Light column digital display instrument reading verification method, system, device and storage medium

Abstract

A method, system, device, and storage medium for verifying readings of a bar graph digital display are disclosed. The method includes: first, acquiring an initial image of the instrument area and preprocessing the initial image; then, segmenting the preprocessed initial image to generate a digital display area and a bar graph display area; identifying the digital display area to obtain a first target parameter, and / or identifying the bar graph display area to obtain a second target parameter; finally, comparing the first target parameter and the second target parameter, and if the difference between the first target parameter and the second target parameter is within a preset error threshold range, then outputting the first target parameter and the second target parameter as valid parameters. This method solves the technical problem of unstable recognition results and low reliability caused by the limitations of a single recognition model in complex and ever-changing industrial environments, especially with varying lighting conditions, significantly improving the accuracy and environmental robustness of digital readings.

Description

Technical Field

[0001] This application relates to the field of image recognition technology, specifically to a method, system, device, and storage medium for verifying the readings of a bar graph digital display. Background Technology

[0002] With the continuous improvement of industrial automation and intelligence, the automatic, real-time, and highly reliable acquisition and verification of readings from various instruments (such as pressure gauges, thermometers, and flow meters) in production sites has become a key requirement for achieving equipment status monitoring, process optimization, and safe production. Barcode digital displays are widely used in industrial settings because they can simultaneously provide intuitive analog indications and accurate digital readings; however, their automatic identification has always faced significant challenges.

[0003] In related technologies, achieving automatic instrument reading mainly relies on two types of methods: one is based on traditional computer vision techniques, such as image preprocessing, edge detection, template matching, or projection analysis to locate and identify the position of the pointer or light bar, and then calculate the reading; the other is based on deep learning techniques, especially using object detection models (such as YOLO series, SSD) or character recognition models (such as CRNN, CTC network) to directly perform end-to-end recognition of the digital regions in the instrument image. These methods can achieve certain results in controlled or well-lit laboratory environments.

[0004] However, in real-world, complex industrial environments, lighting conditions are highly variable and uncontrollable, including alternating periods of strong and weak light, localized reflections, and shadows. This exposes significant shortcomings in existing technologies: traditional image processing methods typically have fixed algorithm features and parameters, resulting in poor robustness to environmental changes. Under lighting interference, feature extraction easily fails, leading to a sharp drop in recognition rate. While methods based on single deep learning models possess some feature learning capabilities, they are prone to false positives, false negatives, or character recognition errors under extreme lighting conditions where training data is insufficient. Furthermore, their outputs lack self-verification mechanisms, causing the entire system to fail upon error. More fundamentally, existing solutions often employ a single recognition path, whether recognizing only numbers or only light beams, failing to fully utilize the multiple information sources provided by the instrument itself to build a system with inherent fault tolerance and verification capabilities. Therefore, designing a highly reliable automatic reading solution for industrial instruments that can operate stably and accurately under complex lighting interference remains a pressing technical challenge in this field. Summary of the Invention

[0005] This application provides a method, apparatus, device, and storage medium for verifying the readings of a bar graph digital display, which can solve the technical problem in the prior art that the bar graph digital display readings cannot be stably and accurately identified under complex lighting interference.

[0006] In a first aspect, embodiments of this application provide a method for verifying the readings of a bar graph digital display, the method comprising:

[0007] Acquire an initial image of the instrument area and preprocess the initial image;

[0008] The preprocessed initial image is segmented to generate a digital display area and a light bar display area;

[0009] The digital display area is identified to obtain a first target parameter, and / or the light bar display area is identified to obtain a second target parameter; the identification of the digital display area to obtain the first target parameter includes: simultaneously inputting the image of the digital display area into a first recognition model and a second recognition model, and obtaining the recognition parameter values ​​of the two models; calculating the difference between the recognition parameter values ​​of the two models and a preset theoretical reference value, and selecting the recognition parameter value with the smallest error between the recognition parameter value of the two models and the theoretical reference value as the first target parameter;

[0010] The first target parameter is compared with the second target parameter. If the difference between the first target parameter and the second target parameter is within a preset error threshold range, then the first target parameter and the second target parameter are output as valid parameters.

