Instrument intelligent analysis method and system based on video images

Abstract

The invention particularly relates to an instrument intelligent analysis method and system based on a video image, and the method comprises the steps: carrying out the preprocessing based on the instrument video image, and carrying out the illumination contrast parameter adjustment, thereby improving the visual effect and definition of the image. Then, image segmentation operation is carried out based on the preprocessed instrument image, the image segmentation operation is used for accurately positioning a feature region needing to be identified, and the accuracy of the key features of the instrument is improved; and further, instrument feature extraction is carried out on the instrument region segmentation map by utilizing an instrument feature recognition model, when the instrument type is a pointer type instrument, pointer positioning refinement and scale calibration are carried out based on the key features of the instrument to obtain calibration data, and then instrument reading calculation is carried out based on the calibration data to determine the instrument reading. And an instrument data identification mode combining a classic image processing technology and convolutional neural network model features is integrated, so that the accuracy of instrument data reading is improved.

Description

Technical Field

[0001] The present application relates to the technical field of instrument detection, and in particular to an instrument intelligent analysis method and system based on video images. Background Art

[0002] In many fields such as industrial production, transportation, and environmental monitoring, instruments are key equipment parameter indicator tools. The accuracy and real-time reading of instruments are crucial to ensuring the stable operation of the system, preventing safety accidents, and optimizing production efficiency. Traditional instrument readings rely on manual observation and recording, which is not only time-consuming and labor-intensive, but also prone to inaccurate data due to human factors, which in turn affects subsequent data analysis and decision-making.

[0003] With the rapid development of computer vision technology and deep learning algorithms, intelligent instrument analysis methods have emerged, aiming to achieve fast and accurate recognition and recording of instrument readings through automated and intelligent means. However, related intelligent instrument analysis methods are only applicable to specific types of instruments in simple scenarios. When dealing with complex scenarios such as lighting changes, the recognition accuracy needs to be improved, and it is difficult to meet the needs of high-precision instrument data collection.

[0004] Therefore, how to improve the recognition accuracy of the instrument is an urgent problem to be solved by those skilled in the art. Summary of the invention

[0005] The purpose of this application is to provide an instrument intelligent analysis method and system based on video images to solve at least one of the above technical problems.

[0006] The above invention objectives of the present application are achieved through the following technical solutions: In a first aspect, the present application provides a video image-based instrument intelligent analysis method, which adopts the following technical solution: An instrument intelligent analysis method based on video images, comprising: Acquire an instrument video image, and perform preprocessing based on the instrument video image to obtain a preprocessed instrument image, wherein the preprocessing includes: filtering and denoising, grayscale conversion, and illumination contrast parameter adjustment; Performing an image segmentation operation based on the preprocessed instrument image to obtain an instrument region segmentation map, wherein the instrument region segmentation map is used to divide different functional areas in the instrument into zones; Acquire an instrument feature recognition model, and use the instrument feature recognition model to extract instrument features from the instrument region segmentation map to determine instrument key features, wherein the instrument key features include: pointer position features and scale distribution features; When the instrument type is a pointer-type instrument, the pointer positioning is refined and the scale is calibrated based on the key features of the instrument to obtain calibration data, and the instrument reading is calculated based on the calibration data to determine the instrument reading.

[0007] By adopting the above technical solution, preprocessing is performed based on the instrument video image to obtain a preprocessed instrument image. The preprocessing includes adjusting the illumination contrast parameters. The illumination contrast parameters are adjusted on the instrument video image so that the brightness and contrast of the processed instrument image reach a suitable range, thereby enhancing the visual effect and clarity of the image. Then, an image segmentation operation is performed based on the preprocessed instrument image to obtain an instrument area segmentation map. The image segmentation operation is used to accurately locate the feature area to be identified and improve the accuracy of the key features of the instrument. Furthermore, the instrument feature recognition model is used to extract instrument features from the instrument area segmentation map to determine the key features of the instrument. After that, the pointer positioning is refined and the scale is calibrated based on the key features of the instrument to obtain calibration data, and the instrument reading is calculated based on the calibration data to determine the instrument reading. The instrument data recognition method that combines classical image processing technology and convolutional neural network model feature extraction for data recognition improves the accuracy of instrument data reading.

[0008] In a preferred example, the present application may be further configured as follows: after calculating the meter reading based on the calibration data and determining the meter reading, the method further includes: Storing the meter readings in a data monitoring database, wherein the data monitoring database is used to store meter readings corresponding to different nodes of each meter; Extracting an instrument reading sequence corresponding to a target instrument from the data monitoring database, wherein the target instrument is any instrument stored in the data monitoring database, and the instrument reading sequence consists of a plurality of instrument readings arranged in chronological order; Reading prediction is performed based on the instrument reading sequence to determine prediction data, and preventive evaluation is performed based on the prediction data to obtain preventive maintenance information.

[0009] In a preferred example, the present application may be further configured as follows: before acquiring the instrument video image, the present application may further include: Acquiring illumination information collected by a light sensor, wherein the light sensor is deployed at a preset position of the instrument and is used to detect the light intensity and illumination conditions of the instrument's surrounding environment in real time; Based on the illumination information, working parameter analysis is performed to determine working parameters corresponding to the image acquisition device, and the working parameters are sent to the image acquisition device to control the image acquisition device to perform image acquisition according to the working parameters, wherein the working parameters include: exposure time, shutter speed, white balance speed and ISO value.

[0010] In a preferred example, the present application may be further configured as follows: after obtaining the illumination information collected by the light sensor, the method further includes: Performing a lighting condition evaluation based on the lighting information to determine a lighting condition evaluation result, wherein the lighting condition evaluation result includes: satisfying image acquisition requirements and not satisfying image acquisition requirements; When the illumination condition evaluation result is that the image acquisition requirement is not met, a fill light analysis is performed based on the illumination information to determine a fill light operation, wherein the fill light operation includes: adding a light source and adjusting a light direction.

