Old equipment data acquisition system and method based on image recognition

Abstract

The invention belongs to the technical field of measurement and detection, and particularly relates to an old equipment data acquisition system and method based on image recognition. The invention provides an old equipment data acquisition system based on image recognition. The old equipment data acquisition system comprises the following modules: an image acquisition module, an image preprocessing module, an image recognition module, a data analysis and verification module and a data uploading and management module. The specific acquisition method comprises the steps of image acquisition, image preprocessing, image recognition, data analysis and verification, data uploading and management and the like. The invention provides an efficient, accurate and automatic data acquisition system and method.

Description

Technical Field

[0001] The present invention belongs to the technical field of measurement and detection, and specifically relates to a system and method for collecting data of old equipment based on image recognition. Background Art

[0002] In the field of metrology and testing, old equipment such as digital scales and digital hardness testers are often unable to directly export readable files through data interfaces due to technical limitations or equipment aging. The data reading of these devices traditionally relies on manual operation, that is, the operator needs to directly observe the display screen of the device and record the readings. However, this method has many disadvantages:

[0003] Inefficiency: Traditional manual reading methods require a lot of time and effort, especially in scenarios where large-scale testing or frequent data reading is required, which will significantly increase working time and costs and is extremely inefficient.

[0004] Error-prone: Manual reading of data is easily affected by factors such as visual fatigue and distraction, resulting in reading errors, which in turn affects the accuracy of subsequent measurement and test results.

[0005] High cost: A large amount of human resources is required to perform data reading tasks, which increases the unit's operating costs.

[0006] Low degree of automation: In the existing technology, there are few automated data collection solutions for this type of old equipment, which cannot meet the needs of modern laboratories for efficient and accurate data collection.

[0007] With the advancement of science and technology, especially the development of image processing technology and artificial intelligence technology, new solutions have been provided for the data collection of old equipment. Image recognition technology can automatically analyze the image content and extract useful information, thereby realizing the automatic reading of old equipment. However, the current image recognition technology used to automatically read the reading of old equipment has the following technical problems:

[0008] Poor versatility: Existing technologies are often optimized for specific types of old equipment and lack broad adaptability to different types of equipment. This means that different recognition algorithms may need to be developed for old equipment of different brands, models or types, increasing development and maintenance costs.

[0009] Low recognition accuracy: The display screen of old equipment may have problems such as wear, stains, and reflections, which leads to reduced image quality and affects the accuracy of image recognition. In addition, the complex reading methods of some devices, such as multiple scales and non-standard fonts, also increase the difficulty of recognition.

[0010] Complex operation: The operation process of existing technical solutions is relatively complicated, requiring users to have a certain technical background or professional knowledge, which increases the difficulty of learning and using. At the same time, the poor compatibility and interoperability between different solutions also affects the user experience.

[0011] Insufficient data security and privacy protection: During data transmission and storage, existing technologies may lack sufficient security measures to protect the integrity and privacy of data. This may lead to data leakage or illegal use, bringing potential risks to users and organizations.

[0012] High maintenance costs: Due to the slow replacement of old equipment and the lack of corresponding technical support and maintenance services in the market, the long-term maintenance costs of existing technical solutions may be high. In addition, as technology continues to develop, existing solutions may need to be constantly updated and upgraded to adapt to new needs and challenges.

[0013] Aiming at the problems of low efficiency, easy error, high cost and low automation in data collection of the above-mentioned old equipment, the present invention proposes a data collection system and method for old equipment based on image recognition to solve the deficiencies and problems in data collection of old equipment, improve the efficiency and accuracy of data collection, reduce the user's operation difficulty and maintenance cost, and strengthen the security and privacy protection of data. Summary of the invention

[0014] The present invention proposes a data collection system and method for old equipment based on image recognition, which can efficiently, accurately and automatically collect, transmit, store and use data.

[0015] In order to achieve the above object, the present invention solves the above technical problems through the following technical solutions:

[0016] An old equipment data collection system based on image recognition includes the following modules:

[0017] Image acquisition module: used for image acquisition to obtain clear and complete image data;

[0018] Image preprocessing module: preprocess the image to improve image quality;

[0019] Image recognition module: performs text recognition on preprocessed images;

[0020] Data analysis and verification module: structure the identification data and perform data verification;

[0021] Data upload and management module: upload data to the designated storage platform for user convenience.

