Automatic reading method and device of float flowmeter, terminal equipment and storage medium

Abstract

The invention discloses an automatic reading method and device for a float flowmeter, terminal equipment and a storage medium, and the method comprises the steps: obtaining a to-be-recognized equipment image and the style type of the float flowmeter in the to-be-recognized equipment image; according to the to-be-identified equipment image and the style type, determining a float flowmeter area image in the to-be-identified equipment image through a deep learning image identification technology; according to the float flowmeter area image and the style type, determining float position information in the float flowmeter area image through a deep learning image recognition technology; according to the area image of the float flowmeter, scale information in the area image of the float flowmeter is determined through a deep learning image recognition technology; and the reading of the float flowmeter is determined according to the float position information and the scale information. According to the automatic reading method for the float flowmeter, the reading of the float flowmeter can be automatically read, manpower is saved, and the reading efficiency and accuracy of the float flowmeter are improved.

Description

Technical Field

[0001] The invention relates to the technical field of float flowmeter reading, and in particular to a float flowmeter automatic reading method, device, terminal equipment and storage medium. Background Art

[0002] With the development of automation construction in the industrial field, inspection robots, remote video monitoring and other means have begun to gradually replace relevant staff to complete the reading and recognition of various instruments, thereby creating a more accurate and safe automated industrial environment. Float flowmeter is also one of the instruments that need to be monitored in real time. It is widely used in industrial process control, energy management, water treatment and supply, laboratory research and other fields, and is used to measure the flow of various liquid / gas media. At present, the reading of float flowmeter is mainly completed by the staff's human eye recognition. The human eye is prone to fatigue, low efficiency, and unstable accuracy. It cannot ensure the reliability of the reading results and easily reduces the accuracy of the work. Summary of the invention

[0003] The present invention provides a float flowmeter automatic reading method, device, terminal equipment and storage medium to solve the technical problem that the prior art relies on manual reading resulting in low result accuracy.

[0004] In order to solve the above technical problems, an embodiment of the present invention provides a method for automatically reading a float flowmeter, comprising:

[0005] Acquire an image of a device to be identified and a type of a float flowmeter in the image of the device to be identified;

[0006] According to the image of the device to be identified and the type of pattern, a float flowmeter area image in the image of the device to be identified is determined by deep learning image recognition technology;

[0007] According to the float flowmeter area image and the style type, the float position information in the float flowmeter area image is determined by deep learning image recognition technology;

[0008] According to the float flowmeter area image, the scale information in the float flowmeter area image is determined by deep learning image recognition technology;

[0009] The reading of the float flowmeter is determined according to the float position information and the scale information.

[0010] As a preferred solution, the method of determining the float flowmeter area image in the image of the device to be identified by deep learning image recognition technology according to the image of the device to be identified and the type of pattern includes:

[0011] According to the style type, determining whether a first training set of a trained float flowmeter target detection model contains a sample device image corresponding to a float flowmeter of the style type;

[0012] If the first training set contains sample device images corresponding to the float flowmeter of the style type, the device image to be identified is input into the float flowmeter target detection model, so that the float flowmeter target detection model identifies and outputs the float flowmeter area image in the device image to be identified; wherein the float flowmeter target detection model is a target detection model that takes the device image as input and takes the image of the float flowmeter area in the device image as output;

[0013] If the first training set does not contain a sample device image corresponding to the float flowmeter of the style type, then obtain the float flowmeter marked area image of the device image to be identified, and register the float flowmeter marked area image as the base library image of the float flowmeter matching and recognition model; input the device image to be identified into the float flowmeter matching and recognition model, so that the float flowmeter matching and recognition model recognizes and outputs the float flowmeter area image of the device image to be identified; wherein the float flowmeter matching and recognition model is a matching and recognition model that takes the device image as input and the image of the float flowmeter area in the device image as output.

