Method for improving accuracy of AI identification data of optical cable cross-connecting box
By preprocessing and feature extraction of the data of the optical cable junction box, and using an optimized AI model for identification and feedback optimization, the problems of low accuracy and lack of real-time performance of the optical cable junction box AI recognition technology in complex environments are solved, and efficient and accurate optical cable junction box management is achieved.
Patent Information
- Application Number
- CN202510021569.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-07
- Publication Date
- 2025-04-15
AI Technical Summary
The AI recognition technology of optical cable junction boxes is low in complex environments, making it difficult to adapt to different brands and models of optical cable junction boxes, and lacks real-time and efficient.
Through the steps of data acquisition, preprocessing, resource capacity center construction, AI model introduction and optimization, AI model evaluation and optimization, and real-time identification and comparison, the accuracy of AI recognition in optical cable junction box is improved. Specifically, it includes preprocessing and feature extraction of image data, identifying using an optimized AI model, and optimizing the model through real-time feedback.
It improves the accuracy and stability of AI identification of optical cable junction box, realizes accurate and efficient management of optical cable junction box identification data, reduces labor costs, and improves the management efficiency of communication networks.
Smart Images
Figure CN119832277A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of communication information technology, and in particular to a method for improving the accuracy of AI recognition data in an optical cable junction box. Background Art
[0002] The optical cable junction box, also commonly known as the street cabinet, is usually placed on the trunk optical cable. It is a junction device that provides optical cable termination and jumpering for the trunk layer optical cable and the distribution layer optical cable. However, in the management and maintenance process of the optical cable junction box, the traditional method often relies on manual inspection and identification, which is not only inefficient but also costly.
[0003] AI recognition capability refers to a technology that uses computer algorithms and models to simulate human recognition capabilities. It can automatically recognize various types of information such as images, voices, and texts by learning from large amounts of data. In the application of AI recognition in optical cable junction boxes, previous technologies have many shortcomings.
[0004] First of all, due to the complexity of the environment in which the optical cable junction box is located (such as light changes, obstructions, dirt accumulation, etc.), the accuracy of traditional AI recognition technology is often difficult to guarantee, and misjudgments and missed judgments are prone to occur.
[0005] Secondly, there are many types of optical cable junction boxes, and different brands and models differ in appearance and structure, which puts higher requirements on the generalization ability of AI models, and previous technologies have performed poorly in this regard.
[0006] Finally, traditional AI recognition systems often lack real-time and efficiency.
[0007] In order to solve the above problems, the present invention proposes a method for improving the accuracy of AI recognition data of optical cable junction box. Summary of the invention
[0008] In order to make up for the defects of the prior art, the present invention provides a simple and efficient method for improving the accuracy of AI recognition data of optical cable junction boxes.
[0009] The present invention is achieved through the following technical solutions:
[0010] A method for improving the accuracy of AI recognition data of an optical cable junction box, characterized in that it comprises the following steps:
[0011] Step S1, data collection
[0012] The data acquisition module takes photos and collects data from the optical cable junction box equipment, and transmits the image data to the resource capability center; Step S2, data preprocessing
[0013] The preprocessing module performs preprocessing operations on the collected image data to improve the image quality and data consistency;
[0014] In step S2, the preprocessing operations include denoising, missing value filling, outlier processing, enhancement, cropping, feature selection, extraction and dimensionality reduction.
[0015] Step S3: Resource capability center construction
[0016] Integrate basic data, operation status and historical fault information of optical cable junction boxes in the resource capability center;
[0017] Step S4: AI model introduction and optimization
[0018] Customize and introduce AI models according to scenario requirements, build training data sets using historical fault information, and train and optimize parameters of the selected AI models;
[0019] In step S4, the AI model selects an optical cross-connect box idle terminal image recognition model or an optical cross-connect box flange component image recognition model.
[0020] In step S4, the best parameter combination is found through grid search and Bayesian optimization method during the training process to improve the recognition accuracy of the model.