[0011] In conjunction with the first aspect, in one embodiment, the step of simultaneously inputting the image of the digital display area into the first recognition model and the second recognition model, and obtaining the recognition parameter values ​​of the two models, includes:

[0012] The image of the digital display area is input into the trained YOLOv5s model and CRNN model respectively;

[0013] Obtain the character sequence after the YOLOv5s model recognizes a single character within the digital display area, and / or obtain the character sequence after the CRNN model recognizes a single character within the digital display area, and use the character sequence recognized by the YOLOv5s model and / or the CRNN model as the recognition parameter value.

[0014] In conjunction with the first aspect, in one embodiment, identifying the light bar display area to obtain the second target parameter includes:

[0015] The light column display area is subjected to image binarization processing;

[0016] Obtain the pixel height parameter of the light bar display area within the light bar display area after image binarization;

[0017] The actual height value of the light column is read based on the pixel height parameter of the light column display area.

[0018] In conjunction with the first aspect, in one embodiment, reading the actual height value of the light column based on the pixel height parameter of the light column display area includes:

[0019] The pixel height of the light column display area is divided into several equal parts according to a ratio, and the reading is taken according to the actual position of the light column in the light column display area according to the pre-divided equal parts.

[0020] In conjunction with the first aspect, in one embodiment, the step of performing image segmentation on the preprocessed initial image to generate a digital display area and a light bar display area includes:

[0021] A first coordinate is generated within the instrument area, and a first boundary line is generated based on the first coordinate; and / or, a second coordinate is generated within the instrument area, and a second boundary line is generated based on the second coordinate;

[0022] The area enclosed by the first boundary line is defined as the digital display area, and the area enclosed by the second boundary line is defined as the light bar display area.

[0023] In conjunction with the first aspect, in one embodiment, acquiring the initial image of the instrument area further includes:

[0024] Detect the brightness value of the instrument area;

[0025] Determine whether the preset image brightness value acquisition conditions are met. If not, acquire the brightness value that meets the preset image acquisition conditions.

[0026] In conjunction with the first aspect, in one embodiment, the preprocessing of the initial image includes:

[0027] The initial image of the instrument area is processed to grayscale and contrast is enhanced.

[0028] Secondly, embodiments of this application provide a reading verification system for a bar graph digital display, the system comprising:

[0029] An image acquisition module is used to acquire an initial image of the instrument area and preprocess the initial image;

[0030] The image segmentation module is used to segment the preprocessed initial image to generate a digital display area and a light bar display area;

[0031] A first image recognition module and a second image recognition module, wherein the first image recognition module is used to recognize the digital display area to obtain a first target parameter, and / or the second image recognition module is used to recognize the light bar display area to obtain a second target parameter; the recognition of the digital display area to obtain the first target parameter includes: simultaneously inputting the image of the digital display area into the first recognition model and the second recognition model, obtaining the recognition parameter values ​​of the two models, calculating the difference between the recognition parameter values ​​of the two models and a preset theoretical reference value, and selecting the recognition parameter value with the smallest error between the recognition parameter value of the two models and the theoretical reference value as the first target parameter;

[0032] The control module is used to compare the first target parameter with the second target parameter. If the difference between the first target parameter and the second target parameter is within a preset error threshold range, the first target parameter and the second target parameter are output as valid parameters.

[0033] In conjunction with the second aspect, in one embodiment, the image acquisition module further includes a light sensor, a fill light, and an industrial camera;

[0034] The optical sensor is used to detect the brightness value of the instrument area;

[0035] If the brightness value does not meet the preset image brightness value acquisition conditions, the supplementary light will provide supplementary lighting until the brightness meets the preset image acquisition conditions, and then the industrial camera will acquire the image.

[0036] Thirdly, embodiments of this application provide an electronic device, which includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the steps of the bar graph digital display reading verification method described in any of the above claims.

[0037] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the steps of the bar graph digital display reading verification method described in any of the above claims.