[0014] As a preferred solution, the method of determining the float position information in the float flowmeter area image by deep learning image recognition technology based on the float flowmeter area image and the style type includes:

[0015] According to the style type, determining whether a second training set of the trained float target detection model contains a sample float flowmeter image corresponding to the float flowmeter of the style type;

[0016] If the second training set contains a sample float flowmeter image corresponding to the float flowmeter of the style type, the float flowmeter area image is input into the float target detection model, so that the float target detection model identifies the float area in the float flowmeter area image and outputs the coordinates of the float area as the float position information; wherein the float target detection model is a target detection model that takes the float flowmeter image as input and takes the coordinates of the float area in the float flowmeter image as output;

[0017] If the second training set does not contain a sample float flowmeter image corresponding to the float flowmeter of the style type, then obtain the float mark area image of the float flowmeter area image, and register the float mark area image as the base library picture of the float matching recognition model; input the float flowmeter area image into the float matching recognition model, so that the float matching recognition model recognizes the float area in the float flowmeter area image, and outputs the coordinates of the float area as the float position information; wherein the float matching recognition model is a matching recognition model that takes the float flowmeter image as input and takes the coordinates of the float area in the float flowmeter image as output.

[0018] As a preferred solution, the method of determining the scale information in the float flowmeter area image by deep learning image recognition technology based on the float flowmeter area image includes:

[0019] Inputting the float flowmeter area image into a trained scale target detection model so that the scale target detection model can recognize the scale area image in the float flowmeter area image; wherein the scale target detection model is a target detection model that takes the float flowmeter image as input and takes the image of the scale area in the float flowmeter image as output;

[0020] Inputting the scale area image into a trained scale value recognition model so that the scale value recognition model recognizes the scale recognition value in the scale area image; wherein the scale value recognition model is a text recognition model that takes the scale image as input and the scale value as output;

[0021] Determine a corresponding scale correction rule according to the style type;

[0022] According to the scale correction rule, the scale identification value is corrected to obtain the scale information.

[0023] As a preferred solution, the float position information is the coordinates of the lower right corner of the float area;

[0024] The scale information is the scale value corresponding to the position of each scale in the scale area image;

[0025] Determining the reading of the float flowmeter according to the float position information and the scale information includes:

[0026] According to the float position information and the scale information, two upper and lower adjacent target scales closest to the position where the lower right corner coordinate is located are determined, and the target scale coordinates and corresponding target scale values ​​of the target scale are determined;

[0027] The reading of the float flowmeter is calculated according to the coordinates of the lower right corner of the float area, the target scale coordinates and the target scale value.

[0028] As a preferred solution, the calculation formula for the reading of the float flowmeter is:

[0029]

[0030] Where, R represents the reading of the float flowmeter; K up Indicates the scale value of the target scale located above; K down Indicates the scale value of the target scale below; Y up Indicates the y-axis coordinate of the target scale located above; Y down Represents the y-axis coordinate of the target scale located above; b represents the y-axis value in the coordinate of the lower right corner of the float area.

[0031] As a preferred solution, the scale correction rule is used to standardize the possible value set of scale values ​​and scale differences; wherein the scale difference refers to the difference between adjacent scales;

[0032] The step of correcting the scale identification value according to the scale correction rule to obtain the scale information includes:

[0033] According to the scale correction rule, determining a scale identification value belonging to the possible value set as a to-be-determined valid scale value;

[0034] According to the scale coordinates of all the pending valid scale values ​​and the scale differences, target scale values ​​corresponding to all scales in the scale area image are determined, and the target scale values ​​are used as the scale information.