[0011] In a preferred example, the present application may be further configured as follows: after calculating the meter reading based on the calibration data and determining the meter reading, the method further includes: Acquire instrument environment data, perform environmental interference analysis based on the instrument environment data, and determine an environmental interference analysis result, wherein the environmental interference analysis result includes: interference exists and no interference exists; When the environmental interference analysis result indicates that interference exists, a corresponding target sensor is determined based on the type of the instrument reading, and industrial data collected by the target sensor is obtained, and the instrument reading is updated based on the industrial data.

[0012] In a preferred example, the present application can be further configured as follows: the pointer positioning refinement and scale calibration based on the key features of the instrument to obtain calibration data includes: Using an edge detection algorithm to refine the pointer position feature in the key feature of the instrument to determine the position of the pointer tip; Using template matching and regression analysis to calibrate the scale distribution features in the key features of the instrument to determine scale distribution information; The calibration data is obtained by combining the pointer tip position and the scale distribution information.

[0013] In the second aspect, the present application provides an instrument intelligent analysis system based on video images, which adopts the following technical solutions: A preprocessing module, used for acquiring an instrument video image, and performing preprocessing based on the instrument video image to obtain a preprocessed instrument image, wherein the preprocessing includes: filtering and denoising, grayscale conversion, and illumination contrast parameter adjustment; An image segmentation module, used for performing an image segmentation operation based on the pre-processed instrument image to obtain an instrument region segmentation map, wherein the instrument region segmentation map is used for dividing different functional regions in the instrument; A feature extraction module is used to obtain an instrument feature recognition model, and use the instrument feature recognition model to extract instrument features from the instrument region segmentation map to determine instrument key features, wherein the instrument key features include: pointer position features and scale distribution features; The refinement calibration module is used to perform pointer positioning refinement and scale calibration based on the key features of the instrument when the instrument type is a pointer instrument, obtain calibration data, and calculate the instrument reading based on the calibration data to determine the instrument reading.

[0014] In a third aspect, the present application provides an electronic device, which adopts the following technical solution: at least one processor; Memory; At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the above-mentioned video image-based instrument intelligent analysis method.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program, which, when executed in a computer, causes the computer to execute the above-mentioned video image-based instrument intelligent analysis method.

[0016] In summary, the present application includes at least one of the following beneficial technical effects: Preprocessing is performed based on the instrument video image to obtain a preprocessed instrument image. The preprocessing includes adjusting the illumination contrast parameters. The illumination contrast parameters are adjusted on the instrument video image so that the brightness and contrast of the processed instrument image reach a suitable range, thereby enhancing the visual effect and clarity of the image. Then, an image segmentation operation is performed based on the preprocessed instrument image to obtain an instrument region segmentation map. The image segmentation operation is used to accurately locate the feature area to be identified and improve the accuracy of the key features of the instrument. Furthermore, the instrument feature recognition model is used to extract instrument features from the instrument region segmentation map to determine the key features of the instrument. After that, the pointer positioning is refined and the scale is calibrated based on the key features of the instrument to obtain calibration data, and the instrument reading is calculated based on the calibration data to determine the instrument reading. The instrument data recognition method that combines the classical image processing technology and the convolutional neural network model feature extraction for data recognition improves the accuracy of instrument data reading.

[0017] The instrument reading sequence corresponding to the target instrument is extracted from the data monitoring database. Then, the reading prediction is performed based on the instrument reading sequence, the prediction data is determined, and the preventive evaluation is performed based on the prediction data to obtain preventive maintenance information. The instrument reading is usually closely related to the operating status of the equipment. Therefore, performing reading prediction and preventive evaluation operations can help to timely discover and deal with potential problems, reduce the occurrence of equipment failures or accidents to a certain extent, and thus improve the reliability of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flowchart of an instrument intelligent analysis method based on video images according to one embodiment of the present application; Figure 2 It is a structural schematic diagram of an instrument intelligent analysis system based on video images according to one embodiment of the present application; Figure 3 It is a structural schematic diagram of an electronic device according to one embodiment of the present application. DETAILED DESCRIPTION

[0019] The following combination Figures 1 to 3 This application is described in further detail.

[0020] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, a person skilled in the art may make non-creative modifications to the present embodiment as needed, but such modifications are protected by the patent law as long as they are within the scope of the present application.

[0021] In order to make the purpose, technical scheme and advantages of the embodiment of the present application clearer, the technical scheme in the embodiment of the present application will be clearly and completely described in conjunction with the drawings in the embodiment of the present application. Obviously, the described embodiment is a part of the embodiment of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application. It should be noted that in the optional embodiments of the present application, the object information and other related data involved, when the embodiments in the present application are applied to specific products or technologies, need to obtain the permission or consent of the object, and the collection, use and processing of the relevant data need to comply with the relevant laws, regulations and standards of the relevant countries and regions. In other words, if the data related to the object is involved in the embodiment of the present application, it needs to be obtained through the authorization and consent of the object, the authorization and consent of the relevant departments, and in accordance with the relevant laws, regulations and standards of the country and region. If personal information is involved in the embodiment, the acquisition of all personal information needs to obtain the consent of the individual. If sensitive information is involved, the separate consent of the information subject needs to be obtained, and the embodiment also needs to be implemented with the authorization and consent of the object.

[0022] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0023] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0024] The embodiment of the present application provides an instrument intelligent analysis method based on video images, which is executed by an electronic device, which can be a server or a terminal device, wherein the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present application. Figure 1 As shown, the method includes step S101, step S102, step S103 and step S104, wherein: Step S101: Acquire an instrument video image, and perform preprocessing based on the instrument video image to obtain a preprocessed instrument image, wherein the preprocessing includes: filtering and denoising, grayscale conversion, and illumination contrast parameter adjustment.

