] Embedding layer: Use Bert to convert each character into a character vector;

[0115] BiLSTM layer: uses bidirectional LSTM;

[0116] Encode layer: obtains the occurrence probability and maps the high-dimensional vector into a dimension that matches the label;

[0117] CRF layer: It is essentially a direction matrix, which obtains the transition probability, represents the state transition information between labels, and uses the CEF feature function set;

[0118] S2-4, classification model classification error type: multi-layer perceptron, the result of the classification model training after the key information extracted by NER can classify a large amount of data into useless data, various types of error codes, and various types of protocols. The labeled data has greatly reduced the amount of original data. Next, use a multi-layer perceptron to classify these error types. The perceptron is usually composed of a three-layer model: input layer, hidden layer, output layer:

[0119] The hidden layer outputs f(W1X+b1), where W is the weight, b is the bias, and the activation function f formula is: Sigmoid(x)=1 / (1+e -x ); Output layer multi-class logistic regression formula: I k =∑w kj O y +θ k (k∈[j-1,n]); through continuous iteration, find the most suitable weights and biases to minimize the loss.

[0120] During training, there may be multiple errors in the input of the same device. Classification generally balances the classification results by setting the weight of each error.

[0121] Balance degree δ=W1f+W2μ+W3c…+WnΦ, where w1+w2+..Wn=1;

[0122] Finally, the corresponding output result obtained by the model should be a specific and single error code. For example, when the platform login fails, the device fails and the cloud service cannot be used. Through data NER sorting, the HTTPS hexadecimal error code of the platform login error is found in a large amount of data, and then this error code (the above target identification code) is input into the perceptron, and finally classified as 0x10040004: the current user permission check fails;

[0123] S3, rule engine recommended solution: DROOLS, because there are many reasons for camera errors, covering various factors such as network, hardware, software, environment, etc., and as the functionality becomes increasingly complex, if IF-ELSE is used to handle the final error results and solution development efficiency is too low, the maintenance cost is very high. In order to improve efficiency, automation should also be achieved in the management process, so the rule engine DROOLS is introduced. Figure 7 is a schematic diagram of another optional data processing method according to an embodiment of the present application, such as Figure 7 As shown:

[0124] S3-1, create a fact object:

[0125] Based on OO thinking, the rule factors required by the routing strategy are generated, the Fact*JavaBean objects are constructed and then handed over to the rule engine for execution. The implementation in the code is new router() and set the rule parameters.

[0126] S3-2, generate rule-drl data file:

[0127] Quickly generate multiple rules through fact code objects, usually in the pseudo-code -drl file format. Exporting this rule file can be used in multiple projects that use the rule engine. (It can be maintained in the database later without changing the code).

[0128] S3-3, rule file preloading:

[0129] The drl file needs to be loaded into the drools working memory first, that is, loaded into the drools container. After verifying the legitimacy of the drl rule file, a rule engine container is finally built (the drools management center API provides CRUD, construction, management and execution interfaces).

[0130] S3-4, fact object hit rule:

[0131] Provide data to the rule engine, which will match rules based on the provided data. After activating the rule engine, if the rule matches successfully, the rule will be executed.

[0132] S4, self-repair module:

[0133] Usually, the solutions to errors are restarting, updating, reauthorizing, initializing the device, prompting the user to operate manually, etc. Therefore, after obtaining the specific solution to the error in the self-detection module, the final repair step can be activated by presetting specific instructions in the solution. Through this full-dimensional preset solution, manual operations can be reduced to the greatest extent. Except for physical repairs that require prompting the user to manually operate on site, all other repairs can be automatically repaired through the analysis of specific errors.

[0134] Exemplarily, the embodiments of the present application include the following modules: a communication module, an acquisition module, an analysis module, a detection module, and a repair module: Communication module: used for communication between the camera and the detection system, such as serial communication and network communication; Acquisition module: used for collecting camera operation logs, packet capture logs, and dump logs; Analysis module: used to analyze the information collected by the acquisition module and extract fault characteristic values; Detection module: used for fault characteristic value detection and providing a repair plan; Repair module: executes the repair plan to repair the camera.

