Data center equipment image recognition method, system and equipment based on deep learning and medium thereof

Abstract

The invention provides a deep learning-based data center equipment image recognition method and system, equipment and a medium thereof. The method comprises the following steps of: 1, acquiring key image information of the data center equipment and recording equipment related information; the data center equipment image is automatically identified through the deep learning model, compared with traditional manual inspection, the identification efficiency is greatly improved, the labor cost and the time consumption are reduced, meanwhile, the deep learning model can accurately identify information such as the category, the model and the state of the equipment through mass data training, and the identification efficiency is improved. Subjective difference and omission errors caused by manual inspection are avoided; the deep learning model is trained by using a large amount of data, and the performance of the deep learning model is evaluated and optimized regularly, so that the method can adapt to a complex and changeable environment of a data center, is not limited by fixed rules and templates, can be updated and optimized along with the upgrading of equipment and the technical development, and has a wide application prospect. Therefore, management and maintenance requirements are met in time.

Description

Technical Field

[0001] The present invention relates to the field of image recognition technology, and in particular to a data center equipment image recognition method, system, equipment and medium thereof based on deep learning. Background Art

[0002] With the rapid development of information technology, data centers, as the core hub for information storage, processing and transmission, are continuing to expand in size. Today, large data centers house tens of thousands of devices of various types, including servers, storage devices, network devices, refrigeration equipment and other fields. This large-scale and diverse equipment layout poses great challenges to the management and maintenance of equipment, and thus places more stringent requirements on the methods of management and maintenance. However, traditional equipment identification methods have gradually revealed many drawbacks in dealing with this complex situation. 1. Traditional equipment identification methods mainly rely on manual inspections. Manual inspections require operation and maintenance personnel to inspect and identify equipment one by one. This process is extremely time-consuming. Long inspections will make operation and maintenance personnel tired, prone to omissions and errors. Different operation and maintenance personnel may have different understandings and judgment standards for the status and characteristics of equipment, which leads to large subjective differences in the equipment identification process, thereby reducing the accuracy of identification. Second, traditional equipment identification methods are mainly based on rule-based image analysis. The environment of the data center is complex and changeable. Rule-based image analysis methods usually rely on pre-set fixed rules and templates to identify equipment, and have poor adaptability to these complex environmental factors. In addition, with the continuous upgrading of data center equipment and the development of technology, the characteristics and identification requirements of equipment are also constantly changing. Rule-based image analysis methods are difficult to quickly adapt to these changes and cannot meet the requirements of data center equipment management and maintenance for efficient and accurate identification in a timely manner. To this end, a data center equipment image recognition method, system, equipment and medium based on deep learning are proposed. Summary of the invention

[0003] In view of this, an embodiment of the present invention provides a data center equipment image recognition method, system, device and medium based on deep learning to solve or alleviate the technical problems existing in the prior art and at least provide a beneficial option.

[0004] To solve the above technical problems, a technical solution adopted in this application is: a data center equipment image recognition method based on deep learning, comprising the following steps: Step 1: Obtain key image information of data center equipment and record equipment-related information; Step 2: pre-process the acquired key image information based on image processing technology; Step 3: Based on the processed key image information and deep learning algorithm, a deep learning model is constructed; Step 4: Obtain the device image of the device to be identified and perform preprocessing; Step 5: Input the device image of the device to be identified into the deep learning model for identification, and generate a device identification result; Step 6: Store the device identification results in the database and output them in a visual way to generate a device identification report; Step 7: Set a regular evaluation cycle to evaluate the performance of the deep learning model, conduct in-depth analysis of the misidentification cases that occur during the evaluation process, and generate analysis results; Step 8. Update and optimize the deep learning model based on the analysis results.

[0005] Provided as a further preferred embodiment of the present technical solution, in step one, the key image information is collected by a camera and an image acquisition card installed at a key position in the data center; the camera installed at the key position in the data center has autofocus and low-light compensation functions, and the autofocus method adopts a hybrid focus technology that combines phase detection autofocus with contrast detection autofocus; the device-related information includes device number, device location, device brand, device model, equipment production date and the last maintenance time of the equipment.