[0035] Based on the above embodiment, another embodiment of the present invention provides a float flowmeter automatic reading device, including: a data acquisition module, a float flowmeter identification module, a float identification module, a scale identification module and a reading calculation module;

[0036] The data acquisition module is used to acquire the image of the device to be identified and the type of the float flowmeter in the image of the device to be identified;

[0037] The float flowmeter identification module is used to determine the float flowmeter area image in the device image to be identified by deep learning image recognition technology according to the device image to be identified and the style type;

[0038] The float recognition module is used to determine the float position information in the float flowmeter area image by deep learning image recognition technology according to the float flowmeter area image and the style type;

[0039] The scale recognition module is used to determine the scale information in the float flowmeter area image by deep learning image recognition technology according to the float flowmeter area image;

[0040] The reading calculation module is used to determine the reading of the float flowmeter according to the float position information and the scale information.

[0041] Based on the above embodiments, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the automatic reading method of the float flowmeter described in the above invention embodiment is implemented.

[0042] Based on the above embodiment, another embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the float flowmeter automatic reading method described in the above invention embodiment.

[0043] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0044] The present invention obtains an image of a device to be identified and a style type of a float flowmeter in the image of the device to be identified; according to the image of the device to be identified and the style type, the float flowmeter area image in the image of the device to be identified is determined by deep learning image recognition technology; according to the float flowmeter area image and the style type, the float position information in the float flowmeter area image is determined by deep learning image recognition technology; according to the float flowmeter area image, the scale information in the float flowmeter area image is determined by deep learning image recognition technology; according to the float flowmeter area image, the float flowmeter area image is determined by deep learning image recognition technology; the float flowmeter reading is determined according to the float position information and the scale information. The float flowmeter automatic reading method provided by the present invention can automatically read the float flowmeter reading, save manpower, and improve the efficiency and accuracy of the float flowmeter reading. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flow chart of an automatic reading method of a float flowmeter provided by an embodiment of the present invention;

[0046] Figure 2 It is a device image schematic diagram of a float flowmeter;

[0047] Figure 3 A schematic diagram of a float flowmeter area image of a float flowmeter;

[0048] Figure 4A schematic diagram of a float area in a float flowmeter area image of a float flowmeter;

[0049] Figure 5 A schematic diagram of a scale area in a float flowmeter area image of a float flowmeter;

[0050] Figure 6 A schematic structural diagram of an automatic reading device for a float flowmeter provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] Embodiment 1

[0053] Please refer to Figure 1 , a method for automatically reading a float flowmeter provided by an embodiment of the present invention, comprising:

[0054] S1. Acquire an image of a device to be identified and a type of a float flowmeter in the image of the device to be identified.

[0055] It should be noted that the styles and types of float flowmeters can be finely divided according to the scale design requirements of the float flowmeter in different application scenarios. For example, a certain float flowmeter has a reading range of 100 to 1000 and a scale difference of 50.

[0056] Please refer to Figure 2 , Figure 2 This is a device image schematic diagram of a float flowmeter. The glass tube float flowmeter is suitable for two usage scenarios. The reading ranges are 40-100 and 20-190, and the corresponding scale differences are 10 and 5 respectively.

[0057] S2. According to the image of the device to be identified and the type of pattern, determine the float flowmeter area image in the image of the device to be identified by deep learning image recognition technology.

[0058] In a preferred embodiment, the method of determining the float flowmeter area image in the image of the device to be identified by deep learning image recognition technology based on the image of the device to be identified and the type of pattern includes:

[0059] According to the style type, determining whether a first training set of a trained float flowmeter target detection model contains a sample device image corresponding to a float flowmeter of the style type;

[0060] If the first training set contains sample device images corresponding to the float flowmeter of the style type, the device image to be identified is input into the float flowmeter target detection model, so that the float flowmeter target detection model identifies and outputs the float flowmeter area image in the device image to be identified; wherein the float flowmeter target detection model is a target detection model that takes the device image as input and takes the image of the float flowmeter area in the device image as output;

[0061] If the first training set does not contain a sample device image corresponding to the float flowmeter of the style type, then obtain the float flowmeter marked area image of the device image to be identified, and register the float flowmeter marked area image as the base library image of the float flowmeter matching and recognition model; input the device image to be identified into the float flowmeter matching and recognition model, so that the float flowmeter matching and recognition model recognizes and outputs the float flowmeter area image of the device image to be identified; wherein the float flowmeter matching and recognition model is a matching and recognition model that takes the device image as input and the image of the float flowmeter area in the device image as output.