[0135] In an exemplary embodiment, in a scenario where a large number of cameras are down and users are not aware of the downtime, the following steps are performed:

[0136] S1, the camera is down and the communication module is disconnected, the camera protection program is started and the communication module is connected, and the signal and camera information are automatically collected to the collection module;

[0137] S2, the acquisition module receives the abnormal signal, triggers the acquisition operation, and receives the log file and dump file reported by the camera;

[0138] S3, the acquisition module sends the analysis signal and the file to be analyzed to the analysis module, the analysis module performs analysis and extracts the characteristic value, and after the analysis is completed, sends the detection signal and the characteristic value to the self-detection module;

[0139] S4, after receiving the detection signal, the self-detection module processes the characteristic value, and after analysis finds that an upgrade is needed, provides the upgrade package download path and upgrade instructions to the repair module;

[0140] S5, the repair module sends the upgrade package and upgrade instructions to the camera via the communication module, and the camera completes the upgrade;

[0141] S6, the camera's log recording program crashes and automatically repairs the log.

[0142] In an exemplary embodiment, in a scenario where a large number of cameras are downtime processed when the user actively triggers the downtime:

[0143] S1, the user finds that a large number of camera videos display abnormalities, and issues a repair instruction in the detection and repair system;

[0144] S2, the collection module receives the repair signal, triggers the collection operation, collects the device operation log, and captures the packet;

[0145] S3, the acquisition module sends the analysis signal and the file to be analyzed to the analysis module, the analysis module performs analysis and extracts the characteristic value, and after the analysis is completed, sends the detection signal and the characteristic value to the self-detection module;

[0146] S4, after receiving the detection signal, the detection module processes the characteristic value and analyzes that a certain platform frequently triggers the camera to restart and the video abnormality is also caused by the frequent restart;

[0147] S5, the repair module sends a message to the camera via the communication module to disconnect the link of a certain platform, and informs the user of the abnormal reason through a text message;

[0148] S6, the camera records the repair log.

[0149] Through the embodiments of the present application, in the case of a large number of equipment logs, packet captures, and various information, effective data can be quickly extracted and automatic detection and maintenance integration can be achieved, thereby improving the detection and maintenance efficiency while reducing manual workload.

[0150] It is understandable that in the specific implementation of this application, related data such as user information is involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0151] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0152] According to another aspect of the embodiment of the present application, a data processing device for implementing the above-mentioned data processing method is also provided. Figure 8 As shown, the device comprises:

[0153] An acquisition module 802 is used to acquire abnormal data associated with the image acquisition device, wherein the abnormal data represents the corresponding operation data when the image acquisition device is abnormal;

[0154] A first determination module 804 is used to input the abnormal data into a target recognition model to obtain a target identification code, wherein the target recognition model is used to identify the abnormal data and determine the abnormal type of the abnormal data, and the target identification code is used to identify the abnormal type;

[0155] A second determination module 806 is used to determine a target processing instruction based on a rule mapping relationship and a target identification code, wherein the rule mapping relationship includes a plurality of exception identification codes, a plurality of processing instructions, and a mapping relationship between the processing instructions and the exception identification codes, and there is a target mapping relationship between the target identification code and the target processing instruction;

[0156] The repair module 808 is used to repair the image acquisition device based on the target processing instruction.

[0157] As an optional solution, the above-mentioned device is used to input the above-mentioned abnormal data into a target recognition model to obtain a target identification code in the following manner: using a word segmentation processing module to convert characters in the above-mentioned abnormal data into character vectors; inputting the above-mentioned character vectors into a bidirectional long short-term memory network module to determine the occurrence probability of the above-mentioned characters belonging to each identification code in the identification code set, and determining the identification code whose occurrence probability meets a preset probability condition as the initial identification code; inputting the above-mentioned initial identification code into a sequence labeling module to determine the above-mentioned target identification code, wherein the above-mentioned target recognition model includes the above-mentioned word segmentation processing module, the above-mentioned bidirectional long short-term memory network module, and the above-mentioned sequence labeling module, and the above-mentioned sequence labeling module is used to adjust the positions of each named entity in the above-mentioned initial identification code.