[0022] A method for collecting data of old equipment based on image recognition comprises the following steps:

[0023] S1: Image acquisition: Use a high-resolution camera to take real-time or timed photos of the display screen of old equipment that does not have the data export function to obtain clear and complete image data;

[0024] S2: Image preprocessing: grayscale conversion, adaptive binarization, contrast adjustment, blur processing and other preprocessing operations are performed on the images collected in S1 to improve image quality and facilitate subsequent image recognition processing;

[0025] S3: Image recognition: Use optical character recognition technology to perform text recognition on the image pre-processed by S2 to extract the digital or character information on the display screen;

[0026] S4: Data parsing and verification: Parse the data identified in S3, convert it into a structured data format, and implement a strict data verification strategy to ensure the accuracy and reliability of the data;

[0027] S5: Data upload and management: Upload the data verified by S4 to the designated data storage platform or laboratory information management system through a security protocol, and provide data management and query functions to facilitate users to view and analyze equipment data at any time.

[0028] In S1, a high-definition camera with automatic focus and exposure control functions is selected; appropriate image resolution and frame rate are set to balance image quality and processing speed; exposure time and white balance settings are automatically or manually adjusted to ensure accurate image color and moderate brightness.

[0029] The specific method in S2 is:

[0030] Grayscale processing: Use OpenCV's image processing function to convert the color image into a grayscale image;

[0031] Adaptive binarization: dynamically calculate the threshold value based on the local brightness of the grayscale image to adapt to uneven lighting conditions;

[0032] Contrast adjustment: adjust the contrast and brightness through the two parameters of alpha and beta;

[0033] Blur processing: Applying Gaussian blur to smooth the image can reduce image noise and provide smoother input for subsequent operations.

[0034] The specific steps in S3 are:

[0035] Text detection: Use the EAST model and the DB model to locate all text regions in the image;

[0036] Text direction classification: The system automatically detects the text direction and adjusts it accordingly to ensure that the text can be recognized correctly;

[0037] Text recognition: Use optical character recognition (OCR) technology to efficiently and accurately identify the text area extracted from the image;

[0038] Post-processing: After text recognition is completed, post-processing operations are performed to optimize the recognition results;

[0039] Output result: The result after the above processing, output text box position, recognized text and text direction information.

[0040] The data in S4 is parsed using regular expression matching or a custom parsing algorithm.

[0041] The data verification in S4 needs to verify the reasonable range of data and perform logical consistency check on data with logical relationship.

[0042] The S5 includes the steps of data encryption and transmission and data storage and management.

[0043] The data is encrypted before transmission, and then the HTTPS protocol is used to encrypt the data for transmission.

[0044] Upload the verified data to the designated storage platform or LIMS, and provide data management and query functions.

[0045] Based on the implementation of the above technical solution, the present invention can achieve the following technical effects:

[0046] 1. The present invention provides a data collection system and method for old equipment based on image recognition. By optimizing image preprocessing and recognition algorithms, the adaptability to display screens of different types of old equipment is improved, so that the method can be widely used in old equipment of various brands, models and types, thereby improving versatility.

[0047] 2. The present invention provides a data collection system and method for old equipment based on image recognition, which introduces more advanced image recognition technology and data verification strategies to improve the recognition accuracy and reliability in complex environments. At the same time, in view of the common problems of wear and stains on old equipment, a special image processing algorithm is designed to reduce its impact on the recognition effect and improve the recognition accuracy.

[0048] 3. The present invention provides a data collection system and method for old equipment based on image recognition, and designs an intuitive and easy-to-use operation interface and process to reduce the user's learning cost and difficulty of use. At the same time, it ensures that the method provided by the present invention has good compatibility and interoperability with the mainstream data management systems and platforms on the market, and simplifies the operation process.

[0049] 4. The present invention provides a data collection system and method for old equipment based on image recognition. In the process of data transmission and storage, encryption, verification and other security measures are used to protect the integrity and privacy of data. At the same time, relevant laws, regulations and industry standards are followed to ensure that the method meets the regulatory requirements when processing sensitive information and strengthen data security and privacy protection.

[0050] 5. The present invention provides a data collection system and method for old equipment based on image recognition, which provides flexible configuration options and scalable architecture design, so that the method provided by the present invention can be easily upgraded and expanded to adapt to new needs and challenges. At the same time, detailed technical documentation and user support services are provided to reduce the user's maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 It is a data collection process diagram provided by the present invention;

[0053] Figure 2 It is an image recognition process diagram provided by the present invention;

[0054] Figure 3 The present invention provides a laboratory information management system for obtaining equipment data. DETAILED DESCRIPTION

[0055] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0056] In the description of the embodiments of the present invention, it should be noted that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limitations on the embodiments of the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0057] In the description of the embodiments of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the embodiments of the present invention can be understood according to specific circumstances.