[0006] As a further preferred embodiment of the present technical solution, in step 2, the key image information preprocessing includes image cleaning, image annotation and image enhancement; the image cleaning uses median filtering and Gaussian filtering to remove salt and pepper noise and Gaussian noise in the image; The calculation formula for removing salt and pepper noise in the image by using median filtering is: ; in, is the original image at coordinates The pixel value at is the pixel value of the filtered image at that point, is the filter window, usually or Rectangular area, median means taking the median of the pixel values ​​in the window; The Gaussian filter is used to remove Gaussian noise in the image, and its calculation formula is: ; in, are pixel coordinates, is the mean (usually 0), is the standard deviation, which determines the width of the Gaussian function. Filtering is achieved by convolving each pixel in the image with the Gaussian function. By calculating the clarity index of the image, the Laplace operator is used to detect the blurred part. The Laplace operator is defined as: ; In the discrete case, its template can be expressed as: or ; The template is convolved with the image. The larger the value obtained, the higher the clarity of the point. For areas with values ​​below a certain threshold, they can be judged as blurred parts and processed accordingly.

[0007] As a further preferred embodiment of the present technical solution, in step three, the method for constructing a deep learning model comprises the following steps: Step 1: Select the convolutional neural network architecture in the deep learning algorithm as the architecture of the deep learning model; Step 2: Divide the preprocessed key image information into a training set, a validation set, and a test set, where the training set accounts for 70%, the validation set accounts for 20%, and the test set accounts for 10%; Step 3: Select the cross entropy loss function and Adam optimizer as the loss function and optimizer required for deep learning model training; Step 4: Use the training set to train the deep learning model, monitor the performance on the validation set, and adjust the model parameters to prevent overfitting; Step 5: After training is completed, use the test set to evaluate the recognition accuracy and recall rate of the deep learning model and generate test results; Step 6: Tune the deep learning model based on the test results, including adjusting the network structure and hyperparameters.

[0008] As a further preferred embodiment of the present technical solution, in step 5, the device identification result includes device category, device model, device status, device identification time and device confidence; The method for generating a device identification result comprises the following steps: Step S1, extracting features of the input device image of the device to be identified based on the convolution layer and pooling layer of the deep learning model to obtain image features; Step S2: input the extracted image features into the fully connected layer of the deep learning model for classification and recognition, and output the preliminary recognition results; Step S3: Calculate the confidence of the preliminary recognition result. If the confidence is lower than a preset threshold, mark it as an uncertain result. Step S4: verify and optimize the uncertain results based on multi-model fusion to obtain the final recognition result.

[0009] As a further preferred embodiment of the present technical solution, the confidence calculation of the device recognition result utilizes the maximum value output by the softmax function as the confidence value, and the preset threshold is dynamically adjusted according to the importance and recognition difficulty of the device. If the confidence of the preliminary recognition result is lower than the corresponding threshold, it is marked as an uncertain result.

[0010] As a further preferred embodiment of the present technical solution, the deep learning model is deployed on a cloud server, using a cloud computing platform to achieve efficient data processing and storage, and providing a remote monitoring interface, allowing remote users to view device recognition results and monitoring videos in real time through a web interface or mobile application.

[0011] To solve the above technical problems, another technical solution adopted by the present application is: a data center equipment image recognition system based on deep learning, comprising: an image acquisition module, an image preprocessing module, a model building module, an image processing module to be identified, an equipment identification module, a result output and storage module, a model evaluation and optimization module and a remote monitoring interface module; The image acquisition module is used to obtain key image information of the data center equipment through the camera and image acquisition card installed at the key position of the data center, and record the relevant information of the equipment, wherein the camera has the functions of autofocus and low light compensation, and the autofocus method adopts the hybrid focus technology combining phase detection autofocus and contrast detection autofocus; The image preprocessing module is used to preprocess the acquired key image information, including image cleaning, image annotation and image enhancement, wherein the image cleaning uses median filtering and Gaussian filtering to remove salt and pepper noise and Gaussian noise in the image; The model building module is used to build a deep learning model based on a deep learning algorithm according to the processed key image information, and the deep learning model uses a convolutional neural network architecture to be trained, verified and tuned through a training set, a verification set and a test set; The to-be-recognized image processing module is used to obtain a device image of the to-be-recognized device and pre-process the device image, and input the device image into a deep learning model for recognition; The device identification module is used to input the device image of the device to be identified into the deep learning model for identification, and generate a device identification result including the device category, device model, device status, device identification time and device confidence; The result output and storage module is used to store the device identification results in a database and output them in a visual manner to generate a device identification report; The model evaluation and optimization module is used to set a regular evaluation cycle, evaluate the performance of the deep learning model, conduct in-depth analysis of the misrecognition cases that occur during the evaluation process, and update and optimize the deep learning model based on the analysis results; The remote monitoring interface module is used to deploy the deep learning model on a cloud server, use the cloud computing platform to achieve efficient data processing and storage, and provide a remote monitoring interface, allowing remote users to view device recognition results and monitoring videos in real time through a web interface or mobile application.