[0158] As an optional solution, the above-mentioned device is used to input the above-mentioned initial identification code into the sequence labeling module to determine the above-mentioned target identification code in the following manner: input the above-mentioned initial identification code into the above-mentioned sequence labeling module to determine multiple first identification codes; input the above-mentioned multiple first identification codes into a multi-layer perceptron to determine the weight value corresponding to each of the above-mentioned first identification codes; perform a weighted operation on each of the above-mentioned first identification codes according to the above-mentioned weight value to obtain the above-mentioned target identification code.

[0159] As an optional solution, the above-mentioned device is also used to: input the above-mentioned initial identification code into the above-mentioned sequence labeling module, and after determining multiple first identification codes, obtain the second identification code sent by the target playback platform, wherein the above-mentioned target playback platform is used to receive the image captured by the above-mentioned image acquisition device, and the above-mentioned second identification code is used to indicate that the communication between the above-mentioned target playback platform and the above-mentioned image acquisition device is abnormal; input the above-mentioned multiple first identification codes and the above-mentioned second identification code into a multi-layer perceptron to determine the weight value corresponding to each of the above-mentioned first identification code and the above-mentioned second identification code; perform a weighted sum operation on each of the above-mentioned first identification code and the above-mentioned second identification code according to the above-mentioned weight value to obtain the above-mentioned target identification code.

[0160] As an optional solution, the above-mentioned device is used to determine the target processing instruction based on the rule mapping relationship and the above-mentioned target identification code in the following manner: obtain the rule mapping file, wherein the above-mentioned rule mapping file is used to store the above-mentioned rule mapping relationship; load the above-mentioned rule mapping file into the rule engine container, wherein the above-mentioned rule engine container is used to perform rule matching according to the above-mentioned rule mapping relationship; input the above-mentioned target identification code into the above-mentioned rule engine container, and determine the above-mentioned target processing instruction corresponding to the above-mentioned target identification code from the multiple above-mentioned processing instructions based on the above-mentioned target mapping relationship.

[0161] As an optional scheme, the above-mentioned device is used to obtain the abnormal data associated with the above-mentioned image acquisition device in the following manner: when the communication connection between the above-mentioned image acquisition device and the target playback platform is disconnected, and the communication connection between the above-mentioned image acquisition device and the target detection platform is not disconnected, obtain a first trigger request issued by the above-mentioned image acquisition device, wherein the above-mentioned first trigger request indicates that the above-mentioned image acquisition device requests to stop sending the image to the above-mentioned target playback platform, and send the above-mentioned abnormal data to the target detection platform; in response to the above-mentioned first trigger request, obtain the above-mentioned abnormal data.

[0162] As an optional solution, the above-mentioned device is used to obtain the abnormal data associated with the above-mentioned image acquisition device in the following manner: in response to the target interaction operation, a second trigger request is sent to the above-mentioned image acquisition device, wherein the above-mentioned second trigger request is used to request the above-mentioned image acquisition device to stop sending images to the target playback platform, and send the above-mentioned abnormal data to the target detection platform; in response to receiving the target confirmation message sent by the above-mentioned image acquisition device, the above-mentioned abnormal data is obtained.

[0163] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0164] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0165] According to one aspect of the present application, a computer program product is provided. The computer program product includes a computer program.

[0166] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0167] Fig. 9 The structure block diagram of a computer system for implementing an electronic device according to an embodiment of the present application is schematically shown.

[0168] It should be noted that Fig. 9 The computer system 900 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0169] like Fig. 9As shown, the computer system 900 includes a central processing unit 901 (CPU), which can perform various appropriate actions and processes according to the program stored in the read-only memory 902 (ROM) or the program loaded from the storage part 908 to the random access memory 903 (RAM). Various programs and data required for system operation are also stored in the random access memory 903. The central processing unit 901, the read-only memory 902 and the random access memory 903 are connected to each other through a bus 904. An input / output interface 905 (Input / Output interface, i.e., I / O interface) is also connected to the bus 904.

[0170] The following components are connected to the input / output interface 905: an input section 906 including a keyboard, a mouse, etc.; an output section 907 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, a modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the input / output interface 905 as needed. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 910 as needed so that a computer program read therefrom is installed into the storage section 908 as needed.