[0021] Step S5: AI model evaluation and optimization
[0022] Use the test data set to evaluate the trained AI model and evaluate the performance of the AI model by calculating the accuracy and recall rate;
[0023] Optimize the AI model based on the evaluation results, including adjusting the network structure, increasing the amount of training data, and customizing the optimization algorithm to further improve the accuracy of the model;
[0024] Step S6: real-time identification and comparison
[0025] Deploy the evaluated and optimized AI model to actual scenarios to perform real-time identification of optical cable junction boxes;
[0026] By comparing the recognition results with the actual situation, evaluating the accuracy of the model and collecting feedback data, the AI model is continuously optimized and improved based on real-time feedback data to improve data accuracy.
[0027] A system for improving the accuracy of AI recognition data of an optical cable junction box, comprising a data acquisition module, a preprocessing module, a resource capability center, an AI model introduction and optimization module, an AI model evaluation and optimization module, and a real-time recognition and comparison module;
[0028] The data acquisition module is responsible for taking photos and collecting data from the optical cable junction box equipment, and transmitting the image data to the resource capability center;
[0029] The preprocessing module is responsible for preprocessing the collected image data to improve the image quality and data consistency;
[0030] The preprocessing operations of the preprocessing module include denoising, missing value filling, outlier processing, enhancement, cropping, feature selection, extraction and dimensionality reduction.
[0031] Resource capability center, responsible for integrating the basic data, operation status and historical fault information of the optical cable junction box;
[0032] The AI model introduction and optimization module is responsible for customizing the introduction of AI models according to scenario requirements, building training data sets using historical fault information, and training and parameter tuning the selected AI models;
[0033] The AI model introduction and optimization module uses an optical cross-connect box idle terminal image recognition model or an optical cross-connect box flange component image recognition model.
[0034] During the training process, the AI model introduction and optimization module searches for the best parameter combination through grid search and Bayesian optimization methods to improve the recognition accuracy of the model.
[0035] The AI model evaluation and optimization module is responsible for evaluating the trained AI model using the test data set, evaluating the performance of the AI model by calculating the accuracy and recall rate; and optimizing the AI model based on the evaluation results, including adjusting the network structure, increasing the amount of training data, and customizing the optimization algorithm to further improve the accuracy of the model;
[0036] The real-time identification and comparison module is responsible for deploying the evaluated and optimized AI model to the actual scenario and performing real-time identification of the optical cable junction box. It also evaluates the accuracy of the model and collects feedback data by comparing the identification results with the actual situation. It continuously optimizes and improves the AI model based on the real-time feedback data to improve data accuracy.
[0037] A device for improving the accuracy of AI recognition data of an optical cable junction box, characterized in that it includes a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the above method steps when executing the computer program.
[0038] A readable storage medium, characterized in that: a computer program is stored on the readable storage medium, and the computer program implements the above method steps when executed by a processor.
[0039] The beneficial effects of the present invention are as follows: the method for improving the accuracy of AI recognition data of the optical cable junction box effectively reduces the error and redundancy of image data and improves the generalization ability of the model through preprocessing and feature extraction of image data; the optimized algorithm and model structure are adopted to better adapt to the characteristics of the optical cable junction box image and improve the accuracy and stability of recognition; through the introduction, optimization and real-time recognition and feedback of the AI model, accurate and efficient management of the recognition data of the optical cable junction box is achieved, the labor cost is reduced, and the management efficiency of the communication network is improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0041] Attached Figure 1 A schematic diagram of a method for improving the accuracy of AI recognition data in an optical cable junction box according to the present invention. DETAILED DESCRIPTION
[0042] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0043] The method for improving the accuracy of AI recognition data of an optical cable junction box comprises the following steps:
[0044] Step S1, data collection
[0045] The data acquisition module takes photos and collects data from the optical cable junction box equipment, and transmits the image data to the resource capability center; Step S2, data preprocessing
[0046] The preprocessing module performs preprocessing operations on the collected image data to improve the image quality and data consistency;
[0047] In step S2, the preprocessing operations include denoising, missing value filling, outlier processing, enhancement, cropping, feature selection, extraction and dimensionality reduction.