[0062] Please note that, please refer to Figure 3 , Figure 3 A schematic diagram of an image of a float flowmeter area of ​​a float flowmeter.

[0063] In this embodiment, two methods for recognizing the image of the float flowmeter region in the image of the device to be recognized are provided, one is to use the float flowmeter target detection model, and the other is to use the float flowmeter matching recognition model.

[0064] 1. Use the float flowmeter target detection model: When the first training set already contains sample device images corresponding to the float flowmeter of this style, the image of the device to be identified is directly input into the float flowmeter target detection model to obtain the float flowmeter area image;

[0065] 2. Use the float flowmeter matching and recognition model: When the first training set does not contain the sample device image corresponding to the float flowmeter of this style type, manually mark the float flowmeter area in the image, and obtain the marked float flowmeter marked area image, and register the float flowmeter marked area image as the float flowmeter matching and recognition model base library image. After that, the float flowmeter matching and recognition model can retrieve the area in the image that is most similar to the registered target according to the characteristics of the registered target, that is, the matching and recognition algorithm can directly output the float flowmeter area image of the registered float flowmeter.

[0066] S3. Determine the float position information in the float flowmeter area image according to the float flowmeter area image and the style type by using deep learning image recognition technology.

[0067] In a preferred embodiment, the determining the float position information in the float flowmeter area image by deep learning image recognition technology according to the float flowmeter area image and the style type includes:

[0068] According to the style type, determining whether a second training set of the trained float target detection model contains a sample float flowmeter image corresponding to the float flowmeter of the style type;

[0069] If the second training set contains a sample float flowmeter image corresponding to the float flowmeter of the style type, the float flowmeter area image is input into the float target detection model, so that the float target detection model identifies the float area in the float flowmeter area image and outputs the coordinates of the float area as the float position information; wherein the float target detection model is a target detection model that takes the float flowmeter image as input and takes the coordinates of the float area in the float flowmeter image as output;

[0070] If the second training set does not contain a sample float flowmeter image corresponding to the float flowmeter of the style type, then obtain the float mark area image of the float flowmeter area image, and register the float mark area image as the base library picture of the float matching recognition model; input the float flowmeter area image into the float matching recognition model, so that the float matching recognition model recognizes the float area in the float flowmeter area image, and outputs the coordinates of the float area as the float position information; wherein the float matching recognition model is a matching recognition model that takes the float flowmeter image as input and takes the coordinates of the float area in the float flowmeter image as output.

[0071] Please note that, please refer to Figure 4 , Figure 4 It is a schematic diagram of a float area in a float flowmeter area image of a float flowmeter, wherein the framed portion is the float area.

[0072] In this embodiment, two methods for identifying float position information in a float flowmeter area image are provided, one is to use a float target detection model, and the other is to use a float matching recognition model.

[0073] 1. Use the float target detection model: When the second training set already contains sample float flowmeter images corresponding to the float flowmeter of this style, the float flowmeter area image is directly input into the float target detection model to obtain the float position information;

[0074] 2. Use the float matching recognition model: When the second training set does not contain a sample float flowmeter image corresponding to this style of float flowmeter, manually mark the float area in the image, obtain the marked float marked area image, and register the float flowmeter marked area image as the base library image of the float matching recognition model. After that, the float matching recognition model can retrieve the area in the image that is most similar to the registered target based on the characteristics of the registered target, that is, the matching recognition algorithm can directly output the float position information of the registered float.

[0075] S4. Determine the scale information in the float flowmeter area image by deep learning image recognition technology according to the float flowmeter area image.