[0038] The beneficial effects of the technical solutions provided in this application include:

[0039] The method for verifying readings of a bar graph digital display disclosed in this application solves the technical problem in related technologies where a single recognition model suffers from unstable and unreliable recognition results due to its own limitations when facing complex and variable industrial environments (especially with varying lighting conditions). This method simultaneously inputs the image of the digital display area into a first recognition model and a second recognition model, and selects the optimal result as the first target parameter by comparing the error with the theoretical reference value. The solution constructs a dynamic and adaptive decision-making arbitration mechanism, effectively integrating the advantages of different models in terms of anti-interference or recognition accuracy, thereby significantly improving the accuracy and environmental robustness of digital readings.

[0040] Meanwhile, by comparing and verifying the first target parameter identified based on the digital display area with the second target parameter identified based on the light bar display area, and outputting both as valid parameters when the difference meets the requirements, this solves the technical problem in related technologies where a single identification path (identifying only numbers or only light bars) is used, where the system lacks an inherent fault tolerance and verification mechanism, leading to the failure of the entire verification process if an identification error occurs. This solution creatively utilizes the instrument's own dual display information sources to construct a dual-redundant verification path, achieving cross-verification and mutual backup of results, greatly enhancing the reliability and industrial availability of the entire reading verification system. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating one embodiment of the bar graph digital display reading verification method of this application;

[0042] Figure 2 This is a flowchart illustrating another embodiment of the bar graph digital display reading verification method of this application;

[0043] Figure 3 For this application Figure 1 A detailed flowchart illustrating the process of identifying the light bar display area to obtain the second target parameters in step S3;

[0044] Figure 4 This is a schematic diagram of the hardware structure of the electronic device involved in the embodiments of this application. Detailed Implementation

[0045] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0047] In a first aspect, embodiments of this application provide a method for verifying the readings of a bar graph digital display.

[0048] In one embodiment, reference is made to Figures 1-3 , Figure 1 This is a flowchart illustrating the first embodiment of the bar graph digital display reading verification method of this application. Figure 1 As shown, the reading verification method for the bar graph digital display includes:

[0049] Acquire the initial image of the instrument area and preprocess the initial image to make the initial graphic have the basis for processing. This can be understood as using image optimization software or algorithms to process the image's clarity, contrast, etc.

[0050] The preprocessed initial image is segmented to generate a digital display area and a light bar display area. This step mainly involves finding the key areas for subsequent identification based on the initial image. That is, the areas for displaying numbers and the areas for displaying light bars in the instrument image are extracted separately as the identification parts.

[0051] Then, the digital display area is identified to obtain a first target parameter, and / or the light bar display area is identified to obtain a second target parameter. It should be noted that the first target parameter refers to the numerical string displayed within the digital area, and the second target parameter refers to the display height range of the light bar on the light bar display area being converted into the corresponding numerical string. For the step of identifying the digital display area to obtain the first target parameter, the method further includes: simultaneously inputting the image of the digital display area into a first recognition model and a second recognition model, obtaining the recognition parameter values ​​of the two models, calculating the difference between the recognition parameter values ​​of the two models and a pre-set theoretical reference value, and selecting the recognition parameter value with the smallest error between the two models and the theoretical reference value as the first target parameter. This ensures that the reading of the first target parameter is verified through two different models, guaranteeing the accuracy of the first target parameter. It should be noted that the pre-set theoretical reference value is the basis for arbitrating the recognition parameter values ​​of the two models, and its source can be historical data, i.e., the last reading deemed reliable. The calculated value from the optical beam path is obtained by directly using the currently acquired second target parameter. This means that when digital recognition is interfered with, the system can use the results from the optical beam path to assist in arbitrating the digital path. Multi-frame filtering is performed by applying a moving average or median filter to the recognition results of the first N frames in a continuous video stream.

[0052] Finally, the first target parameter is compared with the second target parameter. If the difference between the first target parameter and the second target parameter is within the preset error threshold range, the first target parameter and the second target parameter are output as valid parameters.