[0048] Step S3: Resource capability center construction
[0049] Integrate basic data, operation status and historical fault information of optical cable junction boxes in the resource capability center;
[0050] Step S4: AI model introduction and optimization
[0051] Customize and introduce AI models according to scenario requirements, build training data sets using historical fault information, and train and optimize parameters of the selected AI models;
[0052] In step S4, the AI model selects an optical cross-connect box idle terminal image recognition model or an optical cross-connect box flange component image recognition model.
[0053] In step S4, the best parameter combination is found through grid search and Bayesian optimization method during the training process to improve the recognition accuracy of the model.
[0054] Step S5: AI model evaluation and optimization
[0055] Use the test data set to evaluate the trained AI model and evaluate the performance of the AI model by calculating the accuracy and recall rate;
[0056] Optimize the AI model based on the evaluation results, including adjusting the network structure, increasing the amount of training data, and customizing the optimization algorithm to further improve the accuracy of the model;
[0057] Step S6: real-time identification and comparison
[0058] Deploy the evaluated and optimized AI model to actual scenarios to perform real-time identification of optical cable junction boxes;
[0059] By comparing the recognition results with the actual situation, evaluating the accuracy of the model and collecting feedback data, the AI model is continuously optimized and improved based on real-time feedback data to improve data accuracy.
[0060] The system for improving the accuracy of AI recognition data in the optical cable junction box includes a data acquisition module, a preprocessing module, a resource capability center, an AI model introduction and optimization module, an AI model evaluation and optimization module, and a real-time recognition and comparison module;
[0061] The data acquisition module is responsible for taking photos and collecting data from the optical cable junction box equipment, and transmitting the image data to the resource capability center;
[0062] The preprocessing module is responsible for preprocessing the collected image data to improve the image quality and data consistency;
[0063] The preprocessing operations of the preprocessing module include denoising, missing value filling, outlier processing, enhancement, cropping, feature selection, extraction and dimensionality reduction.
[0064] Resource capability center, responsible for integrating the basic data, operation status and historical fault information of the optical cable junction box;
[0065] The AI model introduction and optimization module is responsible for customizing the introduction of AI models according to scenario requirements, building training data sets using historical fault information, and training and parameter tuning the selected AI models;
[0066] The AI model introduction and optimization module uses an optical cross-connect box idle terminal image recognition model or an optical cross-connect box flange component image recognition model.
[0067] During the training process, the AI model introduction and optimization module searches for the best parameter combination through grid search and Bayesian optimization methods to improve the recognition accuracy of the model.
[0068] The AI model evaluation and optimization module is responsible for evaluating the trained AI model using the test data set, evaluating the performance of the AI model by calculating the accuracy and recall rate; and optimizing the AI model based on the evaluation results, including adjusting the network structure, increasing the amount of training data, and customizing the optimization algorithm to further improve the accuracy of the model;
[0069] The real-time identification and comparison module is responsible for deploying the evaluated and optimized AI model to the actual scenario and performing real-time identification of the optical cable junction box. It also evaluates the accuracy of the model and collects feedback data by comparing the identification results with the actual situation. It continuously optimizes and improves the AI model based on the real-time feedback data to improve data accuracy.
[0070] The device for improving the accuracy of AI recognition data in an optical cable junction box comprises a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the above method steps when executing the computer program.
[0071] The readable storage medium stores a computer program, and when the computer program is executed by a processor, the above method steps are implemented.
[0072] Compared with the existing technology, this method for improving the accuracy of AI recognition data of optical cable junction boxes has the following characteristics:
[0073] By preprocessing and extracting features of image data, the error and redundancy of image data are effectively reduced, and the generalization ability of the model is improved;
[0074] The optimized algorithm and model structure can better adapt to the characteristics of the optical cable junction box image and improve the accuracy and stability of recognition;
[0075] Through the introduction, optimization, real-time identification and feedback of AI models, accurate and efficient management of optical cable junction box identification data is achieved, which reduces labor costs and improves the management efficiency of communication networks.
