Liquid column meter reading recognition method and device, storage medium, terminal

Abstract

A method and device, storage medium, and terminal for recognizing the reading of a liquid column type meter, the method comprising: obtaining a plurality of first sample liquid level images, the plurality of first sample liquid level images comprising a range area, the range area being divided into a liquid column area and a non-liquid column area by the liquid level; training a preset recognition model using a set of sample liquid level images to obtain a liquid level recognition model; obtaining a liquid level image to be recognized, the liquid level image to be recognized comprising an image of a liquid column type meter to be recognized; inputting the liquid level image to be recognized into the liquid level recognition model to obtain a liquid level ratio output by the liquid level recognition model; determining the reading of the liquid column type meter to be recognized according to the liquid level ratio output by the liquid level recognition model and the range of the liquid column type meter to be recognized. Through the scheme provided by the present invention, the reading of the liquid column type meter can be determined efficiently and accurately.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a reading recognition method and device, a storage medium, and a terminal for a liquid column type meter. Background Art

[0002] The liquid column type meter is a traditional meter that uses the height of the liquid column to reflect the measurement results. It is widely used in production and life. In some industrial scenarios, liquid column type meters are usually set up to monitor production equipment, production environment, etc. In the prior art, the readings of the liquid column type meter are mainly read manually to determine whether the readings are within the normal range to achieve the purpose of monitoring. Such a solution requires a lot of manpower and is inefficient. In addition, the human eye is prone to misjudgment due to factors such as fatigue, which may cause production safety accidents.

[0003] Therefore, there is an urgent need for a reading recognition method for a liquid column type meter that can efficiently and accurately determine the reading of the liquid column type meter. Summary of the invention

[0004] The technical problem solved by the invention is how to efficiently and accurately determine the reading of a liquid column type instrument.

[0005] To solve the above technical problems, an embodiment of the present invention provides a method for identifying the reading of a liquid column type meter, the method comprising: obtaining a plurality of first sample liquid level images, the plurality of first sample liquid level images comprising a range area, the range area being divided into a liquid column area and a non-liquid column area by liquid level, the range area being an area covered by the range of the liquid column type meter, each first sample liquid level image having a pre-marked liquid level ratio, the liquid level ratio referring to the ratio of the liquid column area to the range area in the scale direction; training a preset recognition model using a sample liquid level image set to obtain a liquid level recognition model, the sample liquid level image set comprising the plurality of first sample liquid level images; obtaining a liquid level image to be identified, the liquid level image to be identified comprising an image of the liquid column type meter to be identified; inputting the liquid level image to be identified into the liquid level recognition model to obtain the liquid level ratio output by the liquid level recognition model; determining the reading of the liquid column type meter to be identified according to the liquid level ratio output by the liquid level recognition model and the range of the liquid column type meter to be identified.

[0006] Optionally, the sample liquid level image set also includes multiple second sample liquid level images. Before using the sample liquid level image set to train the preset recognition model, the method also includes: cutting the multiple first sample liquid level images to obtain the multiple second sample liquid level images, wherein the liquid level ratio of the second sample liquid level image is different from the liquid level ratio of the corresponding first sample liquid level image.

[0007] Optionally, cutting the multiple first sample liquid level images includes: cutting the multiple first sample liquid level images at any position in the non-liquid column area to obtain the multiple second sample liquid level images, wherein the liquid level ratio of the second sample liquid level image is greater than the liquid level ratio of the corresponding first sample liquid level image.

[0008] Optionally, acquiring multiple first sample liquid level images includes: acquiring multiple first sample device images, the multiple first sample device images including an image of a target device, the liquid column type instrument being arranged on the target device; performing data enhancement on the multiple first sample device images to obtain multiple second sample device images; and intercepting images in a sample device image set to obtain the multiple first sample liquid level images, wherein the sample device image set includes the multiple first sample device images and the multiple second sample device images.

[0009] Optionally, the data enhancement is selected from one or more of the following: brightness transformation, chromaticity transformation and saturation transformation.

[0010] Optionally, the images in the sample device image set have pre-labeled labels, and the labels are used to indicate the positions of the liquid column type instruments in the images. Before acquiring the images of the liquid level meters to be identified, the method further includes: using the sample device image set to train a preset detection model to obtain an instrument detection model.