[0053] In this embodiment, by simultaneously inputting the image of the digital display area into both the first and second recognition models, and using the error comparison with the theoretical reference value to select the optimal result as the first target parameter, the technical problem of unstable and unreliable recognition results caused by the limitations of a single recognition model in complex and ever-changing industrial environments (especially with varying lighting conditions) is solved. This solution constructs a dynamic and adaptive decision-making arbitration mechanism, effectively integrating the advantages of different models in terms of anti-interference capability and recognition accuracy, thereby significantly improving the accuracy and environmental robustness of digital readings.

[0054] Meanwhile, by comparing and verifying the first target parameter identified based on the digital display area with the second target parameter identified based on the light bar display area, and outputting both as valid parameters when the difference meets the requirements, this solves the technical problem in related technologies where a single identification path (identifying only numbers or only light bars) is used, where the system lacks an inherent fault tolerance and verification mechanism, leading to the failure of the entire verification process if an identification error occurs. This solution creatively utilizes the instrument's own dual display information sources to construct a dual-redundant verification path, achieving cross-verification and mutual backup of results, greatly enhancing the reliability and industrial availability of the entire reading verification system.

[0055] Furthermore, in one embodiment, simultaneously inputting the image of the digital display area into the first recognition model and the second recognition model, and obtaining the recognition parameter values ​​of the two models includes:

[0056] The images of the digital display area are input into the trained YOLOv5s and CRNN models, respectively. The YOLOv5s model is a character-level detection model specifically designed for detecting and recognizing individual characters. Its key feature lies in its training using a database of instrument digital samples that includes various lighting conditions such as simulated strong light, weak light, side light, and reflected light, allowing it to learn lighting-invariant features and possess strong resistance to lighting interference. The CRNN model is a high-precision character sequence recognition model, exhibiting extremely high recognition accuracy under good lighting conditions.

[0057] Obtain the character sequence after the YOLOv5s model recognizes a single character within the digit display area, and / or obtain the character sequence after the CRNN model recognizes a single character within the digit display area, and use the character sequence recognized by the YOLOv5s model and / or the CRNN model as the recognition parameter value.

[0058] In this embodiment, by inputting the image of the digital display area into a YOLOv5s character detection model specifically trained to resist illumination interference and a high-precision CRNN sequence recognition model, and combining the recognition results of the two, the technical problem in related technologies that a single character recognition model cannot simultaneously guarantee high accuracy and high robustness under drastic changes in ambient lighting is solved. This solution is not a simple stacking of models, but rather allows each model to leverage its specific advantages: the YOLOv5s model, due to its illumination invariance obtained from training on samples containing various extreme lighting conditions, ensures basic recognition capability and stability under poor lighting conditions; the CRNN model guarantees the upper limit of recognition accuracy under good lighting conditions. The two constitute a complementary collaborative recognition unit, enabling the system to adaptively cover various lighting scenarios from ideal to extreme conditions, thereby achieving stable and accurate digital recognition in complex industrial environments.

[0059] Furthermore, in one embodiment, identifying the light pillar display area to obtain the second target parameter includes: performing image binarization processing on the light pillar display area; obtaining the pixel height parameter of the light pillar display area within the light pillar display area after image binarization processing; and reading the actual height value of the light pillar based on the pixel height parameter of the light pillar display area.

[0060] In this embodiment, by performing image binarization processing on the light column display area and obtaining its pixel height to convert it into an actual value, the technical problem in related technologies is that the image quality of the light column area is poor (such as blurred boundaries, broken areas, or noise) due to uneven lighting, reflection, or poor contrast, which leads to inaccurate height measurement and large errors in the final reading conversion. This solution, through targeted image binarization, stably and clearly separates the light column area from the background, thereby reliably extracting the key parameter of pixel height used for quantization calculation. This lays a solid foundation for accurately converting image information into actual instrument readings in the physical world, and improves the anti-interference capability and overall accuracy of the light column recognition path.

[0061] Furthermore, in one embodiment, reading the actual height value of the light column based on the pixel height parameter of the light column display area includes:

[0062] The pixel height of the light bar display area is divided into several equal parts according to a ratio. The reading is taken according to the actual position of the light bar in the light bar display area and according to the pre-divided equal parts.