[0076] In a preferred embodiment, the determining the scale information in the float flowmeter area image by deep learning image recognition technology according to the float flowmeter area image includes:

[0077] Inputting the float flowmeter area image into a trained scale target detection model so that the scale target detection model can recognize the scale area image in the float flowmeter area image; wherein the scale target detection model is a target detection model that takes the float flowmeter image as input and takes the image of the scale area in the float flowmeter image as output;

[0078] Inputting the scale area image into a trained scale value recognition model so that the scale value recognition model recognizes the scale recognition value in the scale area image; wherein the scale value recognition model is a text recognition model that takes the scale image as input and the scale value as output;

[0079] Determine a corresponding scale correction rule according to the style type;

[0080] According to the scale correction rule, the scale identification value is corrected to obtain the scale information.

[0081] Please note that, please refer to Figure 5 , Figure 5 The figure is a schematic diagram of a scale area in a float flowmeter area image of a float flowmeter, wherein the framed portion is the scale area.

[0082] In this embodiment, the float flowmeter area image is input into the scale target detection model to obtain the scale area image in the float flowmeter area image, and then the scale area image is input into the value recognition model to obtain the scale recognition value.

[0083] After obtaining the scale recognition value, since the text recognition algorithm may have scale recognition errors in complex environments such as shadows and uneven brightness, it is necessary to perform rule screening on it to obtain the correct scale information.

[0084] In a preferred embodiment, the scale correction rule is used to standardize the possible value set of scale values ​​and scale differences; wherein the scale difference refers to the difference between adjacent scales;

[0085] The step of correcting the scale identification value according to the scale correction rule to obtain the scale information includes:

[0086] According to the scale correction rule, determining a scale identification value belonging to the possible value set as a to-be-determined valid scale value;

[0087] According to the scale coordinates of all the pending valid scale values ​​and the scale differences, target scale values ​​corresponding to all scales in the scale area image are determined, and the target scale values ​​are used as the scale information.

[0088] It should be noted that if Figure 5 The scale correction rules of the example diagram are: the left scale is valid, the scale increases from bottom to top, and the scale difference is 50. Assuming that the text recognition algorithm gives a set of left scales [5*, 180, 150, 208, 250, 300, 350, 480], the real scale [50, 100, 150, 200, 250, 300, 350, 400] can be corrected according to the scale correction rules.

[0089] S5. Determine a reading of the float flowmeter according to the float position information and the scale information.

[0090] In a preferred embodiment, the float position information is the coordinates of the lower right corner of the float area;

[0091] The scale information is the scale value corresponding to the position of each scale in the scale area image;

[0092] Determining the reading of the float flowmeter according to the float position information and the scale information includes:

[0093] According to the float position information and the scale information, two upper and lower adjacent target scales closest to the position where the lower right corner coordinate is located are determined, and the target scale coordinates and corresponding target scale values ​​of the target scale are determined;

[0094] The reading of the float flowmeter is calculated according to the coordinates of the lower right corner of the float area, the target scale coordinates and the target scale value.

[0095] In a preferred embodiment, the calculation formula of the reading of the float flowmeter is:

[0096]

[0097] Where, R represents the reading of the float flowmeter; K up Indicates the scale value of the target scale located above; K down Indicates the scale value of the target scale below; Y up Indicates the y-axis coordinate of the target scale located above; Y down Represents the y-axis coordinate of the target scale located above; b represents the y-axis value in the coordinate of the lower right corner of the float area.

[0098] Embodiment 2

[0099] Please refer to Figure 2 , is a schematic structural diagram of an automatic reading device for a float flowmeter provided by an embodiment of the present invention, comprising: a data acquisition module, a float flowmeter identification module, a float identification module, a scale identification module and a reading calculation module;

[0100] The data acquisition module is used to acquire the image of the device to be identified and the type of the float flowmeter in the image of the device to be identified;

[0101] The float flowmeter identification module is used to determine the float flowmeter area image in the device image to be identified by deep learning image recognition technology according to the device image to be identified and the style type;

[0102] The float recognition module is used to determine the float position information in the float flowmeter area image by deep learning image recognition technology according to the float flowmeter area image and the style type;

[0103] The scale recognition module is used to determine the scale information in the float flowmeter area image by deep learning image recognition technology according to the float flowmeter area image;

[0104] The reading calculation module is used to determine the reading of the float flowmeter according to the float position information and the scale information.