[0025] For the embodiments of the present application, in order to solve the defect that the intelligent analysis method of the instrument in the related art cannot realize high-precision data reading in the scene with complex lighting changes, the lighting contrast parameter of the instrument video image is adjusted so that the brightness and contrast of the processed instrument image reach the appropriate range, thereby enhancing the visual effect and clarity of the image. After that, the pre-processed instrument image is subjected to image segmentation operation, so that the instrument feature recognition model can accurately locate the feature area to be identified when extracting the instrument features from the instrument area segmentation map, thereby improving the accuracy of the key features of the instrument. At the same time, the key features of the instrument extracted by the convolutional neural network model are subjected to pointer positioning refinement and scale calibration by using classical image processing technology. The instrument data recognition method combining the two methods improves the accuracy of instrument data reading.

[0026] Specifically, a high-definition image acquisition device is deployed near the meter. Preferably, the image acquisition device is a high-performance camera, and the deployment position should ensure that the camera can clearly capture the image of the meter. Therefore, the image acquisition device can record the dial video of the meter and ensure that the video quality is clear and stable. The electronic device is connected to the image acquisition device deployed near the meter by wireless means, so that the electronic device can obtain the meter video image from the image acquisition device.

[0027] In order to improve the clarity and quality of the instrument image and improve the efficiency of subsequent instrument data reading, preprocessing is performed based on the instrument video image to obtain a preprocessed instrument image. The preprocessing includes but is not limited to: filtering denoising, grayscale conversion and illumination contrast parameter adjustment. Through filtering denoising, various noises generated by the instrument video image during the acquisition and transmission process are effectively reduced or eliminated, the clarity and quality of the image are improved, and more accurate input is provided for subsequent processing. Since color images contain multiple color channels, the amount of data is large and the processing is relatively complex, while grayscale images only contain brightness information, the amount of data is small, and the processing speed is faster. Therefore, performing grayscale conversion can reduce the computational complexity of subsequent processing and improve processing efficiency. At the same time, the lighting conditions of the instrument video image may vary depending on the environment, resulting in uneven brightness and contrast of the image, which affects the visual effect of the image and the accuracy of subsequent processing to a certain extent. Therefore, by adjusting the illumination contrast parameters, the brightness and contrast of the image are within a suitable range, the visual effect and clarity of the image are enhanced, and more accurate input is provided for subsequent processing.

[0028] Step S102: performing an image segmentation operation based on the pre-processed instrument image to obtain an instrument region segmentation map, wherein the instrument region segmentation map is used to divide different functional regions in the instrument into zones.

[0029] For the embodiment of the present application, based on the pre-processed instrument image, an image segmentation method is selected, and the image segmentation method includes but is not limited to: a threshold-based image segmentation method, a region-based image segmentation method, an edge-based image segmentation method, and a machine learning-based image segmentation method. Different image segmentation methods have different characteristics, and the characteristics of the instrument image they are suitable for will also be different. Therefore, the electronic device pre-stores the correspondence between the characteristics of the instrument image and the image segmentation method, and the correspondence is used to match the most suitable image segmentation method for the pre-processed instrument image. For example, the threshold-based image segmentation method is suitable for situations where the grayscale value difference between the target and background in the instrument image is obvious; the region-based image segmentation method is suitable for situations where the target area and the background area in the instrument image have obvious differences in color, texture, etc.; the edge-based image segmentation method is suitable for situations where the edge between the target area and the background area in the instrument image is clear and obvious; the machine learning-based image segmentation method is suitable for situations where the instrument image is complex and the difference between the target area and the background area is not obvious.

[0030] Then, based on the selected image segmentation method, the pre-processed instrument image is subjected to image segmentation operation to obtain an instrument region segmentation map, in which different functional areas are divided into different blocks, such as speedometer area, fuel gauge area, tachometer area, temperature gauge area, pressure gauge area, etc. By performing the image segmentation operation, not only can the feature area to be identified be accurately located, other irrelevant information can be ignored to reduce the amount of calculation for subsequent feature extraction and improve the accuracy of the key features of the instrument; it can also decompose the complex instrument image into simpler sub-problems, reducing the complexity of subsequent instrument feature extraction.

[0031] Step S103: Acquire an instrument feature recognition model, use the instrument feature recognition model to extract instrument features from the instrument region segmentation map, and determine instrument key features, wherein the instrument key features include: pointer position features and scale distribution features.

[0032] For the embodiment of the present application, the electronic device pre-stores the instrument feature recognition model, which is obtained by training the convolutional neural network model with a large number of training samples, wherein the convolutional neural network model includes: input layer, convolution layer, pooling layer, fully connected layer and other structures to achieve effective feature extraction of the instrument image, and the training samples are labeled instrument readings and corresponding instrument area segmentation maps. In the process of training the convolutional neural network model with training samples, it is necessary to continuously optimize the parameters of the model to improve its ability to recognize the key features of the instrument. At the same time, during the training process, it is also necessary to regularly evaluate the model to check its recognition performance; if the performance is not good, the model needs to be further optimized, such as adjusting the network structure, increasing training data, etc.; when the recognition performance meets the requirements, the trained model is used as the instrument feature recognition model.

[0033] Then, the instrument area segmentation map to be identified is input into the instrument feature recognition model, which extracts instrument features from each functional area block in the instrument area segmentation map to determine the instrument key features, which include: pointer position features and scale distribution features, where the pointer position features are used to characterize the specific position of the pointer in the instrument, and the scale distribution features are used to characterize the numerical range and accuracy of the scale on the instrument. Of course, for digital display instruments, instrument key features can also include: displayed numbers and symbols. When the instrument type is a digital display instrument, the displayed numbers in the instrument key features are combined with the corresponding symbols to obtain a complete instrument reading.