[0063] In this embodiment, by pre-dividing the pixel height of the light column display area into several equal measurement portions according to a predetermined ratio, and taking readings based on the light column position within this measurement system, the technical problem in related technologies where continuous proportional conversion based directly on the original pixel height leads to the accumulation of system errors caused by factors such as image perspective distortion, camera installation angle deviation, or sensor nonlinearity, thus affecting the accuracy and consistency of the final reading. This solution introduces a "virtual scale" mechanism, transforming continuous analog quantity measurement into a mapping reading based on a discrete scale. This not only mimics the logic of manual reading and improves the intuitiveness and reliability of the results, but also effectively suppresses the amplified impact of local pixel errors on the overall results through equal division, thereby significantly improving the accuracy, stability, and adaptability to different installation conditions of the light column parameter recognition.

[0064] Furthermore, in one embodiment, image segmentation is performed on the preprocessed initial image to generate a digital display area and a light bar display area, including:

[0065] A first coordinate is generated within the instrument area, and a first boundary line is generated based on the first coordinate; and / or, a second coordinate is generated within the instrument area, and a second boundary line is generated based on the second coordinate; the area enclosed by the first boundary line is determined as the digital display area, and the area enclosed by the second boundary line is determined as the light bar display area.

[0066] In this embodiment, the display areas for numbers and light bars are precisely divided by dynamically generating coordinates and boundary lines within the instrument area. This solves the technical problems in related technologies, such as poor adaptability to differences in instrument models, installation angles, and perspective distortions due to reliance on fixed templates or single feature segmentation, as well as inaccurate area positioning and blurred boundaries under uneven lighting and cluttered backgrounds. This solution achieves adaptive area detection and extraction, robustly handling different instrument structures and imaging conditions, providing stable and accurate input for subsequent independent recognition of numbers and light bars, and ensuring the reliability of the entire recognition process from the source.

[0067] Furthermore, in one embodiment, acquiring the initial image of the instrument area further includes:

[0068] The system detects the brightness value of the instrument area; then it determines whether the preset image brightness value acquisition conditions are met. If not, it acquires the brightness value that meets the preset image acquisition conditions. Specifically, it can control the light source component to start and provide supplementary lighting, and the brightness value can be adjusted according to the desired optimal brightness value.

[0069] Furthermore, in one embodiment, preprocessing the initial image includes:

[0070] The initial image of the instrument area is converted to grayscale and its contrast is enhanced.

[0071] Therefore, grayscale conversion can be handled using the averaging method. The averaging method can be calculated using the following formula: Gray = (R + G + B) / 3;

[0072] Where R, G, and B represent the brightness values ​​of a specific pixel in the red, green, and blue channels of the image, respectively, and Gray is the calculated grayscale value of that pixel. The calculation method is to add the brightness values ​​of the red, green, and blue channels of the same pixel and then take the arithmetic mean.

[0073] Secondly, embodiments of this application also provide a reading verification system for a bar graph digital display.

[0074] In one embodiment, the system includes: an image acquisition module, an image segmentation module, a first image recognition module, a second image recognition module, and a control module;

[0075] The system includes an image acquisition module for acquiring an initial image of the instrument area and preprocessing the initial image; an image segmentation module for segmenting the preprocessed initial image to generate a digital display area and a light bar display area; a first image recognition module for recognizing the digital display area to obtain a first target parameter, and / or a second image recognition module for recognizing the light bar display area to obtain a second target parameter; recognizing the digital display area to obtain the first target parameter includes: simultaneously inputting the image of the digital display area into a first recognition model and a second recognition model, obtaining the recognition parameter values ​​of the two models, calculating the difference between the recognition parameter values ​​of the two models and a preset theoretical reference value, and selecting the recognition parameter value with the smallest error between the recognition parameter value of the two models and the theoretical reference value as the first target parameter; and a control module for comparing the first target parameter with the second target parameter, and if the difference between the first target parameter and the second target parameter is within a preset error threshold range, then outputting the first target parameter and the second target parameter as valid parameters.