[0058] In the embodiments of the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0059] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0060] Combine the following Figure 1-Figure 3 The present invention provides a data collection system and method for old equipment based on image recognition.

[0061] In an embodiment of the present invention, a data collection system for old equipment based on image recognition includes the following modules:

[0062] Image acquisition module: used for image acquisition, obtaining clean and complete image data;

[0063] Image preprocessing module: preprocess the image to improve image quality;

[0064] Image recognition module: performs text recognition on preprocessed images;

[0065] Data analysis and verification module: structure the identification data and perform data verification;

[0066] Data upload and management module: upload data to the designated storage platform for user convenience.

[0067] Furthermore, a method for collecting data of old equipment based on image recognition is provided, comprising the following steps:

[0068] S1: Image acquisition: Use a high-resolution camera to take real-time or timed photos of the display screen of old equipment that does not have the data export function to obtain clear and complete image data;

[0069] S2: Image preprocessing: grayscale conversion, contrast adjustment, blur processing, adaptive binarization and other preprocessing operations are performed on the images collected in S1 to improve image quality and facilitate subsequent image recognition processing;

[0070] S3: Image recognition: Use optical character recognition technology to perform text recognition on the image pre-processed by S2 to extract the digital or character information on the display screen;

[0071] S4: Data parsing and verification: Parse the data identified in S3, convert it into a structured data format, and implement a strict data verification strategy to ensure the accuracy and reliability of the data;

[0072] S5: Data upload and management: Upload the data verified by S4 to the designated data storage platform or laboratory information management system through a security protocol, and provide data management and query functions to facilitate users to view and analyze equipment data at any time.

[0073] Combination Figure 1 As shown, the S1 image acquisition steps are as follows:

[0074] (a) Hardware equipment selection:

[0075] HD camera: To ensure that clear and complete images can be captured under different lighting conditions, the camera should have functions such as automatic focus and exposure control to adapt to the display or dial characteristics of different devices.

[0076] Fixing device: Design a special fixing device according to the height of the shooting equipment to ensure that the camera can be stably aimed at the target device and reduce jitter and deviation during shooting.

[0077] (c) Shooting parameter settings:

[0078] Resolution and frame rate: According to the display characteristics of the target device, set the appropriate image resolution and frame rate to balance image quality and processing speed. Currently, the supported display resolutions are 480P or below, and the frame rate is 30 frames or below.

[0079] Exposure and White Balance: Automatically or manually adjust exposure time and white balance settings based on ambient light conditions to ensure accurate image color and moderate brightness.

[0080] Further, the S2 image preprocessing is specifically performed as follows:

[0081] (a) Grayscale processing: Use OpenCV's image processing functions to convert color images into grayscale images. The purpose of this step is to reduce channel information and retain only brightness information to simplify subsequent operations. If grayscale is not enabled and a grayscale image is required in subsequent steps (such as binarization), the code will automatically process it into grayscale.

[0082] (b) Adaptive binarization: Dynamically calculate the threshold value based on the local brightness of the grayscaled image to adapt to uneven lighting conditions. Before performing this operation, the code ensures that the input image is already a grayscale image and processes it according to the block size (blockSize) and constant adjustment value (C) provided by the configuration file. The block size needs to be an odd number. If the value is an even number, the code will automatically add 1 to correct it to ensure normal execution of the algorithm.

[0083] (c) Contrast adjustment: Contrast adjustment is controlled by two parameters: alpha is used to adjust contrast, and beta is used to adjust brightness. The code reads the corresponding value in the configuration file, and then calls the OpenCV function to perform a linear transformation on each pixel of the image, thereby improving the visual effect and making the image clearer.

[0084] (d) Blur processing (denoising): Apply Gaussian blur to smooth the image. This is achieved by specifying the size of a blur kernel. The larger the kernel, the more significant the blur effect. This process can reduce image noise and provide smoother input for subsequent operations.

[0085] Combination Figure 2 As shown, in the S3 image recognition step, the specific steps are:

[0086] (a) Text Detection: Text detection is the first step in the recognition process, and its goal is to locate all text regions in the image. The models used are EAST (Efficient and Accurate Scene Text) and DB (Differentiable Binarization) models.