[0012] To solve the above-mentioned technical problems, another technical solution adopted in the present application is: an electronic device, comprising a processor and a memory coupled to the processor, wherein program instructions are stored in the memory, and when the program instructions are executed by the processor, the processor executes the steps of a data center equipment image recognition method based on deep learning as described above.

[0013] In order to solve the above technical problems, another technical solution adopted in this application is: a storage medium storing program instructions capable of implementing a data center equipment image recognition method based on deep learning as described above.

[0014] The embodiment of the present invention has the following advantages due to the adoption of the above technical solution: 1. The present invention uses a deep learning model to automatically identify data center equipment images, which greatly improves the recognition efficiency and reduces labor costs and time consumption compared to traditional manual inspections. At the same time, the deep learning model can accurately identify the type, model, status and other information of the equipment through a large amount of data training, avoiding subjective differences and omissions in manual inspections. 2. The present invention uses a large amount of data to train the deep learning model and regularly evaluates and optimizes the performance of the deep learning model. It can adapt to the complex and changing environment of the data center, is not restricted by fixed rules and templates, and can be updated and optimized as the equipment is updated and the technology develops, so as to meet the management and maintenance needs in a timely manner; 3. The present invention deploys the deep learning model on a cloud server and provides a remote monitoring interface, allowing remote users to view device recognition results and monitoring videos in real time through a web interface or mobile application, thereby improving management efficiency and flexibility.

[0015] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present invention will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A schematic diagram of a process of a data center equipment image recognition method based on deep learning according to the present invention; Figure 2 A schematic diagram of a process for constructing a deep learning model according to the present invention; Figure 3 A schematic diagram of a flow chart of a method for generating device identification results according to the present invention; Figure 4 This is a schematic diagram of the functional modules of a data center equipment image recognition system based on deep learning according to the present invention; Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0019] It should be clear that the following embodiments of the present disclosure are described by specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other in the absence of conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.

[0020] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0021] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure. The drawings only show components related to the present disclosure rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0022] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0023] Figure 1 is a flowchart of a method for image recognition of data center equipment based on deep learning according to an embodiment of the present invention. It should be noted that if there are substantially the same results, the method of the present application is not based on Figure 1 The process sequence shown is limited. Figure 1-Figure 3 As shown: A data center equipment image recognition method based on deep learning, comprising the following steps: Step 1: Obtain key image information of data center equipment and record equipment-related information to provide comprehensive data support for subsequent analysis and management; Step 2: Preprocess the acquired key image information based on image processing technology, thereby improving the image quality and the input data quality of the model; Step 3: Based on the processed key image information and the deep learning algorithm, a deep learning model is constructed, and the device image of the device to be identified is identified through the deep learning model; Step 4: Obtain the device image of the device to be identified and perform preprocessing; wherein the device image of the device to be identified is collected by a camera and an image acquisition card installed at a key position in the data center; the preprocessing of the device image of the device to be identified and the preprocessing of the key image information in the step adopt the same processing method; Step 5: Input the device image of the device to be identified into the deep learning model for identification, and generate the device identification result; specifically: In the data center management system, the integration of equipment identification function can realize the automatic management and intelligent operation and maintenance of equipment. For example, in the equipment asset management system, the equipment ledger is automatically updated according to the equipment identification results to realize the real-time inventory and tracking of equipment assets. When a new device is identified or the equipment information changes, the equipment record in the database is automatically updated, including the location, status, model and other information of the equipment; In the operation and maintenance management system, an operation and maintenance work order can be automatically generated based on the equipment identification results and status. If the model identifies that the equipment is in an abnormal or faulty state, the system automatically generates a corresponding operation and maintenance work order and notifies the relevant operation and maintenance personnel to handle it. The work order contains detailed equipment information, fault description, and recommended handling measures, etc. In the monitoring system, the device identification information and real-time image data are combined to realize visual monitoring and early warning of the device operation status. By displaying the device image and identification results, as well as real-time operating parameters (such as temperature, humidity, etc.) on the monitoring interface, the operation and maintenance personnel can intuitively understand the operation status of the equipment. When certain parameters of the equipment exceed the normal range or identify abnormal conditions, the system automatically sends out early warning signals to remind the operation and maintenance personnel to pay attention and deal with them in time; Step 6: Store the device identification results in the database and output them in a visual way to generate a device identification report. Specifically: store the device identification results, including the device category, model, status, identification time, confidence and other information, in a relational database (such as MySQL) or a non-relational database (such as MongoDB) for subsequent query, statistical analysis and historical tracing. At the same time, establish a data index and backup mechanism to ensure data security and accessibility. Step 7: Set a regular evaluation cycle to evaluate the performance of the deep learning model, conduct in-depth analysis of the misrecognition cases that occurred during the evaluation process, and generate analysis results; among them, conduct in-depth analysis of the misrecognition cases that occurred during the evaluation process, including finding out the causes of the errors, such as image quality problems, special equipment status not covered in the training data, model overfitting, etc., and formulate corresponding improvement measures for different causes; Step 8. Update and optimize the deep learning model based on the analysis results; for example, add new training data, including new types and new states of data center equipment or image data in special environments to enhance the generalization ability of the model; adjust the structure and parameters of the model and optimize the algorithm; retrain and fine-tune to improve the performance and accuracy of the model. At the same time, deploy the optimized model in practical applications in a timely manner to ensure the continued efficient operation of the equipment recognition system.