[0076] In this embodiment, a dedicated system hardware architecture comprising image acquisition, image segmentation, dual-channel image recognition, and control modules is constructed. The control module performs collaborative comparison and verification of the results from the two recognition channels. This solves the technical problems inherent in related technologies that use general or loosely combined algorithm modules for instrument recognition. These problems include low system integration, lack of unified scheduling and coordination among processing units, and inability to achieve real-time reliable closed-loop verification, leading to insufficient overall reliability and difficulty in meeting the requirements for continuous and stable industrial-grade operation. This solution solidifies the algorithm flow into clearly defined physical and logical modules, enabling the dual-redundant recognition and verification mechanism to operate efficiently and stably at the system level. This ensures end-to-end controllability and reliability from image acquisition to result output, significantly improving the productization level and engineering application reliability of the entire verification device.

[0077] Further, in one embodiment, the first image recognition module includes a YOLOv5s model, and the second image recognition module includes a CRNN model; the first image recognition module and the second image recognition module are used to input the image of the digital display area into the trained YOLOv5s model and CRNN model respectively; and obtain the character sequence after the YOLOv5s model recognizes a single character in the digital display area, and / or obtain the character sequence after the CRNN model recognizes a single character in the digital display area, and use the character sequence recognized by the YOLOv5s model and / or CRNN model as the recognition parameter value.

[0078] Furthermore, in one embodiment, the second image recognition module is also used to: perform image binarization processing on the light column display area; obtain the pixel height parameters of the light column display area within the light column display area after image binarization processing; and read the actual height value of the light column based on the pixel height parameters of the light column display area.

[0079] Furthermore, in one embodiment, the second image recognition module is also used to: take readings according to a pre-divided measurement division based on the actual position of the light column within the light column display area. Specifically, the pixel height of the light column display area is divided into several measurement divisions according to a ratio.

[0080] Furthermore, in one embodiment, the image segmentation module is also used to generate a first coordinate within the instrument area, and generate a first boundary line based on the first coordinate; and / or, generate a second coordinate within the instrument area, and generate a second boundary line based on the second coordinate; the area enclosed by the first boundary line is determined as a digital display area, and the area enclosed by the second boundary line is determined as a light bar display area.

[0081] The image acquisition module is also used to detect the brightness value of the instrument area and determine whether it meets the preset image brightness value acquisition conditions. If not, it acquires the brightness value that meets the preset image acquisition conditions. Additionally, it performs grayscale processing and contrast enhancement processing on the initial image of the instrument area.

[0082] The functions of each module in the above-mentioned bar graph digital display reading verification system correspond to the steps in the above-mentioned bar graph digital display reading verification method embodiment, and their functions and implementation processes will not be described in detail here.

[0083] Thirdly, embodiments of this application provide an electronic device, which may be a personal computer (PC), a laptop computer, a server, or other device with data processing capabilities.

[0084] Reference Figure 4 , Figure 4 This is a schematic diagram of the hardware structure of the electronic device involved in the embodiments of this application. In the embodiments of this application, the electronic device may include a processor, a memory, a communication interface, and a communication bus.

[0085] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0086] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces used for interconnecting devices within the bar graph digital display reading verification system, as well as interfaces used for interconnecting the system with other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0087] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0088] The processor can be a general-purpose processor, which can call the bar graph digital display reading verification program stored in the memory and execute the bar graph digital display reading verification method provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the bar graph digital display reading verification program is called can refer to the various embodiments of the bar graph digital display reading verification method of this application, and will not be repeated here.

[0089] Those skilled in the art will understand that Figure 4 The hardware structure shown does not constitute a limitation of this application and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0090] Fourthly, embodiments of this application also provide a computer-readable storage medium.

[0091] The present application has a computer program stored on a computer-readable storage medium, wherein when the computer program is executed by a processor, it implements the steps of the above-described bar graph digital display reading verification method.

[0092] The method implemented when the computer program is executed can be referred to in various embodiments of the bar graph digital display reading verification method of this application, and will not be repeated here.