[0105] Embodiment 3

[0106] Accordingly, an embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the automatic reading method of the float flowmeter described in the above-mentioned embodiment of the invention is implemented.

[0107] Embodiment 4

[0108] Accordingly, an embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the float flowmeter automatic reading method described in the above-mentioned embodiment of the invention.

[0109] It should be noted that the device embodiments described above are merely schematic, wherein 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 modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0110] Those skilled in the art can clearly understand that for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0111] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0112] The processor may be a central processing unit (CPU), or 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 any conventional processor, etc. The processor is the control center of the device, and various interfaces and lines are used to connect various parts of the entire device.

[0113] The memory can be used to store the computer program, and the processor realizes various functions of the device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0114] The storage medium is a storage medium, and the computer program is stored in the storage medium. When the computer program is executed by the processor, the steps of each method embodiment described above can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer readable media do not include electric carrier signals and telecommunication signals.

[0115] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for automatic reading of a float flowmeter, characterized in that: include: Acquire an image of a device to be identified and a type of a float flowmeter in the image of the device to be identified; According to the image of the device to be identified and the type of pattern, a float flowmeter area image in the image of the device to be identified is determined by deep learning image recognition technology; According to the float flowmeter area image and the style type, the float position information in the float flowmeter area image is determined by deep learning image recognition technology; According to the float flowmeter area image, the scale information in the float flowmeter area image is determined by deep learning image recognition technology; The reading of the float flowmeter is determined according to the float position information and the scale information.

2. The automatic reading method of a float flowmeter as claimed in claim 1, characterized in that: The method of determining the float flowmeter area image in the image of the device to be identified by deep learning image recognition technology according to the image of the device to be identified and the type of pattern includes: According to the style type, determining whether a first training set of a trained float flowmeter target detection model contains a sample device image corresponding to a float flowmeter of the style type; If the first training set contains sample device images corresponding to the float flowmeter of the style type, the device image to be identified is input into the float flowmeter target detection model, so that the float flowmeter target detection model identifies and outputs the float flowmeter area image in the device image to be identified; wherein the float flowmeter target detection model is a target detection model that takes the device image as input and takes the image of the float flowmeter area in the device image as output; If the first training set does not contain a sample device image corresponding to the float flowmeter of the style type, then obtain the float flowmeter marked area image of the device image to be identified, and register the float flowmeter marked area image as the base library image of the float flowmeter matching and recognition model; input the device image to be identified into the float flowmeter matching and recognition model, so that the float flowmeter matching and recognition model recognizes and outputs the float flowmeter area image of the device image to be identified; wherein the float flowmeter matching and recognition model is a matching and recognition model that takes the device image as input and the image of the float flowmeter area in the device image as output.

3. The automatic reading method of a float flowmeter according to claim 1, characterized in that: The method of determining the float position information in the float flowmeter area image by deep learning image recognition technology according to the float flowmeter area image and the style type includes: According to the style type, determining whether a second training set of the trained float target detection model contains a sample float flowmeter image corresponding to the float flowmeter of the style type; If the second training set contains a sample float flowmeter image corresponding to the float flowmeter of the style type, the float flowmeter area image is input into the float target detection model, so that the float target detection model identifies the float area in the float flowmeter area image and outputs the coordinates of the float area as the float position information; wherein the float target detection model is a target detection model that takes the float flowmeter image as input and takes the coordinates of the float area in the float flowmeter image as output; If the second training set does not contain a sample float flowmeter image corresponding to the float flowmeter of the style type, then obtain the float mark area image of the float flowmeter area image, and register the float mark area image as the base library picture of the float matching recognition model; input the float flowmeter area image into the float matching recognition model, so that the float matching recognition model recognizes the float area in the float flowmeter area image, and outputs the coordinates of the float area as the float position information; wherein the float matching recognition model is a matching recognition model that takes the float flowmeter image as input and takes the coordinates of the float area in the float flowmeter image as output.