[0011] Optionally, the liquid column type instrument to be identified is arranged on the device to be identified, and obtaining the liquid level image to be identified includes: obtaining the image to be detected, the image to be detected includes the image of the device to be identified; inputting the image to be detected into the instrument detection model to obtain the liquid level image to be identified.

[0012] The embodiment of the present invention further provides a reading recognition device for a liquid column type meter, the device comprising: a sample acquisition module, used to acquire a plurality of first sample liquid level images, the plurality of first sample liquid level images comprising a range area, the range area being divided into a liquid column area and a non-liquid column area by liquid level, the range area being an area covered by the range of the liquid column type meter, each first sample liquid level image having a pre-marked liquid level ratio, the liquid level ratio being the ratio of the liquid column area to the range area in the scale direction; an identification training module, used to use a sample liquid level image set to train a preset identification The liquid level recognition model is trained to obtain a liquid level recognition model, wherein the sample liquid level image set includes the multiple first sample liquid level images; an image acquisition module is used to acquire the liquid level image to be recognized, wherein the liquid level image to be recognized includes the image of the liquid column type meter to be recognized; a ratio determination module is used to input the liquid level image to be recognized into the liquid level recognition model to obtain the liquid level ratio output by the liquid level recognition model; and a recognition module is used to determine the reading of the liquid column type meter to be recognized according to the liquid level ratio output by the liquid level recognition model and the range of the liquid column type meter to be recognized.

[0013] An embodiment of the present invention further provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the liquid level identification method of the liquid column type meter are executed.

[0014] An embodiment of the present invention further provides a terminal, including a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor executes the steps of the above-mentioned liquid level identification method for the liquid column type instrument when running the computer program.

[0015] Compared with the prior art, the technical solution of the embodiment of the present invention has the following beneficial effects:

[0016] In the scheme of the embodiment of the present invention, a sample liquid level image set is used to train a preset recognition model. Since the sample liquid level image set includes multiple first sample liquid level images, and the first sample liquid level image has a pre-marked liquid level ratio, the trained liquid level recognition model can be used to identify the ratio of the liquid column area of ​​the liquid column type instrument in the scale direction to the range area in the image. Furthermore, by obtaining the liquid level image to be identified, and inputting the liquid level image to be identified into the liquid level recognition model, the liquid level ratio corresponding to the image to be identified can be obtained, and the reading of the liquid column type instrument to be identified can be obtained in combination with the range of the liquid column type instrument to be identified. With such a scheme, there is no need to identify the scale line and reading of the liquid level in the liquid level image to be identified, which reduces the difficulty of identification and is conducive to improving the accuracy of the reading. In addition, this process does not require human participation, which is also conducive to improving the efficiency of identifying readings.

[0017] Furthermore, in the solution of the embodiment of the present invention, a plurality of first sample liquid level images are cut to obtain a plurality of second sample liquid level images, wherein the liquid level ratio of the second sample liquid level image is different from the liquid level ratio of the first sample liquid level image before cutting. By adopting such a solution, the number of images in the sample liquid level image set can be increased, that is, the amount of training data for subsequent training of the liquid level recognition model can be increased, which is conducive to improving the recognition performance of the model, thereby improving the accuracy of reading recognition.

[0018] Furthermore, in the scheme of the embodiment of the present invention, cutting is performed at any position of the non-liquid column area in each first sample liquid level image to obtain a corresponding second sample liquid level image. Among them, the liquid level ratio of the second sample liquid level image is greater than the liquid level ratio in the first sample liquid level image before cutting. Compared with the scheme of cutting at any position of the liquid column area, such a scheme can increase the liquid level ratio. In other words, it can increase the proportion of the liquid column area in the range area, and can make the liquid level recognition model obtained by subsequent training more sensitive to changes in the height of the liquid column, which is beneficial to improving the performance of the liquid level recognition model. Especially when the reading variation range of the liquid column type meter is small, the accuracy of the reading recognition can still be ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of a method for identifying readings of a liquid column type meter in an embodiment of the present invention;

[0020] Figure 2 is a schematic diagram of a first sample liquid level image in an embodiment of the present invention;

[0021] Figure 3 yes Figure 2 A schematic diagram of a measuring range area;

[0022] Figure 4 yes Figure 1 A flowchart of a specific implementation of step S101;

[0023] Figure 5 is a schematic diagram of a first sample device image in an embodiment of the present invention;

[0024] Figure 6 1 is a schematic diagram of another scenario of a method for identifying readings of a liquid column type meter in an embodiment of the present invention;

[0025] Figure 7 It is a structural schematic diagram of a reading recognition device of a liquid column type meter in an embodiment of the present invention. DETAILED DESCRIPTION

[0026] As described in the background art, there is an urgent need for a reading recognition method for a liquid column type meter, which can efficiently and accurately determine the reading of the liquid column type meter.