[0024] In one embodiment, specifically, in step one, key image information is collected by cameras and image acquisition cards installed at key locations in the data center; the cameras installed at key locations in the data center have autofocus and low-light compensation functions, and the autofocus method adopts a hybrid focus technology that combines phase detection autofocus with contrast detection autofocus; the equipment-related information includes equipment number, equipment location, equipment brand, equipment model, equipment production date, and the last maintenance time of the equipment.

[0025] In one embodiment, specifically, in step 2, the key image information preprocessing includes image cleaning, image annotation and image enhancement; image cleaning uses median filtering and Gaussian filtering to remove salt and pepper noise and Gaussian noise in the image; Use median filtering to remove salt and pepper noise in the image. The calculation formula is: ; in, is the original image at coordinates The pixel value at is the pixel value of the filtered image at that point, is the filter window, usually or Rectangular area, median means taking the median of the pixel values ​​in the window; Gaussian filtering is used to remove Gaussian noise in the image. The calculation formula is: ; in, are pixel coordinates, is the mean (usually 0), is the standard deviation, which determines the width of the Gaussian function. Filtering is achieved by convolving each pixel in the image with the Gaussian function. By calculating the clarity index of the image, the Laplace operator is used to detect the blurred part. The Laplace operator is defined as: ; In the discrete case, its template can be expressed as: or ; The template is convolved with the image. The larger the value obtained, the higher the clarity of the point. For areas with values ​​below a certain threshold, they can be judged as blurred parts and processed accordingly. Image annotation: We organize a professional annotation team to accurately annotate the equipment in the image. The annotation content includes the equipment category, model, status (normal, abnormal, fault type, etc.), location coordinates and other information. We use multi-person annotation and cross-validation to ensure the accuracy and consistency of the annotation. For inconsistent annotations, we conduct in-depth discussions and corrections to improve the quality of the annotation data. Image enhancement uses random rotation, scaling, translation, flipping and other operations to transform images, and makes color adjustments to images, such as changes in brightness, contrast, and saturation, to simulate device images under different lighting conditions and environments.