[0034] Step S104: When the instrument type is a pointer instrument, the pointer positioning is refined and the scale is calibrated based on the key features of the instrument to obtain calibration data, and the instrument reading is calculated based on the calibration data to determine the instrument reading.

[0035] For the embodiment of the present application, in the actual instrument feature extraction process, the instrument image may be affected by factors such as illumination changes, noise interference, and camera angles, which may cause certain errors in the pointer position and scale distribution extracted by the instrument feature recognition model, thereby affecting the accuracy of subsequent instrument readings. In order to further improve the accuracy of instrument reading recognition, the pointer positioning is refined and the scale is calibrated based on the key features of the instrument to obtain calibration data, that is, the pointer position is refined using image processing methods such as edge detection and morphological processing, and the scale is calibrated using methods such as template matching and regression analysis, so that the calibration data can meet the high precision and robustness requirements in practical applications. Furthermore, based on the pointer tip position and scale distribution information in the calibration data, the angle of the pointer tip relative to a certain reference point (for example, the zero scale point of the instrument dial) is calculated, and then, according to the scale range and accuracy in the scale distribution information of the instrument, the calculated pointer angle is converted into the corresponding instrument reading.

[0036] It can be seen that in the embodiment of the present application, preprocessing is performed based on the instrument video image to obtain a preprocessed instrument image. The preprocessing includes adjusting the illumination contrast parameters. The illumination contrast parameters are adjusted on the instrument video image so that the brightness and contrast of the processed instrument image reach a suitable range, thereby enhancing the visual effect and clarity of the image. Then, an image segmentation operation is performed based on the preprocessed instrument image to obtain an instrument area segmentation map. The image segmentation operation is used to accurately locate the feature area to be identified, thereby improving the accuracy of the key features of the instrument. Furthermore, the instrument feature recognition model is used to extract instrument features from the instrument area segmentation map to determine the key features of the instrument. After that, the pointer positioning is refined and the scale is calibrated based on the key features of the instrument to obtain calibration data, and the instrument reading is calculated based on the calibration data to determine the instrument reading. The instrument data recognition method that combines classical image processing technology and convolutional neural network model feature extraction for data recognition improves the accuracy of instrument data reading.

[0037] Furthermore, in order to timely discover and handle potential problems, reduce the occurrence of equipment failures or accidents, and improve the reliability of equipment, in the embodiment of the present application, after calculating the instrument reading based on the calibration data and determining the instrument reading, it also includes: The meter readings are stored in a data monitoring database, wherein the data monitoring database is used to store the meter readings corresponding to different nodes of each meter; Extracting an instrument reading sequence corresponding to a target instrument from a data monitoring database, wherein the target instrument is any instrument stored in the data monitoring database, and the instrument reading sequence consists of a plurality of instrument readings arranged in chronological order; Reading prediction is performed based on the instrument reading sequence to determine the prediction data, and preventive evaluation is performed based on the prediction data to obtain preventive maintenance information.

[0038] For the embodiment of the present application, the instrument reading is usually closely related to the running state of the equipment. By storing the instrument reading in the data monitoring database to realize the centralized management of the data, it is not only helpful for the subsequent data analysis and processing, but also convenient for the tracing and query of historical data. Therefore, the instrument reading is formatted so that the instrument reading is converted into a format suitable for storage, such as a number, a string or a specific data structure, and necessary metadata is added, such as a timestamp, an instrument ID, a reading type, etc., for subsequent data query and analysis. Then, the instrument reading sequence corresponding to the target instrument is extracted from the data monitoring database. The target instrument is any instrument stored in the data monitoring database. The instrument reading sequence consists of a plurality of instrument readings arranged in chronological order. The instrument reading sequence can reflect the changing trend of the readings in the target instrument. Then, the reading prediction is performed based on the instrument reading sequence, the prediction data is determined, and the preventive evaluation is performed based on the prediction data to obtain the preventive maintenance information. The instrument reading is usually closely related to the running state of the equipment. Therefore, performing the reading prediction and preventive evaluation operations can help to timely discover and deal with potential problems, reduce the occurrence of equipment failures or accidents to a certain extent, and thus improve the reliability of the equipment.

[0039] There are many specific implementation methods for reading prediction, which are not limited in the embodiments of the present application. In one feasible method, analyze the instrument reading sequence, identify its long-term trend (e.g., linear growth, exponential growth, etc.), and use the identified trend to predict future readings; check whether there are periodic changes in the instrument reading sequence (e.g., daily, weekly, and monthly periodic fluctuations), and predict future instrument readings based on periodic laws. There are many specific implementation methods for preventive evaluation, which are not limited in the embodiments of the present application. In one feasible method, the electronic device stores evaluation standards corresponding to each type of instrument. The evaluation standards are corresponding standards formulated according to the type, purpose, and working environment of the instrument. The standards include but are not limited to: reading range, rate of change, abnormal mode, etc. Then, according to the evaluation standards corresponding to the instrument, determine the reading threshold, which is used to determine whether the predicted reading is within the normal range and whether preventive maintenance measures need to be taken. Furthermore, the predicted data is compared and analyzed with the set reading threshold to determine whether the instrument reading may exceed the normal range or be abnormal. When the analysis determines that the instrument reading may exceed the normal range, the preventive maintenance information is determined based on the dimension of the abnormal instrument reading. The preventive maintenance information includes but is not limited to: maintenance time, maintenance content, required materials, and maintenance personnel. The maintenance content can be maintenance of the instrument or maintenance of the equipment monitored by the instrument. The specific maintenance object can be selected according to the actual situation, and the embodiments of the present application will no longer limit it.

[0040] It can be seen that in the embodiment of the present application, the instrument reading sequence corresponding to the target instrument is extracted from the data monitoring database. Then, the reading prediction is performed based on the instrument reading sequence, the prediction data is determined, and the preventive evaluation is performed based on the prediction data to obtain preventive maintenance information. The instrument reading is usually closely related to the operating status of the equipment. Therefore, performing reading prediction and preventive evaluation operations helps to timely discover and deal with potential problems, reduce the occurrence of equipment failures or accidents to a certain extent, and thus improve the reliability of the equipment.