[0027] The inventor of the present invention has found through research that the reading corresponding to the scale line can be determined by identifying the scale line corresponding to the liquid level of the liquid column in the image. However, since the quality of the images collected in industrial scenes is usually low, the accuracy of the reading obtained by using this solution is poor.

[0028] In order to solve the above technical problems, an embodiment of the present invention provides a method for identifying the reading of a liquid column type meter. In the scheme of this embodiment, a sample liquid level image set is used to train a preset recognition model. Since the sample liquid level image set includes multiple first sample liquid level images, and the first sample liquid level image has a pre-marked liquid level ratio, the trained liquid level recognition model can be used to identify the ratio of the liquid column area of ​​the liquid column type meter in the scale direction to the range area in the image. Further, the liquid level image to be identified is obtained, and the liquid level image to be identified is input into the liquid level recognition model, so that the liquid level ratio corresponding to the image to be identified can be obtained, and the reading of the liquid column type meter to be identified can be obtained in combination with the range of the liquid column type meter to be identified. With such a scheme, there is no need to identify the scale line and reading of the liquid level in the liquid level image to be identified, which reduces the difficulty of identification and is conducive to improving the accuracy of the reading. In addition, this process does not require manual participation, which is also conducive to improving the efficiency of identifying readings.

[0029] In order to make the above-mentioned objects, features and beneficial effects of the present invention more obvious and easy to understand, specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0030] Reference Figure 1 , Figure 1 It is a reading recognition method of a liquid column type meter in an embodiment of the present invention. The method can be executed by a terminal. The terminal can be various devices with data receiving and data processing functions, for example, a mobile phone, a computer, a tablet computer, an Internet of Things device, a cruise robot, and a server, etc., but is not limited thereto. The liquid column type meter in the embodiment of the present invention can be various existing meters that determine the measurement results by the height of the liquid column. This embodiment does not limit the specific type of the liquid column type meter. Figure 1 The method for identifying the reading of the liquid column type meter shown may include the following steps:

[0031] Step S101: acquiring a plurality of first sample liquid level images;

[0032] Step S102: using a sample liquid level image set to train a preset recognition model to obtain a liquid level recognition model;

[0033] Step S103: Acquire a liquid level image to be identified;

[0034] Step S104: inputting the to-be-recognized liquid level image into the liquid level recognition model to obtain the liquid level ratio output by the liquid level recognition model;

[0035] Step S105: determining the reading of the liquid column type meter to be identified according to the liquid level ratio output by the liquid level identification model and the measuring range of the liquid column type meter to be identified.

[0036] It is understandable that, in a specific implementation, the method can be implemented in the form of a software program, which runs in a processor integrated inside a chip or a chip module; or, the method can be implemented in the form of hardware or a combination of hardware and software.

[0037] In the specific implementation of step S101, multiple first sample liquid level images may be obtained, and the multiple first sample liquid level images may include images of the liquid column type instrument. Specifically, each first sample liquid level image includes a range area, and the range area is the area covered by the range of the liquid column type instrument.

[0038] It should be noted that the liquid column type instrument in this embodiment includes a liquid column, and the height of the liquid column in the instrument reflects the measurement result. For example, a liquid column type thermometer can reflect the temperature value through the height of the liquid column. For another example, a liquid column type liquid level meter can reflect the position of the liquid level through the height of the liquid column.

[0039] Furthermore, each first sample liquid level image has a pre-marked liquid level ratio, wherein the liquid level ratio refers to the ratio of the liquid column area to the range area in the scale direction of the liquid column type instrument.

