CN111353502A

CN111353502A - Digital table identification method and device and electronic equipment

Info

Publication number: CN111353502A
Application number: CN202010118061.3A
Authority: CN
Inventors: 张晨光
Original assignee: Beijing Elitenect Technologies Co ltd
Current assignee: Beijing Elitenect Technologies Co ltd
Priority date: 2020-02-25
Filing date: 2020-02-25
Publication date: 2020-06-30
Anticipated expiration: 2040-02-25
Also published as: CN111353502B

Abstract

The application provides a digital table identification method, a digital table identification device and electronic equipment, relates to the technical field of image identification, and solves the technical problem that the integrity of the currently identified digital representation number is low. The method comprises the following steps: acquiring a digital table image; determining a region to be identified containing text in the digital table image, wherein the text comprises a reading and a physical unit; wherein the physical unit closest to the reading in the digital table image is the physical unit corresponding to the reading; and identifying the text in the area to be identified to obtain a number identification result and a physical unit identification result corresponding to the number identification result.

Description

Digital table identification method and device and electronic equipment

Technical Field

The present application relates to the field of image recognition technologies, and in particular, to a method and an apparatus for recognizing a digital table, and an electronic device.

Background

With the development of scientific technology, digital watches are becoming more and more popular as an important data display tool.

At present, in the existing digital table recognition and reading process, an extracted digital area is divided, and then, the number of the divided image is recognized, so that the number of the digital table is finally obtained. However, the digital representation obtained by this method has a low degree of completeness and has a small meaning relative to the digital representation.

Disclosure of Invention

The invention aims to provide a digital table identification method, a digital table identification device and electronic equipment, which are used for solving the technical problem that the integrity of the currently identified digital representation number is low.

In a first aspect, an embodiment of the present application provides a method for identifying a digital table, where the method includes:

acquiring a digital table image;

determining a region to be identified containing text in the digital table image, wherein the text comprises a reading and a physical unit; wherein the physical unit closest to the reading in the digital table image is the physical unit corresponding to the reading;

and identifying the text in the area to be identified to obtain a number identification result and a physical unit identification result corresponding to the number identification result.

In one possible implementation, the digital table image includes a plurality of digital table pictures that are successively acquired; the method further comprises the following steps:

acquiring a plurality of the number identification results corresponding to the same physical unit from the plurality of the digital table pictures;

and determining a final reading result corresponding to the same physical unit based on a plurality of reading identification results.

In one possible implementation, the step of determining a final reading result corresponding to the same physical unit based on a plurality of reading recognition results includes:

filtering the registration identification result exceeding a preset numerical range from the plurality of registration identification results to obtain a normal registration identification result;

and calculating the average value of the plurality of normal reading recognition results to obtain a final reading result.

determining the last acquired target reading identification result in the reading identification results corresponding to the same physical unit according to the acquisition time sequence;

and determining the target reading identification result as a final reading result corresponding to the same physical unit.

In one possible implementation, the step of obtaining the digital table image includes:

acquiring an original image, wherein the original image comprises a digital table;

and detecting the position and the size of the digital table in the original image by using a target detection neural network model to obtain a digital table image.

In one possible implementation, the target detection neural network model is a single neural network-based target detection (young Only Look one, Yolo for short) neural network model.

In one possible implementation, the text further includes dial characters of the digital table, the dial characters being used to represent information of the digital table;

the step of recognizing the text in the area to be recognized to obtain a number recognition result and a physical unit recognition result corresponding to the number recognition result includes:

identifying the text in the area to be identified to obtain a text identification result;

and screening the text recognition results, taking the text recognition results in the number indication form as the number indication recognition results, determining the text recognition results in the physical unit form closest to the number indication recognition results as the physical unit recognition results corresponding to the number indication recognition results, and determining the text recognition results except the number indication recognition results and the physical unit recognition results as the dial character recognition results.

In a second aspect, there is provided a digital table identifying apparatus comprising:

the acquisition module is used for acquiring a digital table image;

the determining module is used for determining a region to be identified containing a text in the digital table image, wherein the text comprises a reading and a physical unit; wherein the physical unit closest to the reading in the digital table image is the physical unit corresponding to the reading;

and the recognition module is used for recognizing the text in the area to be recognized to obtain a number recognition result and a physical unit recognition result corresponding to the number recognition result.

In a third aspect, an embodiment of the present application further provides an electronic device, which includes a memory and a processor, where the memory stores a computer program that is executable on the processor, and the processor implements the method of the first aspect when executing the computer program.

In a fourth aspect, this embodiment of the present application further provides a computer-readable storage medium storing machine executable instructions, which, when invoked and executed by a processor, cause the processor to perform the method of the first aspect.