[0026] In one embodiment, specifically, in step three, the method for constructing a deep learning model includes the following steps: Step 1. Select the convolutional neural network architecture in the deep learning algorithm as the architecture of the deep learning model. Among them, the convolutional neural network architecture can be ResNet, VGG or Inception. According to the characteristics and actual needs of the data center device images, preliminarily screen out several possibly suitable architectures. For example, if the device images are rich in details and have many categories, VGG or Inception can be given priority. If you need to process large-scale device image data and have requirements for the depth of the model, ResNet may be a better choice. Step 2: Divide the preprocessed key image information into a training set, a validation set, and a test set, where the training set accounts for 70%, the validation set accounts for 20%, and the test set accounts for 10%. Specifically: Use a random number generator or a specific data set partitioning tool to randomly partition the preprocessed key image information to ensure that each device category has a certain proportion of sample distribution in the training set, validation set, and test set, so as to ensure that the model can be effectively learned and evaluated on different data sets. For example, you can first group the images according to the device category, and then randomly select a certain proportion of images in each category and put them into the training set, validation set, and test set respectively. Step 3. Select the cross entropy loss function and Adam optimizer as the loss function and optimizer required for deep learning model training. For the optimizer, you can choose stochastic gradient descent (SGD) or Adam. The Adam algorithm is a commonly used optimization algorithm that combines the advantages of adaptive learning rate and momentum. It can automatically adjust the learning rate during training, so that the model converges to the optimal solution faster. The update formula of the Adam algorithm is as follows: Compute the gradient: ; in, is the loss function, are model parameters, is the current iteration number; Compute the first moment estimate: ; in, Usually 0.999 is taken; Compute bias-corrected first- and second-order moment estimates: ; Update parameters: ; in, is the learning rate, is a small constant (usually ), used to prevent the denominator from being 0; The calculation formula of the cross entropy loss function is: ; in, is the sample size, is the number of categories, is an indicator variable (if the sample Belongs to category ,but ,otherwise ), is the model prediction sample Belongs to category probability; Step 4: Use the training set to train the deep learning model, monitor the performance on the validation set, and adjust the model parameters to prevent overfitting. Specifically: During the training process, regularly (such as every round or every few rounds) calculate the model's loss and accuracy on the training set and validation set to monitor the model's training process and performance. You can visually observe the model's training status by drawing loss curves and accuracy curves. If you find that the model's loss on the training set continues to decrease, but the loss on the validation set begins to increase or the accuracy no longer increases, it may mean that the model is overfitting. At this time, you can take some measures to alleviate overfitting, such as increasing the proportion of the Dropout layer, reducing the complexity of the model (such as reducing the number of layers or nodes of the convolutional layer or the fully connected layer), and using regularization methods (such as L1 regularization or L2 regularization). At the same time, pay attention to the training speed and memory usage of the model. If the training speed is too slow, you can consider optimizing the code, using more efficient hardware devices, or using distributed training to improve training efficiency. If the memory usage is too high, you can appropriately reduce the batch size or optimize the model structure to reduce memory requirements. Step 5: After training is completed, use the test set to evaluate the recognition accuracy and recall rate of the deep learning model and generate test results. Specifically: During the training process, use the validation set to validate the model regularly. According to the performance indicators on the validation set, adjust the model's hyperparameters and structure. For example, if the accuracy of the model on the validation set is low, try adjusting the learning rate, increasing the strength of data enhancement, or changing the model's architecture. Step 6: Tune the deep learning model according to the test results, including adjusting the network structure and hyperparameters. Specifically, you can use hyperparameter optimization methods such as grid search or random search to search for the best hyperparameter combination within a certain range. Grid search is an exhaustive search method that traverses all possible combinations in a given hyperparameter space, calculates the performance of the model on the validation set under each combination, and then selects the combination with the best performance. Random search randomly samples some combinations in the hyperparameter space for testing, which is more efficient than grid search, but may not be able to find the global optimal solution. Hyperparameter optimization can improve the performance and generalization ability of the model.

[0027] In one embodiment, specifically, in step five, the device identification result includes device category, device model, device status, device identification time, and device confidence; The method for generating a device identification result comprises the following steps: Step S1, based on the convolution layer and pooling layer of the deep learning model, feature extraction is performed on the input device image of the device to be identified to obtain image features; wherein the image features include information such as shape, texture, and color of the device; Step S2: Input the extracted image features into the fully connected layer of the deep learning model for classification and recognition, and output a preliminary recognition result; wherein the recognition result includes the category, model, status, etc. of the device; Step S3: Calculate the confidence of the preliminary recognition result. If the confidence is lower than the preset threshold, mark it as an uncertain result. Specifically: Assume that the output of the classifier is: ; in, is the number of categories, Indicates that the image belongs to The probability of the class can be calculated by the softmax function: ; in, is the output of the fully connected layer; Confidence can be defined as the maximum probability value, that is ; Step S4, verify and optimize the uncertain results based on the multi-model fusion method to obtain the final recognition result; identify the same device image to be identified by using multiple deep learning models with different architectures, and then combine the results of multiple models according to strategies such as voting method or weighted average method to determine the final recognition result; in addition, the recognition result can be further verified and optimized in combination with the historical data and operating status information of the device to improve the accuracy and reliability of recognition.