[0040] Reference Figure 2 , Figure 2 FIG. 1 is a schematic diagram of a first sample liquid level image in an embodiment of the present invention. Figure 2 As shown, the first sample liquid level image 20 includes an image of a liquid column type meter 21. More specifically, the first sample liquid level image 20 includes a range area 22, which is an area covered by the range of the liquid column type meter 21. It should be noted that this embodiment does not limit the shape of the range area 22. Furthermore, the liquid column type meter 21 has a liquid column 23 inside, and the first sample liquid level image 20 also includes an image of the liquid column 23. More specifically, Figure 2 The x direction shown in is the scale direction of the liquid column type meter 21 .

[0041] Reference Figure 3 , Figure 3 yes Figure 2 A schematic diagram of a measuring range region 22 is shown in FIG. Figure 3As shown, the range area 22 is divided into a liquid column area 221 and a non-liquid column area 222 by the liquid level 220 of the liquid column 23. The liquid level 220 is the liquid surface position of the liquid column; the liquid column area 221 is the area covered by the liquid column 23 in the first sample liquid level image 20, in other words, the liquid column area 221 is the area between the zero scale line and the liquid level 220 in the range area 22; the non-liquid column area 222 is the area in the range area 22 except the liquid column area 221.

[0042] Further, the liquid level ratio may be the ratio of the length of the liquid column region 221 in the scale direction to the length of the range region 22 in the scale direction. More specifically, Figure 3 The length of the middle liquid column area 221 in the scale direction is H1, and the length of the range area 22 in the scale direction is (H1+H2), then the liquid level ratio is

[0043] It should be noted that the solution of this embodiment does not focus on the widths of the liquid column area, the non-liquid column area and the range area, but focuses on the ratio of the heights of the liquid column area and the range area (that is, the length in the scale direction). Therefore, this embodiment does not limit the width of the range area.

[0044] Continue to refer Figure 1 This embodiment does not limit the method for obtaining multiple first sample liquid level images. The multiple first sample liquid level images may be pre-stored in the terminal or obtained from the outside, but is not limited thereto.

[0045] It is understandable that the plurality of first sample liquid level images may be collected in a target scene, which may be an actual business scene using a liquid column type instrument and having a reading recognition requirement. In a specific example, the target scene is a scene of monitoring the temperature of equipment in an oil plant by a thermometer.

[0046] Reference Figure 4 , Figure 4 yes Figure 1 A flowchart of a specific implementation of step S101 in FIG. Figure 4 The illustrated step S101 may include the following steps:

[0047] Step S401: Acquire multiple first sample device images;

[0048] Step S402: performing data enhancement on the plurality of first sample device images to obtain a plurality of second sample device images;

[0049] Step S403: intercepting images in the sample device image set to obtain the plurality of first sample liquid level images.

[0050] In the specific implementation of step S401, the liquid column meter is usually set on the target device to monitor the operating state or operating environment of the target device. Thus, the target device can be photographed to obtain multiple first sample device images, wherein the first sample device images include the image of the target device.

[0051] refer to Figure 5 , Figure 5 is a schematic diagram of a first sample device image in an embodiment of the present invention, wherein, in order to ensure the clarity of the schematic diagram, Figure 5 The background part of the first sample device image is not shown in FIG. 1 , that is, only the image of the liquid column type instrument in the first sample device image is shown. Figure 5 As shown, the first sample device image includes the image of the liquid column type instrument. It should be noted that the background of the first sample device image is complex and the image of the liquid column type instrument occupies a relatively small proportion in the first sample device image.

[0052] Continue to refer Figure 4 In the specific implementation of step S402, data augmentation may be performed on the plurality of first sample device images to obtain a plurality of second sample device images.

[0053] Specifically, random data augmentation (RandomAugmentation) may be performed on the plurality of first sample device images, or automatic data augmentation (AutoAugmentation) may be performed on the plurality of first sample device images, which is not limited in this embodiment.

[0054] More specifically, data enhancement may include one or more of the following: brightness transformation, chromaticity transformation, and saturation transformation. Such a scheme can simulate the diversity of background and light in the target scene, which can increase the richness of training data and enhance the authenticity of training data.

[0055] It should be noted that the first sample device image may also be processed by using various other existing data enhancement methods, and this embodiment does not limit this.

[0056] In a specific implementation of step S403, the sample device image set may include the first sample device image and the second sample device image mentioned above, wherein the second sample device image is obtained by performing data enhancement on the first sample device image.

[0057] Furthermore, the images in the sample device image set may be intercepted, and more specifically, the area where the liquid column type instrument is located in the image may be intercepted to obtain a plurality of first sample liquid level images.