4. The automatic reading method of a float flowmeter as claimed in claim 3, characterized in that: The method of determining the scale information in the float flowmeter area image by deep learning image recognition technology according to the float flowmeter area image includes: Inputting the float flowmeter area image into a trained scale target detection model so that the scale target detection model can recognize the scale area image in the float flowmeter area image; wherein the scale target detection model is a target detection model that takes the float flowmeter image as input and takes the image of the scale area in the float flowmeter image as output; Inputting the scale area image into a trained scale value recognition model so that the scale value recognition model recognizes the scale recognition value in the scale area image; wherein the scale value recognition model is a text recognition model that takes the scale image as input and the scale value as output; Determine a corresponding scale correction rule according to the style type; According to the scale correction rule, the scale identification value is corrected to obtain the scale information.

5. The automatic reading method of a float flowmeter as claimed in claim 4, characterized in that: The float position information is the coordinates of the lower right corner of the float area; The scale information is the scale value corresponding to the position of each scale in the scale area image; Determining the reading of the float flowmeter according to the float position information and the scale information includes: According to the float position information and the scale information, two upper and lower adjacent target scales closest to the position where the lower right corner coordinate is located are determined, and the target scale coordinates and corresponding target scale values ​​of the target scale are determined; The reading of the float flowmeter is calculated according to the coordinates of the lower right corner of the float area, the target scale coordinates and the target scale value.

6. The automatic reading method of a float flowmeter as claimed in claim 5, characterized in that: The calculation formula of the float flowmeter reading is: Where, R represents the reading of the float flowmeter; K up Indicates the scale value of the target scale located above; K down Indicates the scale value of the target scale below; Y up Indicates the y-axis coordinate of the target scale located above; Y down Represents the y-axis coordinate of the target scale located above; b represents the y-axis value in the coordinate of the lower right corner of the float area.

7. The automatic reading method of a float flowmeter as claimed in claim 4, characterized in that: The scale correction rule is used to standardize the possible value set of scale values ​​and scale differences; wherein the scale difference refers to the difference between adjacent scales; The step of correcting the scale identification value according to the scale correction rule to obtain the scale information includes: According to the scale correction rule, determining a scale identification value belonging to the possible value set as a to-be-determined valid scale value; According to the scale coordinates of all the pending valid scale values ​​and the scale differences, target scale values ​​corresponding to all scales in the scale area image are determined, and the target scale values ​​are used as the scale information.

8. An automatic reading device for a float flowmeter, characterized in that: include: Data acquisition module, float flowmeter identification module, float identification module, scale identification module and reading calculation module; The data acquisition module is used to acquire the image of the device to be identified and the type of the float flowmeter in the image of the device to be identified; The float flowmeter identification module is used to determine the float flowmeter area image in the device image to be identified by deep learning image recognition technology according to the device image to be identified and the style type; The float recognition module is used to determine the float position information in the float flowmeter area image by deep learning image recognition technology according to the float flowmeter area image and the style type; The scale recognition module is used to determine the scale information in the float flowmeter area image by deep learning image recognition technology according to the float flowmeter area image; The reading calculation module is used to determine the reading of the float flowmeter according to the float position information and the scale information.

9. A terminal device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the automatic reading method of the float flowmeter according to any one of claims 1 to 7 when executing the computer program.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is executed, the device where the storage medium is located is controlled to execute the automatic reading method of the float flowmeter according to any one of claims 1 to 7.

Images