[0028] In one embodiment, specifically, the confidence calculation of the device recognition result uses the maximum value output by the softmax function as the confidence value, and the preset threshold is dynamically adjusted according to the importance and recognition difficulty of the device. If the confidence of the preliminary recognition result is lower than the corresponding threshold, it is marked as an uncertain result; Specifically: For core key devices, such as server CPUs and storage arrays, the confidence threshold is set to a higher value [θ1], where θ1 is 0.9; for general devices, such as network switches and ordinary server hard disks, the confidence threshold is set to [θ2], where θ2 is 0.8; when the confidence of the preliminary recognition result is lower than the corresponding threshold, it is marked as an uncertain result for subsequent multi-model fusion verification and optimization processing.

[0029] In one embodiment, specifically, the deep learning model is deployed on a cloud server, and the cloud computing platform is used to achieve efficient data processing and storage, and a remote monitoring interface is provided, allowing remote users to view device recognition results and monitoring videos in real time through a web interface or mobile application, thereby improving management efficiency and flexibility; wherein, the cloud server has powerful computing power and scalable storage resources, and can efficiently process a large amount of image data transmitted in real time from the camera of the data center equipment; through the distributed computing architecture of the cloud computing platform, the deep learning model can quickly complete the image recognition task and return the recognition results in real time; the cloud storage solution ensures the reliability and security of the data, while providing a flexible data access and backup mechanism; deploying the deep learning model on a cloud server, and using the cloud computing platform to achieve efficient data processing and storage, as well as providing a remote monitoring interface, is an efficient, flexible, secure and scalable solution that can significantly improve the efficiency and effectiveness of image recognition of data center equipment.

[0030] Figure 4 is a functional module diagram of a data center equipment image recognition system based on deep learning according to an embodiment of the present application, such as Figure 4 As shown, a data center equipment image recognition system based on deep learning includes: an image acquisition module, an image preprocessing module, a model building module, an image processing module to be recognized, an equipment recognition module, a result output and storage module, a model evaluation and optimization module and a remote monitoring interface module; Image acquisition module, used to obtain key image information of data center equipment through cameras and image acquisition cards installed at key locations in the data center, and record relevant information about the equipment. The camera has autofocus and low-light compensation functions. The autofocus method adopts a hybrid focus technology that combines phase detection autofocus with contrast detection autofocus to ensure that clear images can be obtained under different lighting conditions; The image preprocessing module is used to preprocess the acquired key image information, including image cleaning, image annotation and image enhancement. In particular, image cleaning uses median filtering and Gaussian filtering to remove salt and pepper noise and Gaussian noise in the image to improve image quality. A model building module, which is used to build a deep learning model based on a deep learning algorithm according to the processed key image information. The deep learning model uses a convolutional neural network architecture and is trained, verified and tuned through a training set, a validation set and a test set to recognize device images; The image processing module to be identified is used to obtain the device image of the device to be identified, pre-process the device image, and input the device image into the deep learning model for identification; The device identification module is used to input the device image of the device to be identified into the deep learning model for identification, and generate a device identification result including the device category, device model, device status, device identification time and device confidence. The confidence calculation uses the maximum value output by the softmax function as the confidence value, and can dynamically adjust the preset threshold to mark uncertain results according to the importance and identification difficulty of the device; The result output and storage module is used to store the device identification results in the database and output them in a visual way to generate a device identification report; The model evaluation and optimization module is used to set regular evaluation cycles, evaluate the performance of deep learning models, conduct in-depth analysis of misidentification cases that occur during the evaluation process, and update and optimize the deep learning models based on the analysis results; The remote monitoring interface module is used to deploy the deep learning model on the cloud server, use the cloud computing platform to achieve efficient data processing and storage, and provide a remote monitoring interface, allowing remote users to view device recognition results and monitoring videos in real time through the web interface or mobile application.