[0171] In particular, according to an embodiment of the present application, the process described in each method flow chart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer readable medium, and the computer program contains a program code for executing the method shown in the flow chart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 909, and / or installed from the removable medium 911. When the computer program is executed by the central processor 901, various functions defined in the system of the present application are executed.

[0172] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 909, and / or installed from the removable medium 911. When the computer program is executed by the central processor 901, various functions provided by the embodiments of the present application are performed.

[0173] According to another aspect of the embodiment of the present application, an electronic device for implementing the above data processing method is also provided. The electronic device may be Figure 1 The terminal device or server shown in the figure. This embodiment is described by taking the electronic device as a terminal device as an example. Fig.10 As shown, the electronic device includes a memory 1002 and a processor 1004. The memory 1002 stores a computer program, and the processor 1004 is configured to execute the steps in any of the above method embodiments through the computer program.

[0174] Optionally, in this embodiment, the electronic device may be located in at least one network device among a plurality of network devices of a computer network.

[0175] Optionally, in this embodiment, the above-mentioned processor can be configured to execute the methods in each embodiment of the present application through a computer program.

[0176] Alternatively, a person skilled in the art may understand that: Fig.10 The structure shown is for illustration only. Fig.10 The structure of the electronic device is not limited. Fig.10 More or fewer components (such as network interfaces, etc.) as shown in, or with Fig.10 Different configurations are shown.

[0177] Among them, the memory 1002 can be used to store software programs and modules, such as the program instructions / modules corresponding to the data processing method and device in the embodiments of the present application. The processor 1004 executes various functional applications and data processing by running the software programs and modules stored in the memory 1002, that is, realizing the above-mentioned data processing method. The memory 1002 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 1002 may further include a memory remotely located relative to the processor 1004, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. Among them, the memory 1002 can be specifically, but not limited to, used to store information such as abnormal data. As an example, such as Fig.10 As shown, the above-mentioned memory 1002 may include but is not limited to the acquisition module 802, the first determination module 804, the second determination module 806 and the repair module 809 in the processing device of the above-mentioned data. In addition, it may also include but is not limited to other module units in the processing device of the above-mentioned data, which will not be repeated in this example.

[0178] Optionally, the transmission device 1006 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wired network and a wireless network. In one example, the transmission device 1006 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices and routers via a network cable so as to communicate with the Internet or a local area network. In one example, the transmission device 1006 is a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0179] In addition, the electronic device further comprises: a display 1009 for displaying the image acquired by the image acquisition device; and a connection bus 1010 for connecting the various module components in the electronic device.

[0180] In other embodiments, the terminal device or server may be a node in a distributed system, wherein the distributed system may be a blockchain system, and the blockchain system may be a distributed system formed by connecting the multiple nodes through network communication. The nodes may form a peer-to-peer network, and any form of computing device, such as a server, terminal or other electronic device, may become a node in the blockchain system by joining the peer-to-peer network.

[0181] According to one aspect of the present application, a computer-readable storage medium is provided, and a processor of an electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the electronic device executes the data processing method provided in various optional implementations of the above-mentioned data processing aspects.

[0182] Optionally, in this embodiment, the above-mentioned computer-readable storage medium can be configured to store data for executing the methods in various embodiments of the present application.

[0183] Optionally, in this embodiment, a person of ordinary skill in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing hardware related to the terminal device through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0184] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0185] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for one or more electronic devices to execute all or part of the steps of the methods described in each embodiment of the present application.

[0186] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0187] In the several embodiments provided in the present application, it should be understood that the disclosed application can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0188] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0189] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0190] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A data processing method, characterized in that: include: Acquire abnormal data associated with the image acquisition device, wherein the abnormal data represents corresponding operation data when the image acquisition device is abnormal; Inputting the abnormal data into a target recognition model to obtain a target identification code, wherein the target recognition model is used to identify the abnormal data and determine the abnormal type of the abnormal data, and the target identification code is used to identify the abnormal type; Determine a target processing instruction based on a rule mapping relationship and the target identification code, wherein the rule mapping relationship includes a plurality of exception identification codes, a plurality of processing instructions, and a mapping relationship between the processing instructions and the exception identification codes, and a target mapping relationship exists between the target identification code and the target processing instruction; The image acquisition device is repaired based on the target processing instructions.