[0076] The embodiment described above is only one specific implementation of the present invention. Common changes and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for improving the accuracy of AI recognition data of optical cable junction boxes, characterized by: The following steps are involved: Step S1, data collection The data acquisition module takes photos and collects data from the optical cable junction box equipment, and transmits the image data to the resource capability center; Step S2: Data preprocessing The preprocessing module performs preprocessing operations on the collected image data to improve the image quality and data consistency; Step S3: Resource capability center construction Integrate basic data, operation status and historical fault information of optical cable junction boxes in the resource capability center; Step S4: AI model introduction and optimization Customize and introduce AI models according to scenario requirements, build training data sets using historical fault information, and train and optimize parameters of the selected AI models; Step S5: AI model evaluation and optimization Use the test data set to evaluate the trained AI model and evaluate the performance of the AI model by calculating the accuracy and recall rate; Optimize the AI model based on the evaluation results, including adjusting the network structure, increasing the amount of training data, and customizing the optimization algorithm to further improve the accuracy of the model; Step S6: real-time identification and comparison Deploy the evaluated and optimized AI model to actual scenarios to perform real-time identification of optical cable junction boxes; By comparing the recognition results with the actual situation, evaluating the accuracy of the model and collecting feedback data, the AI model is continuously optimized and improved based on real-time feedback data to improve data accuracy.
2. The method for improving the accuracy of AI recognition data of optical cable junction box according to claim 1 is characterized in that: In step S2, the preprocessing operations include denoising, missing value filling, outlier processing, enhancement, cropping, feature selection, extraction and dimensionality reduction.
3. The method for improving the accuracy of AI recognition data of optical cable junction box according to claim 1 is characterized in that: In step S4, the AI model selects an optical cross-connect box idle terminal image recognition model or an optical cross-connect box flange component image recognition model.
4. The method for improving the accuracy of AI recognition data of optical cable junction box according to claim 1 is characterized in that: In step S4, the best parameter combination is found through grid search and Bayesian optimization method during the training process to improve the recognition accuracy of the model.
5. A system for improving the accuracy of AI recognition data in optical cable junction boxes, characterized by: It includes data acquisition module, preprocessing module, resource capability center, AI model introduction and optimization module, AI model evaluation and optimization module, and real-time recognition and comparison module; The data acquisition module is responsible for taking photos and collecting data from the optical cable junction box equipment, and transmitting the image data to the resource capability center; The preprocessing module is responsible for preprocessing the collected image data to improve the image quality and data consistency; Resource capability center, responsible for integrating the basic data, operation status and historical fault information of the optical cable junction box; The AI model introduction and optimization module is responsible for customizing the introduction of AI models according to scenario requirements, building training data sets using historical fault information, and training and parameter tuning the selected AI models; The AI model evaluation and optimization module is responsible for evaluating the trained AI model using the test data set, evaluating the performance of the AI model by calculating the accuracy and recall rate; and optimizing the AI model based on the evaluation results, including adjusting the network structure, increasing the amount of training data, and customizing the optimization algorithm to further improve the accuracy of the model; The real-time identification and comparison module is responsible for deploying the evaluated and optimized AI model to the actual scenario and performing real-time identification of the optical cable junction box. It also evaluates the accuracy of the model and collects feedback data by comparing the identification results with the actual situation. It continuously optimizes and improves the AI model based on the real-time feedback data to improve data accuracy.
6. The system for improving the accuracy of AI recognition data of optical cable junction boxes according to claim 5 is characterized in that: The preprocessing operations of the preprocessing module include denoising, missing value filling, outlier processing, enhancement, cropping, feature selection, extraction and dimensionality reduction.
7. The system for improving the accuracy of AI recognition data of optical cable junction boxes according to claim 5 is characterized in that: The AI model introduction and optimization module uses an optical cross-connect box idle terminal image recognition model or an optical cross-connect box flange component image recognition model.
8. The system for improving the accuracy of AI recognition data of optical cable junction boxes according to claim 5 is characterized in that: During the training process, the AI model introduction and optimization module searches for the best parameter combination through grid search and Bayesian optimization methods to improve the recognition accuracy of the model.
9. A device for improving the accuracy of AI recognition data in optical cable junction boxes, characterized in that: The method comprises a memory and a processor; the memory is used to store a computer program, and the processor is used to implement the method steps as claimed in any one of claims 1 to 4 when executing the computer program.
10. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps according to any one of claims 1 to 4 are implemented.