[0058] It should be noted that, in other embodiments, multiple first sample liquid level images may be captured first, and then data enhancement may be performed on the first sample liquid level images.

[0059] Continue to refer Figure 1 Before executing step S102, the first sample liquid level image may be enhanced. Specifically, multiple first sample liquid level images may be sheared to obtain multiple second sample liquid level images. More specifically, the first sample liquid level image may be sheared along a shearing direction at any position within the range area, wherein the shearing direction is a direction perpendicular to the scale direction. Figure 2 and Figure 3 As shown, the shear direction can be Figure 2 and Figure 3 y direction in .

[0060] It is understandable that, since the shearing is performed along the shearing direction within the measuring range, the liquid level ratio of the second sample liquid level image obtained by shearing is different from the liquid level ratio of the first sample liquid level image before shearing. By adopting such a solution, the number of images in the sample liquid level image set can be increased, that is, the amount of training data for subsequent training of the liquid level recognition model can be increased, which is conducive to improving the recognition performance of the model, thereby improving the accuracy of reading recognition.

[0061] In a non-limiting example, cutting can be performed at any position of the non-liquid column area in each first sample liquid level image to obtain a corresponding second sample liquid level image. The liquid level ratio of the second sample liquid level image is greater than the liquid level ratio in the first sample liquid level image before cutting. Compared with the solution of cutting at any position of the liquid column area, such a solution can increase the liquid level ratio. In other words, it can increase the proportion of the liquid column area in the range area, which can make the liquid level recognition model obtained by subsequent training more sensitive to changes in the height of the liquid column, which is beneficial to improving the performance of the liquid level recognition model. Especially when the reading variation range of the liquid column type meter is small, the accuracy of the reading recognition can still be ensured.

[0062] In the specific implementation of step S102, the preset recognition model can be trained with the sample liquid level image set, and when the preset training conditions are met, the liquid level recognition model can be obtained. The preset recognition model can be any existing network model with learning ability, and this embodiment does not limit the type and structure of the preset recognition model.

[0063] In a specific implementation, a sample device image set can also be used to train a preset detection model to obtain an instrument detection model. Specifically, when the preset training conditions are met, an instrument detection model can be obtained. The images in the sample device image set have pre-labeled labels, which can be used to indicate the position of the liquid column type instrument in the sample device image. Therefore, the instrument detection model obtained by training can be used to detect the liquid column type instrument in the image. Among them, the preset detection model can be various existing network models with learning capabilities, and this embodiment does not limit the type and structure of the preset recognition model.

[0064] More specifically, the trained instrument detection model can be used to detect the area where the liquid column type instrument is located in the image, and the trained liquid level recognition model can be used to recognize the liquid level ratio in the area of ​​the liquid column type instrument.

[0065] In a non-limiting example, the preset detection model can be a backbone network (Backbone Network), for example, it can be a residual network (ResNets), a visual geometry group (Visual Geometry Group, VGG) network, etc.; the preset recognition model can be a prediction head network (Head Network), for example, it can be a YOLO network, RetinaNet and FasterR-CNN, etc.

[0066] It should be noted that the embodiments of the present invention do not limit the process of training a preset recognition model using a sample liquid level image set and the process of training a preset detection model using a sample device image set. It can be an existing appropriate model training method, for example, it can be a gradient descent method, etc., but it is not limited to this.

[0067] In the specific implementation of step S103, an image to be detected may be obtained, and the image to be detected may include an image of a device to be identified. A liquid column type instrument to be identified is provided on the device to be identified. It should be noted that the device to be identified and the target device in the first sample device image above may be the same device or different devices; the liquid column type instrument to be identified and the liquid column type instrument in the sample liquid level image above may be the same instrument or different instruments, and this embodiment does not limit this.

[0068] Furthermore, the image to be detected can be input into the instrument detection model. Since the instrument detection model can be used to detect liquid column type instruments, the position of the liquid column type instrument to be identified in the image to be detected can be determined according to the detection result of the instrument detection model, so as to obtain the liquid level image to be identified. In other words, the liquid level image to be identified contains the image of the liquid column type instrument to be identified.