[0041] Furthermore, in order to ensure that clear instrument images can be captured in various complex lighting environments and improve the recognition accuracy of subsequent instrument readings, in the embodiment of the present application, before acquiring the instrument video image, the following steps are also included: Acquire light information collected by a light sensor, wherein the light sensor is deployed at a preset position of the instrument and is used to detect the light intensity and light conditions of the instrument's surrounding environment in real time; Based on the illumination information, working parameter analysis is performed to determine the working parameters corresponding to the image acquisition device, and the working parameters are sent to the image acquisition device to control the image acquisition device to acquire images according to the working parameters, wherein the working parameters include: exposure time, shutter speed, white balance speed and ISO value.

[0042] For the embodiments of the present application, the instrument may be in various complex lighting conditions. In order to ensure that clear instrument images can be captured in various complex lighting environments, the working parameters of the image acquisition device are dynamically adjusted based on the lighting conditions of the environment in which the instrument is located, thereby significantly reducing image blur, overexposure or underexposure caused by light problems, thereby improving the recognition accuracy of subsequent instrument readings.

[0043] Specifically, a light sensor is pre-deployed at a preset position of the meter. Preferably, a high-precision, high-sensitivity light sensor is selected. Such sensors usually use sensitive light-collecting devices and are equipped with filters, cosine correctors and other components to ensure the accuracy of the measurement. The deployed preset position should be able to fully reflect the light intensity and light conditions of the surrounding environment of the meter. The electronic device is connected to the light sensor via wireless transmission, so that light information is obtained from the light sensor via wireless transmission. The light information includes but is not limited to: light intensity, light color, light direction, light intensity change, etc.

[0044] Furthermore, the electronic device pre-stores the mapping relationship between the illumination information and the working parameters of the image acquisition device. The mapping relationship is determined by testing and analyzing a large number of illumination information and working parameter combinations to determine the optimal working parameter settings. Of course, the mapping relationship can also be optimized using machine learning, deep learning and other algorithms to improve the accuracy and efficiency of the working parameter analysis. Then, the illumination information and the mapping relationship are used to analyze the working parameters to determine the working parameters corresponding to the image acquisition device. The working parameters are the parameters corresponding to the image acquisition device being able to capture the instrument video highlight most clearly under the current illumination information. Finally, the working parameters are sent to the image acquisition device to control the image acquisition device to perform image acquisition according to the working parameters.

[0045] It can be seen that in the embodiment of the present application, in order to ensure that clear instrument images can be collected in various complex lighting environments, the lighting information collected by the light sensor is obtained, the working parameter analysis is performed based on the lighting information, the working parameters corresponding to the image acquisition device are determined, and the working parameters are sent to the image acquisition device to control the image acquisition device to perform image acquisition according to the working parameters. The working parameters of the image acquisition device are dynamically adjusted to significantly reduce problems such as image blur, overexposure or underexposure caused by light problems, thereby improving the recognition accuracy of subsequent instrument readings.

[0046] Furthermore, in order to enable the image acquisition device to work under optimal lighting conditions and improve the image acquisition quality, in the embodiment of the present application, after obtaining the lighting information collected by the light sensor, the following is also included: Performing a lighting condition evaluation based on the lighting information to determine a lighting condition evaluation result, wherein the lighting condition evaluation result includes: meeting the image acquisition requirement and not meeting the image acquisition requirement; When the lighting condition evaluation result is that the image acquisition requirement is not met, a fill light analysis is performed based on the lighting information to determine a fill light operation, wherein the fill light operation includes: adding a light source and adjusting the direction of light.

[0047] For the embodiments of the present application, in an actual working environment, not all lighting conditions meet the image acquisition requirements. Insufficient lighting or excessive lighting will lead to problems such as image blur and color distortion, affecting the quality of the instrument video image and causing errors in the instrument readings. Therefore, in order to improve the image acquisition quality, a lighting condition assessment is performed and when the image acquisition requirements are not met, a fill light operation is automatically performed so that the image acquisition device can work under the best lighting conditions.

[0048] Specifically, the lighting condition evaluation standard pre-stored in the electronic device is obtained. The lighting condition evaluation standard is set according to the technical specifications and application requirements of the image acquisition device, including but not limited to: the range of light intensity, the appropriate value of color temperature, the limitation of light direction, etc. Then, a multi-dimensional analysis is performed based on the lighting information, and the multi-dimensional analysis results are matched with the lighting condition evaluation standard, wherein the multi-dimensional analysis includes but is not limited to: light intensity distribution analysis, color temperature change analysis, and light direction stability analysis. When the multi-dimensional analysis results all match the lighting condition evaluation standard, it is determined that the lighting condition evaluation result meets the image acquisition requirements, otherwise, it is determined that the lighting condition evaluation result does not meet the image acquisition requirements.

[0049] When the result of the lighting condition evaluation is that it meets the image acquisition requirements, no other operations need to be performed; when the result of the lighting condition evaluation is that it does not meet the image acquisition requirements, fill light analysis is performed based on the lighting information to determine the fill light operation, which includes: adding light sources and adjusting the direction of light. For fill light analysis, key parameter evaluation is performed based on lighting information and lighting condition evaluation standards to determine the gap between the current lighting conditions and the standards. If the light intensity is insufficient, the fill light operation is determined to be adding light sources; if the light direction is inappropriate, the fill light operation is determined to be adjusting the light direction, that is, the position of the light source can be adjusted, the light can be guided by a reflector, etc.