[0093] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0094] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0095] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0096] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0097] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0098] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0099] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for verifying the readings of a bar graph digital display, characterized in that, The method includes: Acquire an initial image of the instrument area and preprocess the initial image; The preprocessed initial image is segmented to generate a digital display area and a light bar display area; The digital display area is identified to obtain a first target parameter, and / or the light bar display area is identified to obtain a second target parameter; the identification of the digital display area to obtain the first target parameter includes: simultaneously inputting the image of the digital display area into a first recognition model and a second recognition model, and obtaining the recognition parameter values ​​of the two models; calculating the difference between the recognition parameter values ​​of the two models and a preset theoretical reference value, and selecting the recognition parameter value with the smallest error between the recognition parameter value of the two models and the theoretical reference value as the first target parameter; The first target parameter is compared with the second target parameter. If the difference between the first target parameter and the second target parameter is within a preset error threshold range, then the first target parameter and the second target parameter are output as valid parameters.

2. The method for verifying the readings of a bar graph digital display according to claim 1, characterized in that, The step of simultaneously inputting the image of the digital display area into the first recognition model and the second recognition model, and obtaining the recognition parameter values ​​of the two models, includes: The image of the digital display area is input into the trained YOLOv5s model and CRNN model respectively; Obtain the character sequence after the YOLOv5s model recognizes a single character within the digital display area, and / or obtain the character sequence after the CRNN model recognizes a single character within the digital display area, and use the character sequence recognized by the YOLOv5s model and / or the CRNN model as the recognition parameter value.

3. The method for verifying the readings of a bar graph digital display according to claim 2, characterized in that, The step of identifying the light bar display area to obtain the second target parameter includes: The light column display area is subjected to image binarization processing; Obtain the pixel height parameter of the light bar display area within the light bar display area after image binarization; The actual height value of the light column is read based on the pixel height parameter of the light column display area.

4. The method for verifying the readings of a bar graph digital display according to claim 3, characterized in that, The step of reading the actual height value of the light column based on the pixel height parameter of the light column display area includes: The pixel height of the light column display area is divided into several equal parts according to a ratio, and the reading is taken according to the actual position of the light column in the light column display area according to the pre-divided equal parts.

5. The method for verifying the readings of a bar graph digital display according to any one of claims 1-4, characterized in that, The step of segmenting the preprocessed initial image to generate a digital display area and a light bar display area includes: A first coordinate is generated within the instrument area, and a first boundary line is generated based on the first coordinate; and / or, a second coordinate is generated within the instrument area, and a second boundary line is generated based on the second coordinate; The area enclosed by the first boundary line is defined as the digital display area, and the area enclosed by the second boundary line is defined as the light bar display area.

6. The method for verifying the readings of a bar graph digital display according to claim 1, characterized in that, The initial image acquisition of the instrument area also includes: Detect the brightness value of the instrument area; Determine whether the preset image brightness value acquisition conditions are met. If not, acquire the brightness value that meets the preset image acquisition conditions.

7. The method for verifying the readings of a bar graph digital display according to claim 1, characterized in that, The preprocessing of the initial image includes: The initial image of the instrument area is processed to grayscale and contrast is enhanced.

8. A reading verification system for a bar graph digital display, characterized in that, The system includes: An image acquisition module is used to acquire an initial image of the instrument area and preprocess the initial image; The image segmentation module is used to segment the preprocessed initial image to generate a digital display area and a light bar display area; A first image recognition module and a second image recognition module, wherein the first image recognition module is used to recognize the digital display area to obtain a first target parameter, and / or the second image recognition module is used to recognize the light bar display area to obtain a second target parameter; the recognition of the digital display area to obtain the first target parameter includes: simultaneously inputting the image of the digital display area into the first recognition model and the second recognition model, obtaining the recognition parameter values ​​of the two models, calculating the difference between the recognition parameter values ​​of the two models and a preset theoretical reference value, and selecting the recognition parameter value with the smallest error between the recognition parameter value of the two models and the theoretical reference value as the first target parameter; The control module is used to compare the first target parameter with the second target parameter. If the difference between the first target parameter and the second target parameter is within a preset error threshold range, the first target parameter and the second target parameter are output as valid parameters.

9. An electronic device, characterized in that, The electronic device includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the steps of the bar graph digital display reading verification method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, it implements the steps of the bar graph digital display reading verification method as described in any one of claims 1 to 7.

Images