[0069] In the specific implementation of step S104, the liquid level image to be identified can be input into the liquid level recognition model, and the liquid level ratio output by the liquid level recognition model is obtained. It can be understood that the liquid level ratio output by the liquid level recognition model refers to the ratio of the liquid column area in the liquid level image to be identified to the range area in the scale direction.

[0070] In the specific implementation of step S105, the reading of the liquid column type instrument to be identified can be determined according to the liquid level ratio output by the liquid level recognition model and the measuring range of the liquid column type instrument to be identified. Specifically, the measuring range of the liquid column type instrument to be identified can be preset, and the product of the output liquid level ratio and the measuring range can be calculated to obtain the reading of the liquid column type instrument to be identified.

[0071] As described above, in the solution of the embodiment of the present invention, it is not necessary to identify the scale line and reading of the liquid level in the liquid level image to be identified, which reduces the difficulty of identification and is conducive to improving the accuracy of the reading. In addition, this process does not require manual participation, which is also conducive to improving the efficiency of identifying the reading.

[0072] In a specific implementation, it is also possible to further determine whether the reading of the liquid column type meter to be identified exceeds a preset value range. If so, an alarm signal can be issued to remind the staff to conduct inspections.

[0073] Reference Figure 6 , Figure 6 1 is a schematic diagram of another method for identifying the reading of a liquid column type meter according to an embodiment of the present invention. Figure 6 , a non-limiting description of the training process of the instrument detection model and the liquid level recognition model in this embodiment. Figure 6 As shown, in the solution of this embodiment, the target device can be photographed in the target scene to obtain multiple first sample device images 61. The target device is provided with a liquid column type instrument, and the first sample device image 61 includes an image of the liquid column type instrument.

[0074] It is understandable that, in addition to the image of the liquid column type instrument, the first sample device image 61 also includes images of the target device and other devices in the target scene. The background and light of the first sample device image 61 are very complex, and the image quality is low. More specifically, the target device is large in size, and the area covered by the liquid column type instrument in the first sample device image 61 is small.

[0075] Further, data enhancement processing may be performed on the plurality of first sample device images 61 to obtain second sample device images, thereby obtaining a sample device image set 62, wherein the sample device image set 62 may include the first sample device image 61 and the second sample device image. For the specific content of data enhancement processing on the plurality of first sample device images 61, reference may be made to the above description of step S402, which will not be repeated here.

[0076] Furthermore, the sample device image set 62 may be processed to obtain a sample liquid level image set 64 .

[0077] Specifically, the area where the liquid column type instrument is located in the image of the sample device image set 62 can be intercepted to obtain multiple first sample liquid level images 63. Furthermore, the multiple first sample liquid level images 63 are enhanced to obtain second sample liquid level images, thereby obtaining a sample liquid level image set 64. The sample liquid level image set 64 may include the first sample liquid level image 63 and the second sample liquid level image. For the specific content of enhancing the multiple first sample liquid level images 63, please refer to the above Figure 1 The relevant description will not be repeated here.

[0078] As described above, the area covered by the liquid column type meter in the first sample device image 61 is relatively small. Compared with the solution of directly enhancing the first sample device image for training a preset recognition model, the solution in this embodiment can remove other information except the liquid column type meter in the image, so that the performance of the trained liquid level recognition model 612 is better.

[0079] Furthermore, on one hand, the preset detection model 601 can be trained with the sample device image set 62 to obtain the instrument detection model 611; on the other hand, the preset recognition model 602 can be trained with the sample liquid level image set 64 to obtain the liquid level recognition model 612. In the process of reading recognition, the instrument detection model 611 can be used to detect the liquid column type instrument in the image to be detected to obtain the liquid level image to be recognized. In other words, the liquid level image to be recognized is the area where the image of the liquid column type instrument in the image to be detected is located. Then, the liquid level recognition model 612 is used to recognize the liquid level ratio in the liquid level image to be recognized to obtain a specific reading.

[0080] In a specific implementation, the number of first sample device images 61 captured is usually limited. In order to enhance the performance of the model and more accurately identify the readings of the liquid column type meter in the target scene, it is necessary not only to increase the amount of training data, but also to ensure the authenticity of the training data. In the solution provided in this embodiment, the first sample device images 61 are subjected to data enhancement processing to obtain a sample device image set 62 for training a preset detection model 601; and the sample device image set 62 is further intercepted, enhanced, etc. to obtain a sample liquid level image set 64 for training a preset recognition model 602. Therefore, in the solution of this embodiment, the meter detection model 611 and the liquid level recognition model 612 can be trained based on the limited first sample device images 61, so that the meter detection model 611 and the liquid level recognition model 612 can be used to obtain accurate readings of the liquid column type meter.