[0087] EAST model: EAST is a fast and efficient text detection algorithm that uses a fully convolutional neural network (FCN) to detect text regions in images. The model can recognize text in various directions and forms, and does not rely on the preset text box shape. The key parameters of the EAST model include score_thresh (score threshold, used to filter low-scoring text boxes), box_thresh (box score threshold), and max_candidates (maximum number of candidate text boxes).

[0088] DB model: DB model is an end-to-end text detection method that can perform better when processing rotated and curved text through differentiable binarization technology. Its main parameters include thresh (binarization threshold), det_db_score_mode (scoring mode), and det_db_thresh (threshold for binarization).

[0089] (b) Text orientation classification: In some scenarios, the text in the image may exist in different orientations (e.g., vertical or tilted text). To address this problem, the system automatically detects the orientation of the text and adjusts it accordingly to ensure that the text can be correctly recognized. Common orientation categories include horizontal, 90 degrees, 180 degrees, and 270 degrees. Related parameters include det_max_dim (maximum detection area size) and det_db_thresh (text detection threshold), which are used to determine the orientation of the text.

[0090] (c) Text recognition: Text recognition is a core task of OCR, which uses deep learning models to identify text regions extracted from images. Our model uses SVTR_LCNetV3 as the backbone network, which is a hybrid architecture that combines convolutional neural networks (CNNs) and Transformers. The key feature of this model is the way it handles image text features, using a lighter convolutional network for feature extraction and modeling global context through Transformers, thereby achieving excellent recognition results in different text scenarios.

[0091] LCNetV3 (Light-weight Convolutional Network): LCNet is a lightweight convolutional neural network designed for mobile devices and devices with limited resources. LCNetV3 is optimized on this basis, reducing the amount of calculation while maintaining a high feature extraction capability. It speeds up the inference process by reducing the number of convolutional layers and using efficient convolution operations.

[0092] SVTR (Sequence Vision Transformer): SVTR is a visual Transformer that can capture the global context of an image. The Transformer’s self-attention mechanism is well suited for processing long-distance dependencies in text, so it can better understand character relationships in long texts and complex layouts.

[0093] (d) Post-processing: After text recognition is complete, the OCR model performs post-processing operations to optimize the recognition results. First, the text boxes are fused to merge overlapping or adjacent text boxes to ensure that each text area can be accurately recognized. Then, the system will piece together the recognized characters into complete words or sentences. Finally, if errors or incomplete characters are detected in the recognition results, the model will correct them through a dictionary or language model.

[0094] (e) Output results: After a series of processing, the original output of the model is:

[0095] ① Position of text box: the coordinates of each identified text area.

[0096] ②Recognized text: specific characters or words recognized in each text area.

[0097] ③Text direction information: If direction classification is enabled, output the direction classification result of the text.

[0098] Further, the specific steps of data analysis and verification in S4 are as follows:

[0099] (a) Data analysis:

[0100] Regular expression matching: For data with a fixed format (such as date, time, etc.), regular expressions can be used for matching and parsing.

[0101] Custom parsing algorithm: Write a custom parsing algorithm to process data according to the specific format and rules of the target data.

[0102] (b) Data verification:

[0103] Range check: Check the data according to its reasonable range to ensure that the data is within the expected range.

[0104] Logical consistency check: For data with logical relationships (such as the relationship between maximum and minimum values), a logical consistency check is performed to ensure the correctness of the data.

[0105] Combination Figure 3 As shown, the specific steps of S5 data uploading and management are:

[0106] (a) Data encryption and transmission:

[0107] HTTPS protocol: HTTPS protocol is used to encrypt data transmission to ensure the security of data transmission.

[0108] Data encryption algorithm: Encrypt data before transmission to prevent the data from being stolen or tampered with during transmission.

[0109] (b) Data storage and management:

[0110] Database design: Use MYSQL database to store parsed data, including table structure, indexes, constraints, etc.

[0111] Query interface: Provides a convenient query interface for users to retrieve and analyze data.

[0112] The present invention introduces image recognition technology (OCR algorithm) in the field of data collection of old equipment, realizing the innovation of traditional data collection methods. This technological innovation not only improves the efficiency and accuracy of data collection, but also reduces labor costs and improves the degree of automation.

[0113] The present invention significantly improves the efficiency of data collection and reduces manual intervention and waiting time by automating the collection and processing process. This is particularly important for scenarios where frequent data collection is required, and can greatly improve work efficiency.

[0114] The present invention uses advanced image processing algorithms and deep learning models for image recognition and data analysis, ensuring the accuracy and reliability of the data. This is of great significance for scenarios that require high-precision data and can avoid decision-making errors caused by data errors.