[0050] It can be seen that in the embodiment of the present application, the lighting condition evaluation is performed based on the lighting information to determine the lighting condition evaluation result. In order to improve the image acquisition quality, when the lighting condition evaluation result does not meet the image acquisition requirements, a fill light analysis is performed based on the lighting information to determine the fill light operation so that the image acquisition device can work under the best lighting conditions.

[0051] Further, in order to improve the accuracy and reliability of the meter reading, in the embodiment of the present application, the meter reading is calculated based on the calibration data, and after the meter reading is determined, the following steps are further included: Acquire instrument environment data, perform environmental interference analysis based on the instrument environment data, and determine environmental interference analysis results, wherein the environmental interference analysis results include: interference exists and no interference exists; When the result of the environmental interference analysis is that interference exists, the corresponding target sensor is determined based on the type of the instrument reading, and the industrial data collected by the target sensor is obtained, and the instrument reading is updated based on the industrial data.

[0052] For the embodiments of the present application, since the instrument reading may be affected by many factors, in one case, the instrument zero drift caused by temperature change, the instrument deformation caused by pressure change, etc.; in another case, in some industrial or scientific research environments, the instrument may be under extreme temperature, pressure or other physical conditions; under these conditions, only reading the instrument through video images will cause the reading to be biased or unstable, thereby affecting the accuracy and reliability of the data. Therefore, when the instrument is in an environment that interferes with the reading, a combination of video images and sensor acquisition is used to read the instrument value to improve the accuracy and reliability of the instrument reading.

[0053] Specifically, a variety of environmental sensors are deployed in the relevant environment where the instrument is located, including but not limited to: temperature sensors, pressure sensors, humidity sensors, etc., to monitor environmental changes in real time. The environmental sensors collect environmental data in real time and transmit it to the electronic device wirelessly. Then, according to the characteristics of the instrument and the working environment, an environmental interference identification model is established. The environmental interference identification model should be able to automatically determine whether there are factors that interfere with the instrument reading based on the instrument environmental data, that is, set the interference thresholds of various environmental parameters in the interference identification model. Then, using the environmental interference identification model, the instrument environmental data is analyzed for environmental interference, and the environmental interference analysis results are determined. The environmental interference analysis results include: interference and no interference, that is, when there is at least one environmental data in the instrument environmental data that exceeds the interference threshold, it is determined that there is interference; otherwise, it is determined that there is no interference.

[0054] Furthermore, when the result of the environmental interference analysis is that interference exists, the corresponding target sensor is determined based on the type of instrument reading, and the industrial data collected by the target sensor is obtained, and the instrument reading is updated based on the industrial data. For example, when the instrument reading is temperature, a high-precision temperature sensor is selected as the target sensor, and the industrial data collected by the target sensor is used to update the instrument reading, that is, the industrial data collected by the high-precision and accurate sensor is used to update the reading with deviations, thereby improving the accuracy and reliability of the instrument reading.

[0055] It can be seen that in the embodiment of the present application, the instrument environment data is obtained, the environmental interference analysis is performed based on the instrument environment data, and the environmental interference analysis result is determined. When the environmental interference analysis result is that interference exists, the corresponding target sensor is determined based on the type of instrument reading, and the industrial data collected by the target sensor is obtained, and the instrument reading is updated based on the industrial data. When the instrument is in an environment that interferes with the reading, the instrument value is read by combining video images and sensor acquisition to improve the accuracy and reliability of the instrument reading.

[0056] Furthermore, in order to reduce positioning errors caused by image blur or noise interference and improve the quality and reliability of industrial data, in the embodiment of the present application, pointer positioning refinement and scale calibration are performed based on the key features of the instrument to obtain calibration data, including: The edge detection algorithm is used to refine the pointer location features in the key features of the instrument to determine the position of the pointer tip; Use template matching and regression analysis to calibrate the scale distribution features in the key features of the instrument to determine the scale distribution information; The pointer tip position and scale distribution information are integrated to obtain calibration data.

[0057] For the embodiments of the present application, in the process of extracting instrument features using a convolutional neural network, the instrument image may be affected by factors such as lighting changes, noise interference, and camera angle, which may cause certain errors in the pointer position and scale distribution extracted by the instrument feature recognition model, thereby affecting the accuracy of subsequent instrument readings. In order to further improve the accuracy of instrument reading recognition, the pointer positioning refinement and scale calibration are performed based on the key features of the instrument to obtain calibration data, which are used to reduce positioning errors caused by image blur or noise interference.

[0058] Specifically, the edge detection algorithm is used to refine the pointer position features in the key features of the instrument to determine the position of the pointer tip, that is, the edge detection algorithm is used to accurately identify the pointer edge information, and the outline of the pointer is further refined through morphological processing (such as corrosion, expansion, refinement, etc.), so as to more accurately determine the position of the pointer tip. At the same time, for the case where the instrument scale is regular and stable, the scale distribution features in the instrument image are matched with the preset template through the template matching algorithm, so as to quickly identify the scale distribution information; for instruments with irregular or changing scale distribution, regression analysis statistical methods can be used to fit the scale distribution information based on the position information of the known scale points. Finally, the calibration data is obtained by integrating the pointer tip position and scale distribution information. The execution of pointer positioning refinement and scale calibration operations improves the quality and reliability of industrial data and provides a strong guarantee for the monitoring and optimization of the production process.

[0059] It can be seen that in the embodiment of the present application, the edge detection algorithm is used to refine the pointer positioning of the pointer position feature in the key feature of the instrument to determine the position of the pointer tip. At the same time, the scale distribution feature in the key feature of the instrument is calibrated by template matching and regression analysis to determine the scale distribution information. Finally, the pointer tip position and scale distribution information are combined to obtain calibration data. The pointer positioning refinement and scale calibration are used to reduce the positioning error caused by image blur or noise interference, thereby improving the quality and reliability of industrial data.