[0081] Reference Figure 7 , Figure 7 It is a structural schematic diagram of a reading recognition device of a liquid column type meter in an embodiment of the present invention. Figure 7 The reading identification device shown may include:

[0082] The sample acquisition module 71 is used to acquire a plurality of first sample liquid level images, wherein the plurality of first sample liquid level images include a range area, wherein the range area is divided into a liquid column area and a non-liquid column area by liquid level, wherein the range area is an area covered by the range of the liquid column type instrument, and each first sample liquid level image has a pre-marked liquid level ratio, wherein the liquid level ratio refers to the ratio of the liquid column area to the range area in the scale direction;

[0083] A recognition training module 72, configured to train a preset recognition model using a sample liquid level image set to obtain a liquid level recognition model, wherein the sample liquid level image set includes the plurality of first sample liquid level images;

[0084] An image acquisition module 73, used to acquire a liquid level image to be identified, wherein the liquid level image to be identified includes an image of a liquid column type instrument to be identified;

[0085] A ratio determination module 74, used for inputting the liquid level image to be identified into the liquid level recognition model to obtain the liquid level ratio output by the liquid level recognition model;

[0086] The identification module 75 is used to determine the reading of the liquid column type meter to be identified according to the liquid level ratio output by the liquid level identification model and the measuring range of the liquid column type meter to be identified.

[0087] For more information about the working principle, working method, beneficial effects, etc. of the reading recognition device of the liquid column type meter in the embodiment of the present invention, please refer to the above description of the reading recognition method of the liquid column type meter, which will not be repeated here.

[0088] The embodiment of the present invention further provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for identifying the reading of the liquid column type meter are executed. The storage medium may include ROM, RAM, a magnetic disk or an optical disk, etc. The storage medium may also include a non-volatile memory or a non-transitory memory, etc.

[0089] The embodiment of the present invention further provides a terminal, including a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and the processor executes the steps of the above-mentioned method for identifying the reading of the liquid column meter when running the computer program. The terminal includes but is not limited to terminal devices such as mobile phones, computers, and tablet computers.

[0090] It should be understood that in the embodiments of the present application, the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0091] It should also be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0092] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer program may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless means.

[0093] In the several embodiments provided in the present application, it should be understood that the disclosed methods, devices and systems can be implemented in other ways. For example, the device embodiments described above are merely schematic; for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation; for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0094] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may be physically included separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units. For example, for each device or product applied to or integrated in a chip, each module / unit contained therein may be implemented in the form of hardware such as circuits, or at least part of the modules / units may be implemented in the form of software programs, which run on a processor integrated inside the chip, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits; for each device or product applied to or integrated in a chip module, each module / unit contained therein may be implemented in the form of hardware such as circuits, and different modules / units may be located in the same component (such as a chip, circuit module, etc.) or different components of the chip module, or at least part of the modules / units may be implemented in the form of software programs. The element can be implemented in the form of a software program, which runs on a processor integrated inside the chip module, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits; for various devices and products applied to or integrated in the terminal, the various modules / units contained therein can be implemented in the form of hardware such as circuits, and different modules / units can be located in the same component (for example, chip, circuit module, etc.) or in different components in the terminal, or, at least some modules / units can be implemented in the form of a software program, which runs on a processor integrated inside the terminal, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits.

[0095] It should be understood that the term "and / or" in this article is only a description of the association relationship of the 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 indicates that the associated objects before and after are in an "or" relationship.

[0096] The "multiple" in the embodiments of the present application refers to two or more. The first, second, etc. descriptions in the embodiments of the present application are only used for illustration and distinction of the described objects, and have no order, nor do they represent a special limitation on the number of devices in the embodiments of the present application, and cannot constitute any limitation on the embodiments of the present application.

[0097] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the scope defined by the claims.