[0115] The present invention adopts an automated collection and processing process to reduce dependence on human resources, reduce labor costs, and can directly reduce operating costs and improve economic benefits.

[0116] The present invention can be widely applied to various types of old equipment, such as digital display scales, pointer-type tensile machines, etc., and has broad application prospects and market potential.

[0117] The present invention adopts HTTPS protocol to perform encrypted transmission during data transmission to ensure the security of data transmission, which helps to protect the unit data from being leaked or tampered with and maintain the unit's information security.

[0118] The present invention has been rigorously tested and verified to have high stability and reliability, which helps to ensure the continuity and stability of data acquisition and processing, and avoid data loss or processing interruption due to system failure.

[0119] The method provided by the present invention has been tried out in the metrology and testing center of this unit and can meet the needs of data collection. The accuracy of image recognition reaches more than 98%, which greatly improves the detection efficiency.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An old equipment data acquisition system based on image recognition, characterized in that: Includes the following modules: Image acquisition module: used for image acquisition to obtain clear and complete image data; Image preprocessing module: preprocess the image to improve image quality; Image recognition module: performs text recognition on preprocessed images; Data analysis and verification module: structure the identification data and perform data verification; Data upload and management module: upload data to the designated storage platform for user convenience.

2. A method for collecting data from old equipment based on image recognition, characterized in that: The following steps are involved: S1: Image acquisition: Use a high-resolution camera to take real-time or timed photos of the display screen of old equipment that does not have the data export function to obtain clear and complete image data; S2: Image preprocessing: grayscale conversion, adaptive binarization, contrast adjustment, blur processing and other preprocessing operations are performed on the images collected in S1 to improve image quality and facilitate subsequent image recognition processing; S3: Image recognition: Use optical character recognition technology to perform text recognition on the image pre-processed by S2 to extract the digital or character information on the display screen; S4: Data parsing and verification: Parse the data identified in S3, convert it into a structured data format, and implement a strict data verification strategy to ensure the accuracy and reliability of the data; S5: Data upload and management: Upload the data verified by S4 to the designated data storage platform or laboratory information management system through a security protocol, and provide data management and query functions to facilitate users to view and analyze equipment data at any time.

3. The old equipment data collection method based on image recognition according to claim 2 is characterized in that: In S1, a high-definition camera with automatic focus and exposure control functions is selected; appropriate image resolution and frame rate are set to balance image quality and processing speed; exposure time and white balance settings are automatically or manually adjusted to ensure accurate image color and moderate brightness.

4. The old equipment data collection method based on image recognition according to claim 3 is characterized in that: The specific method in S2 is: Grayscale processing: Use OpenCV's image processing function to convert the color image into a grayscale image; Adaptive binarization: dynamically calculate the threshold value based on the local brightness of the grayscale image to adapt to uneven lighting conditions; Contrast adjustment: adjust the contrast and brightness through the two parameters of alpha and beta; Blur processing: Applying Gaussian blur to smooth the image can reduce image noise and provide smoother input for subsequent operations.

5. The old equipment data collection method based on image recognition according to claim 4 is characterized in that: The specific steps in S3 are: Text detection: Use the EAST model and the DB model to locate all text regions in the image; Text direction classification: The system automatically detects the text direction and adjusts it accordingly to ensure that the text can be recognized correctly; Text recognition: Use optical character recognition (OCR) technology to efficiently and accurately identify the text area extracted from the image; Post-processing: After text recognition is completed, post-processing operations are performed to optimize the recognition results; Output result: The result after the above processing, output text box position, recognized text and text direction information.

6. The old equipment data collection method based on image recognition according to claim 5 is characterized in that: The data in S4 is parsed using regular expression matching or a custom parsing algorithm.

7. The old equipment data collection method based on image recognition according to claim 6 is characterized in that: The data verification in S4 needs to verify the reasonable range of data and perform logical consistency check on data with logical relationship.

8. The old equipment data collection method based on image recognition according to claim 7 is characterized in that: The S5 includes the steps of data encryption and transmission and data storage and management.

9. The old equipment data collection method based on image recognition according to claim 8 is characterized in that: The data is encrypted before transmission, and then the HTTPS protocol is used to encrypt the data for transmission.

10. The old equipment data collection method based on image recognition according to claim 9 is characterized in that: Upload the verified data to the designated storage platform or LIMS, and provide data management and query functions.

Images