[0060] The above embodiment introduces an instrument intelligent analysis method based on video images from the perspective of method flow, and the following embodiment introduces an instrument intelligent analysis system based on video images from the perspective of virtual modules or virtual units. For details, please refer to the following embodiments.

[0061] The present application embodiment provides an instrument intelligent analysis system based on video images, such as Figure 2 As shown, the video image-based instrument intelligent analysis system may specifically include: The preprocessing module 210 is used to obtain the instrument video image, and perform preprocessing based on the instrument video image to obtain a preprocessed instrument image, wherein the preprocessing includes: filtering and denoising, grayscale conversion, and illumination contrast parameter adjustment; An image segmentation module 220 is used to perform an image segmentation operation based on the pre-processed instrument image to obtain an instrument region segmentation map, wherein the instrument region segmentation map is used to divide different functional regions in the instrument into zones; The feature extraction module 230 is used to obtain an instrument feature recognition model, and use the instrument feature recognition model to extract instrument features from the instrument region segmentation map to determine the instrument key features, wherein the instrument key features include: pointer position features and scale distribution features; The refinement and calibration module 240 is used to perform pointer positioning refinement and scale calibration based on the key features of the instrument when the instrument type is a pointer instrument, obtain calibration data, and calculate the instrument reading based on the calibration data to determine the instrument reading.

[0062] For the embodiment of the present application, preprocessing is performed based on the instrument video image to obtain a preprocessed instrument image. The preprocessing includes adjusting the illumination contrast parameters. The illumination contrast parameters are adjusted on the instrument video image so that the brightness and contrast of the processed instrument image reach a suitable range, thereby enhancing the visual effect and clarity of the image. Then, an image segmentation operation is performed based on the preprocessed instrument image to obtain an instrument region segmentation map. The image segmentation operation is used to accurately locate the feature area to be identified, thereby improving the accuracy of the key features of the instrument. Furthermore, the instrument feature recognition model is used to extract instrument features from the instrument region segmentation map to determine the key features of the instrument. Thereafter, the pointer positioning is refined and the scale is calibrated based on the key features of the instrument to obtain calibration data, and the instrument reading is calculated based on the calibration data to determine the instrument reading. The instrument data recognition method that combines classical image processing technology and convolutional neural network model feature extraction for data recognition improves the accuracy of instrument data reading.

[0063] A possible implementation of the embodiment of the present application is a video image-based instrument intelligent analysis system, further comprising: A reading prediction module is used to store the meter readings in a data monitoring database, wherein the data monitoring database is used to store the meter readings corresponding to different nodes of each meter; Extracting an instrument reading sequence corresponding to a target instrument from a data monitoring database, wherein the target instrument is any instrument stored in the data monitoring database, and the instrument reading sequence consists of a plurality of instrument readings arranged in chronological order; Reading prediction is performed based on the instrument reading sequence to determine the prediction data, and preventive evaluation is performed based on the prediction data to obtain preventive maintenance information.

[0064] A possible implementation of the embodiment of the present application is a video image-based instrument intelligent analysis system, further comprising: A light analysis module is used to obtain light information collected by a light sensor, wherein the light sensor is deployed at a preset position of the meter and is used to detect the light intensity and light conditions of the meter's surrounding environment in real time; Based on the illumination information, working parameter analysis is performed to determine the working parameters corresponding to the image acquisition device, and the working parameters are sent to the image acquisition device to control the image acquisition device to acquire images according to the working parameters, wherein the working parameters include: exposure time, shutter speed, white balance speed and ISO value.

[0065] A possible implementation of the embodiment of the present application is a video image-based instrument intelligent analysis system, further comprising: An illumination evaluation module is used to evaluate illumination conditions based on illumination information and determine an illumination condition evaluation result, wherein the illumination condition evaluation result includes: meeting image acquisition requirements and not meeting image acquisition requirements; When the lighting condition evaluation result is that the image acquisition requirement is not met, a fill light analysis is performed based on the lighting information to determine a fill light operation, wherein the fill light operation includes: adding a light source and adjusting the direction of light.

[0066] A possible implementation of the embodiment of the present application is a video image-based instrument intelligent analysis system, further comprising: A data update module is used to obtain instrument environment data, perform environmental interference analysis based on the instrument environment data, and determine the environmental interference analysis result, wherein the environmental interference analysis result includes: interference exists and no interference exists; When the result of the environmental interference analysis is that interference exists, the corresponding target sensor is determined based on the type of the instrument reading, and the industrial data collected by the target sensor is obtained, and the instrument reading is updated based on the industrial data.

[0067] In a possible implementation of the embodiment of the present application, when the refinement calibration module 240 performs pointer positioning refinement and scale calibration based on the key features of the instrument to obtain calibration data, it is used to: The edge detection algorithm is used to refine the pointer location features in the key features of the instrument to determine the position of the pointer tip; Use template matching and regression analysis to calibrate the scale distribution features in the key features of the instrument to determine the scale distribution information; The pointer tip position and scale distribution information are integrated to obtain calibration data.

[0068] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described video image-based instrument intelligent analysis system can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0069] An electronic device is provided in an embodiment of the present application, such as Figure 3 As shown, Figure 3 The electronic device 300 shown includes: a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, such as through a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that in actual applications, the transceiver 304 is not limited to one, and the structure of the electronic device 300 does not constitute a limitation on the embodiments of the present application.

[0070] The processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. The processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0071] The bus 302 may include a path to transmit information between the above components. The bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but it does not mean that there is only one bus or only one type of bus.

[0072] The memory 303 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0073] The memory 303 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the contents shown in the above method embodiment.

[0074] The electronic devices include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and fixed terminals such as digital TVs, desktop computers, etc. It can also be a server, etc. Figure 3 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0075] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding content in the aforementioned method embodiment.