[0031] This embodiment provides a data center equipment image recognition system based on deep learning. By constructing a data center equipment image recognition system based on deep learning, it realizes the automatic recognition and monitoring of key equipment in the data center. The system has the advantages of high accuracy, high real-time performance and high scalability, and can significantly improve the intelligence level of data center operation and maintenance.

[0032] For other details about the technical solutions for implementing each module in the system of the above-mentioned embodiment, please refer to the description of a data center equipment image recognition method, system, equipment and medium based on deep learning in the above-mentioned embodiment, which will not be repeated here.

[0033] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0034] like Figure 5 A schematic diagram of an electronic device 40 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0035] like Figure 5 As shown, the electronic device 40 includes at least one processor 41, and a memory connected to the at least one processor 41 in communication, such as a read-only memory (ROM) 42, a random access memory (RAM) 43, etc., wherein the memory stores a computer program that can be executed by at least one processor 41, and the processor 41 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 42 or the computer program loaded from the storage unit 48 to the random access memory (RAM) 43. In RAM43, various programs and data required for the operation of the electronic device 40 can also be stored. The processor 41, ROM42 and RAM43 are connected to each other through a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0036] A number of components in the electronic device 40 are connected to the I / O interface 45, including: an input unit 46, such as a keyboard, a mouse, etc.; an output unit 47, such as various types of displays, speakers, etc.; a storage unit 48, such as a disk, an optical disk, etc.; and a communication unit 49, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0037] The processor 41 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 41 executes the various methods and processes described above, such as a data center equipment image recognition method based on deep learning.

[0038] In some embodiments, a method for image recognition of data center equipment based on deep learning may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 40 via the ROM 42 and / or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the method for image recognition of data center equipment based on deep learning described above may be performed. Alternatively, in other embodiments, the processor 41 may be configured to perform a method for image recognition of data center equipment based on deep learning in any other appropriate manner (e.g., by means of firmware).

[0039] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0040] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0041] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0042] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0043] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0044] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS (Virtual Private Server) services.

[0045] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.

[0046] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limitations. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.

[0047] In the present disclosure, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. The block diagrams of the devices, devices, equipment, and systems involved in the present disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, devices, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0048] Additionally, as used herein, "or" used in a list of items beginning with "at least one" indicates a separate list, so that, for example, a list of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not mean that the example described is preferred or better than other examples.

[0049] It should also be noted that in the system and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.

[0050] Various changes, substitutions, and modifications of the techniques described herein may be made without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of the present disclosure is not limited to the specific aspects of the processes, machines, manufactures, compositions of events, means, methods, and actions described above. Currently existing or later to be developed processes, machines, manufactures, compositions of events, means, methods, or actions that perform substantially the same functions or achieve substantially the same results as the corresponding aspects described herein may be utilized. Thus, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or actions within their scope.

[0051] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0052] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A data center equipment image recognition method based on deep learning, characterized in that: The following steps are involved: Step 1: Obtain key image information of data center equipment and record equipment-related information; Step 2: pre-process the acquired key image information based on image processing technology; Step 3: Based on the processed key image information and deep learning algorithm, a deep learning model is constructed; Step 4: Obtain the device image of the device to be identified and perform preprocessing; Step 5: Input the device image of the device to be identified into the deep learning model for identification, and generate a device identification result; Step 6: Store the device identification results in the database and output them in a visual way to generate a device identification report; Step 7: Set a regular evaluation cycle to evaluate the performance of the deep learning model, conduct in-depth analysis of the misidentification cases that occur during the evaluation process, and generate analysis results; Step 8. Update and optimize the deep learning model based on the analysis results.

2. The method for data center equipment image recognition based on deep learning according to claim 1, characterized in that: In step one, the key image information is collected by cameras and image acquisition cards installed at key locations in the data center; the cameras installed at key locations in the data center have autofocus and low-light compensation functions, and the autofocus method adopts a hybrid focus technology that combines phase detection autofocus with contrast detection autofocus; the device-related information includes device number, device location, device brand, device model, equipment production date, and the last maintenance time of the equipment.

3. The method for data center equipment image recognition based on deep learning according to claim 1, characterized in that: In step 2, the key image information preprocessing includes image cleaning, image annotation and image enhancement; the image cleaning utilizes median filtering and Gaussian filtering to remove salt and pepper noise and Gaussian noise in the image.