2. The method according to claim 1, characterized in that The step of inputting the abnormal data into a target recognition model to obtain a target identification code comprises: Using a word segmentation processing module to convert characters in the abnormal data into character vectors; Inputting the character vector into a bidirectional long short-term memory network module, determining the probability of the character belonging to each identification code in the identification code set, and determining the identification code whose occurrence probability meets a preset probability condition as the initial identification code; The initial identification code is input into a sequence labeling module to determine the target identification code, wherein the target recognition model includes the word segmentation processing module, the bidirectional long short-term memory network module, and the sequence labeling module, and the sequence labeling module is used to adjust the position of each named entity in the initial identification code.

3. The method according to claim 2, characterized in that The step of inputting the initial identification code into a sequence labeling module to determine the target identification code includes: Inputting the initial identification code into the sequence labeling module to determine a plurality of first identification codes; Input the plurality of first identification codes into a multilayer perceptron to determine a weight value corresponding to each of the first identification codes; A weighted operation is performed on each of the first identification codes according to the weight value to obtain the target identification code.

4. The method according to claim 3, characterized in that After inputting the initial identification code into the sequence labeling module and determining a plurality of first identification codes, the method further includes: Acquire a second identification code sent by a target playback platform, wherein the target playback platform is used to receive the image captured by the image acquisition device, and the second identification code is used to indicate that the communication between the target playback platform and the image acquisition device is abnormal; Input the plurality of first identification codes and the second identification code into a multilayer perceptron, and determine a weight value corresponding to each of the first identification code and the second identification code; A weighted sum operation is performed on each of the first identification code and the second identification code according to the weight value to obtain the target identification code.

5. The method according to claim 1, characterized in that: The determining the target processing instruction based on the rule mapping relationship and the target identification code includes: Obtaining a rule mapping file, wherein the rule mapping file is used to store the rule mapping relationship; Loading the rule mapping file into a rule engine container, wherein the rule engine container is used to perform rule matching according to the rule mapping relationship; The target identification code is input into the rule engine container, and the target processing instruction corresponding to the target identification code is determined from the plurality of processing instructions based on the target mapping relationship.

6. The method according to claim 1, characterized in that The obtaining of abnormal data associated with the image acquisition device includes: When the communication connection between the image acquisition device and the target playback platform is disconnected, and the communication connection between the image acquisition device and the target detection platform is not disconnected, obtaining a first trigger request issued by the image acquisition device, wherein the first trigger request indicates that the image acquisition device requests to stop sending images to the target playback platform and to send the abnormal data to the target detection platform; In response to the first trigger request, the abnormal data is acquired.

7. The method according to claim 1, characterized in that The obtaining of abnormal data associated with the image acquisition device includes: In response to the target interactive operation, sending a second trigger request to the image acquisition device, wherein the second trigger request is used to request the image acquisition device to stop sending images to the target playback platform and send the abnormal data to the target detection platform; In response to receiving a target confirmation message sent by the image acquisition device, the abnormal data is acquired.

8. A data processing device, characterized in that: include: An acquisition module, used to acquire abnormal data associated with an image acquisition device, wherein the abnormal data represents corresponding operation data when an abnormality occurs in the image acquisition device; A first determination module is used to input the abnormal data into a target recognition model to obtain a target identification code, wherein the target recognition model is used to identify the abnormal data and determine the abnormal type of the abnormal data, and the target identification code is used to identify the abnormal type; A second determination module is used to determine a target processing instruction based on a rule mapping relationship and the target identification code, wherein the rule mapping relationship includes a plurality of exception identification codes, a plurality of processing instructions, and a mapping relationship between the processing instructions and the exception identification codes, and there is a target mapping relationship between the target identification code and the target processing instruction; A repair module is used to repair the image acquisition device based on the target processing instruction.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein the computer program can be executed by an electronic device to perform the method described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

11. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 7 through the computer program.

Images