Claims

1. A method for identifying the reading of a liquid column type meter, characterized in that: The method comprises: Acquire a plurality of first sample liquid level images, the plurality of first sample liquid level images comprising a range area, the range area being divided into a liquid column area and a non-liquid column area by liquid level, the range area being an area covered by the range of the liquid column type instrument, each first sample liquid level image having a pre-marked liquid level ratio, the liquid level ratio being the ratio of the liquid column area to the range area in the scale direction; train a preset recognition model using a sample liquid level image set to obtain a liquid level recognition model, the sample liquid level image set comprising the plurality of first sample liquid level images; Acquire a liquid level image to be identified, wherein the liquid level image to be identified includes an image of a liquid column type instrument to be identified; Inputting the to-be-recognized liquid level image into the liquid level recognition model to obtain the liquid level ratio output by the liquid level recognition model; Determining the reading of the liquid column type instrument to be identified according to the liquid level ratio output by the liquid level identification model and the measuring range of the liquid column type instrument to be identified; The sample liquid level image set further includes a plurality of second sample liquid level images, the liquid level ratio of the second sample liquid level images is greater than the liquid level ratio of the corresponding first sample liquid level images, and the second sample liquid level images are obtained by cutting the plurality of first sample liquid level images.

2. The method for identifying the reading of a liquid column meter according to claim 1, characterized in that: Cutting the plurality of first sample liquid level images includes: The plurality of first sample liquid level images are cut at any position in the non-liquid column area to obtain the plurality of second sample liquid level images.

3. The method for identifying the reading of a liquid column meter according to claim 1, characterized in that: Acquiring a plurality of first sample liquid level images includes: Acquire a plurality of first sample device images, the plurality of first sample device images including an image of a target device, the liquid column type meter being disposed on the target device; Performing data enhancement on the plurality of first sample device images to obtain a plurality of second sample device images; The images in the sample device image set are intercepted to obtain the plurality of first sample liquid level images, wherein the sample device image set includes the plurality of first sample device images and the plurality of second sample device images.

4. The method for identifying the reading of a liquid column meter according to claim 3, characterized in that: The data enhancement is selected from one or more of the following: brightness transformation, chroma transformation and saturation transformation.

5. The method for identifying the reading of a liquid column meter according to claim 3, characterized in that: The images in the sample device image set have pre-labeled labels, and the labels are used to indicate the positions of the liquid column type instruments in the images. Before acquiring the image of the liquid level meter to be identified, the method further includes: The sample device image set is used to train a preset detection model to obtain an instrument detection model.

6. The method for identifying the reading of a liquid column meter according to claim 5, characterized in that: The liquid column type instrument to be identified is arranged on the device to be identified, and obtaining the liquid level image to be identified includes: Acquire an image to be detected, wherein the image to be detected includes an image of the device to be identified; The image to be detected is input into the instrument detection model to obtain the liquid level image to be identified.

7. A liquid level identification device for a liquid column type instrument, characterized in that: The device comprises: A sample acquisition module, used to acquire a plurality of first sample liquid level images, wherein the plurality of first sample liquid level images include a range area, wherein the range area is divided into a liquid column area and a non-liquid column area by liquid level, wherein the range area is an area covered by the range of the liquid column type instrument, and each first sample liquid level image has a pre-marked liquid level ratio, wherein the liquid level ratio refers to the ratio of the liquid column area to the range area in the scale direction; A recognition training module, used for training a preset recognition model with a sample liquid level image set to obtain a liquid level recognition model, wherein the sample liquid level image set includes the plurality of first sample liquid level images; an image acquisition module, used for acquiring a liquid level image to be recognized, wherein the liquid level image to be recognized includes an image of a liquid column type instrument to be recognized; A ratio determination module, used for inputting the liquid level image to be identified into the liquid level recognition model to obtain the liquid level ratio output by the liquid level recognition model; An identification module, used to determine the reading of the liquid column type meter to be identified according to the liquid level ratio output by the liquid level identification model and the measuring range of the liquid column type meter to be identified; The sample liquid level image set further includes a plurality of second sample liquid level images, the liquid level ratio of the second sample liquid level images is greater than the liquid level ratio of the corresponding first sample liquid level images, and the second sample liquid level images are obtained by cutting the plurality of first sample liquid level images.

8. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the liquid level identification method of a liquid column type meter according to any one of claims 1 to 6 are executed.

9. A terminal comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, characterized in that: When the processor runs the computer program, the processor executes the steps of the liquid level identification method of the liquid column type meter according to any one of claims 1 to 6.

Images