4. The method for data center equipment image recognition based on deep learning according to claim 1, characterized in that: In step three, the method for constructing a deep learning model includes the following steps: Step 1: Select the convolutional neural network architecture in the deep learning algorithm as the architecture of the deep learning model; Step 2: Divide the preprocessed key image information into a training set, a validation set, and a test set, where the training set accounts for 70%, the validation set accounts for 20%, and the test set accounts for 10%; Step 3: Select the cross entropy loss function and Adam optimizer as the loss function and optimizer required for deep learning model training; Step 4: Use the training set to train the deep learning model, monitor the performance on the validation set, and adjust the model parameters to prevent overfitting; Step 5: After training is completed, use the test set to evaluate the recognition accuracy and recall rate of the deep learning model and generate test results; Step 6: Tune the deep learning model based on the test results, including adjusting the network structure and hyperparameters.

5. The method for data center equipment image recognition based on deep learning according to claim 1, characterized in that: In step 5, the device identification result includes device category, device model, device status, device identification time and device confidence; The method for generating a device identification result comprises the following steps: Step S1, extracting features of the device image of the device to be identified based on the convolution layer and pooling layer of the deep learning model to obtain image features; Step S2: input the extracted image features into the fully connected layer of the deep learning model for classification and recognition, and output the preliminary recognition results; Step S3: Calculate the confidence of the preliminary recognition result. If the confidence is lower than a preset threshold, mark it as an uncertain result. Step S4: verify and optimize the uncertain results based on multi-model fusion to obtain the final recognition result.

6. The method for data center equipment image recognition based on deep learning according to claim 5, characterized in that: The confidence calculation of the device recognition result uses the maximum value output by the softmax function as the confidence value. The preset threshold is dynamically adjusted according to the importance and recognition difficulty of the device. If the confidence of the preliminary recognition result is lower than the corresponding threshold, it is marked as an uncertain result.

7. The method for data center equipment image recognition based on deep learning according to claim 4, characterized in that: The deep learning model is deployed on a cloud server, using a cloud computing platform to achieve efficient data processing and storage, and provides a remote monitoring interface, allowing remote users to view device recognition results and monitoring videos in real time through a web interface or mobile application.

8. A data center equipment image recognition system based on deep learning, characterized in that: include: Image acquisition module, image preprocessing module, model building module, image processing module to be identified, device identification module, result output and storage module, model evaluation and optimization module and remote monitoring interface module; The image acquisition module is used to obtain key image information of the data center equipment through the camera and image acquisition card installed at the key position of the data center, and record the relevant information of the equipment, wherein the camera has the functions of autofocus and low light compensation, and the autofocus method adopts the hybrid focus technology combining phase detection autofocus and contrast detection autofocus; The image preprocessing module is used to preprocess the acquired key image information, including image cleaning, image annotation and image enhancement, wherein the image cleaning uses median filtering and Gaussian filtering to remove salt and pepper noise and Gaussian noise in the image; The model building module is used to build a deep learning model based on a deep learning algorithm according to the processed key image information, and the deep learning model uses a convolutional neural network architecture to be trained, verified and tuned through a training set, a verification set and a test set; The to-be-recognized image processing module is used to obtain a device image of the to-be-recognized device and pre-process the device image, and input the device image into a deep learning model for recognition; The device identification module is used to input the device image of the device to be identified into the deep learning model for identification, and generate a device identification result including the device category, device model, device status, device identification time and device confidence; The result output and storage module is used to store the device identification results in a database and output them in a visual manner to generate a device identification report; The model evaluation and optimization module is used to set a regular evaluation cycle, evaluate the performance of the deep learning model, conduct in-depth analysis of the misrecognition cases that occur during the evaluation process, and update and optimize the deep learning model based on the analysis results; The remote monitoring interface module is used to deploy the deep learning model on a cloud server, use the cloud computing platform to achieve efficient data processing and storage, and provide a remote monitoring interface, allowing remote users to view device recognition results and monitoring videos in real time through a web interface or mobile application.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a data center equipment image recognition method based on deep learning as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, which are used to enable a computer to execute a data center equipment image recognition method based on deep learning as described in any one of claims 1-7.

Images