[0076] The embodiment of the present application provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the method in any of the above embodiments is implemented. Compared with the related art, the embodiment of the present application performs preprocessing based on the instrument video image to obtain a preprocessed instrument image. The preprocessing includes adjusting the illumination contrast parameter. The illumination contrast parameter is adjusted on the instrument video image so that the brightness and contrast of the processed instrument image reach a suitable range, thereby enhancing the visual effect and clarity of the image. Then, an image segmentation operation is performed based on the preprocessed instrument image to obtain an instrument area segmentation map. The image segmentation operation is used to accurately locate the feature area to be identified, thereby improving the accuracy of the key features of the instrument. Furthermore, the instrument feature recognition model is used to extract instrument features from the instrument area segmentation map to determine the key features of the instrument. After that, the pointer positioning refinement and scale calibration are performed based on the key features of the instrument to obtain calibration data, and the instrument reading is calculated based on the calibration data to determine the instrument reading. The instrument data recognition method that combines the classical image processing technology and the convolutional neural network model feature extraction for data recognition improves the accuracy of instrument data reading.

[0077] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0078] The above are only some implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An intelligent instrument analysis method based on video images, characterized in that: include: Acquire an instrument video image, and perform preprocessing based on the instrument video image to obtain a preprocessed instrument image, wherein the preprocessing includes: filtering and denoising, grayscale conversion, and illumination contrast parameter adjustment; Performing an image segmentation operation based on the preprocessed instrument image to obtain an instrument region segmentation map, wherein the instrument region segmentation map is used to divide different functional areas in the instrument into zones; Acquire an instrument feature recognition model, and use the instrument feature recognition model to extract instrument features from the instrument region segmentation map to determine instrument key features, wherein the instrument key features include: pointer position features and scale distribution features; When the instrument type is a pointer-type instrument, the pointer positioning is refined and the scale is calibrated based on the key features of the instrument to obtain calibration data, and the instrument reading is calculated based on the calibration data to determine the instrument reading.

2. The video image-based instrument intelligent analysis method according to claim 1 is characterized in that: After calculating the meter reading based on the calibration data and determining the meter reading, the method further includes: Storing the meter readings in a data monitoring database, wherein the data monitoring database is used to store meter readings corresponding to different nodes of each meter; Extracting an instrument reading sequence corresponding to a target instrument from the data monitoring database, wherein the target instrument is any instrument stored in the data monitoring database, and the instrument reading sequence consists of a plurality of instrument readings arranged in chronological order; Reading prediction is performed based on the instrument reading sequence to determine prediction data, and preventive evaluation is performed based on the prediction data to obtain preventive maintenance information.

3. The video image-based instrument intelligent analysis method according to claim 1 is characterized in that: Before acquiring the instrument video image, the method further includes: Acquiring illumination information collected by a light sensor, wherein the light sensor is deployed at a preset position of the instrument and is used to detect the light intensity and illumination conditions of the instrument's surrounding environment in real time; Based on the illumination information, working parameter analysis is performed to determine working parameters corresponding to the image acquisition device, and the working parameters are sent to the image acquisition device to control the image acquisition device to perform image acquisition according to the working parameters, wherein the working parameters include: exposure time, shutter speed, white balance speed and ISO value.

4. The video image-based instrument intelligent analysis method according to claim 3 is characterized in that: After obtaining the illumination information collected by the light sensor, the method further includes: Performing a lighting condition evaluation based on the lighting information to determine a lighting condition evaluation result, wherein the lighting condition evaluation result includes: satisfying image acquisition requirements and not satisfying image acquisition requirements; When the illumination condition evaluation result is that the image acquisition requirement is not met, a fill light analysis is performed based on the illumination information to determine a fill light operation, wherein the fill light operation includes: adding a light source and adjusting a light direction.

5. The video image-based instrument intelligent analysis method according to claim 1 is characterized in that: After calculating the meter reading based on the calibration data and determining the meter reading, the method further includes: Acquire instrument environment data, perform environmental interference analysis based on the instrument environment data, and determine an environmental interference analysis result, wherein the environmental interference analysis result includes: interference exists and no interference exists; When the environmental interference analysis result indicates that interference exists, a corresponding target sensor is determined based on the type of the instrument reading, and industrial data collected by the target sensor is obtained, and the instrument reading is updated based on the industrial data.

6. The video image-based instrument intelligent analysis method according to claim 1 is characterized in that: The pointer positioning refinement and scale calibration based on the key features of the instrument to obtain calibration data includes: Using an edge detection algorithm to refine the pointer position feature in the key feature of the instrument to determine the position of the pointer tip; Using template matching and regression analysis to calibrate the scale distribution features in the key features of the instrument to determine scale distribution information; The calibration data is obtained by combining the pointer tip position and the scale distribution information.

7. An instrument intelligent analysis system based on video images, characterized in that: include: A preprocessing module, used for acquiring an instrument video image, and performing preprocessing based on the instrument video image to obtain a preprocessed instrument image, wherein the preprocessing includes: filtering and denoising, grayscale conversion, and illumination contrast parameter adjustment; An image segmentation module, used for performing an image segmentation operation based on the pre-processed instrument image to obtain an instrument region segmentation map, wherein the instrument region segmentation map is used for dividing different functional regions in the instrument; A feature extraction module is used to obtain an instrument feature recognition model, and use the instrument feature recognition model to extract instrument features from the instrument region segmentation map to determine instrument key features, wherein the instrument key features include: pointer position features and scale distribution features; The refinement calibration module is used to perform pointer positioning refinement and scale calibration based on the key features of the instrument when the instrument type is a pointer instrument, obtain calibration data, and calculate the instrument reading based on the calibration data to determine the instrument reading.

8. An electronic device, characterized in that: include: at least one processor; Memory; At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the video image-based instrument intelligent analysis method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed in a computer, the computer is caused to execute the video image-based instrument intelligent analysis method as described in any one of claims 1 to 6.

Images