The embodiment of the application brings the following beneficial effects:

according to the digital table identification method, the digital table identification device and the electronic equipment, after a digital table image is obtained, an area to be identified containing a text can be determined in the digital table image, wherein the text comprises a number and a physical unit, the physical unit closest to the number in the digital table image is the physical unit corresponding to the number, the text in the area to be identified is identified to obtain a number identification result and a physical unit identification result corresponding to the number identification result, and in the scheme, the number identification result and the physical unit identification result corresponding to the number identification result can be obtained by identifying the area to be identified including the number and the physical unit in the digital table image. Therefore, when the system identifies and reads the reading of the digital meter, the reading of the digital meter can be read, the reading of the physical unit corresponding to the reading can be read, namely, the physical meaning represented by the reading, the reading result is more complete through the reading with the physical unit, the more complete reading of the digital meter is finally read, and the technical problem that the integrity of the currently identified digital reading is lower is solved.

In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the detailed description of the present application or the technical solutions in the prior art, the drawings needed to be used in the detailed description of the present application or the prior art description will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative efforts.

FIG. 1 is a flow chart diagram illustrating a method for identifying a numeric table according to an embodiment of the present application;

fig. 2 is a schematic view illustrating a processing flow of a plurality of continuously acquired digital table pictures in the digital table recognition method according to an embodiment of the present application;

FIG. 3 is a schematic view of another flowchart of a method for identifying a numeric table according to an embodiment of the present application;

FIG. 4 is a diagram of an example of a digital table image provided by an embodiment of the present application;

FIG. 5 is a schematic structural diagram of a digital table identifying apparatus according to an embodiment of the present application;

fig. 6 is a schematic structural diagram illustrating an electronic device provided in an embodiment of the present application.

Detailed Description

To make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The terms "comprising" and "having," and any variations thereof, as referred to in the embodiments of the present application, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements but may alternatively include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

At present, the traditional mechanical meter is gradually replaced by an intelligent digital display meter which can display physical quantities to be monitored and can also transmit readings to a control system. However, due to the network, the environment of the electric meter box, the technology, and the like, the smart digital display meter cannot transmit the readings to the control system, or even if the readings can be transmitted to the control system, the smart digital display meter may have great instability. Therefore, manual meter reading is sometimes necessary. However, when the number of meters is large, a large amount of data acquisition work needs a certain amount of manpower, and a meter reading error may occur in manual meter reading.

In recent years, with the wide application of artificial neural network technology, especially deep learning methods, the work with extremely high repeatability can be processed by artificial intelligence instead of manpower, so that the manpower is reduced, and the cost is reduced. The digital display meter is identified by using a deep learning method, and the aim of reading the display number of the meter can be realized only by acquiring an instrument image and training. Compared with the traditional image processing method, the deep learning method has higher robustness and adaptability, and even if the environmental factors are changeable, the counting number of the recognition table of the deep learning method is still accurate and reliable.

The existing digital table identification processing process comprises the following steps:

pretreatment: and extracting all straight lines in the electric meter image by using a computer vision method, then carrying out inclination correction on the electric meter image through the straight lines, then carrying out distortion removal processing on the electric meter image and unifying the size of the picture.

Digital area detection: and detecting the digital region in the preprocessed image by using the pre-trained deep neural network, and extracting the digital region in the meter.

Reading the reading number: and segmenting the extracted digital region, and then, identifying the reading by using a deep neural network to finally obtain the reading of the meter.

However, this method has the following problems:

first, the image needs to be preprocessed using conventional image processing methods. The meter image may not be effectively corrected due to factors such as illumination, installation, etc., which may affect subsequent identification.

The identification is performed after dividing the indicated number area, that is, only the indicated number is identified, and information such as a decimal point and an indicated number unit is not identified, but the information has a great significance for the indicated number. However, the size of these characters is usually different from the scale, and they cannot be read by image segmentation together with the scale region in general. Thus, only the reading of the indicator is presently read without physical units, and without the physical meaning that the indicator represents. Such that reading of readings without physical units is incomplete.

In addition, at present, the method is static number meter identification, and the method does not read the number of the meter with the changed number. And a part of the numbers indicate that the numbers are alternately displayed.

Based on this, the embodiment of the application provides a digital table identification method, a digital table identification device and electronic equipment, and the method can solve the technical problem that the integrity of the currently identified digital representation number is low.

Embodiments of the present invention are further described below with reference to the accompanying drawings.

Fig. 1 is a flowchart illustrating a digital table identification method according to an embodiment of the present application. As shown in fig. 1, the method includes:

step S110, a digital table image is acquired.

It should be noted that the digital meter image includes a digital meter, and the digital meter may be any digital meter capable of detecting and displaying various physical quantities, such as an electricity meter, a pressure meter, and a thermometer. Wherein the various physical quantities may comprise a plurality of different meter values, such as current values, voltage values, power values, temperature values, pressure values, etc.

Step S120, determining a region to be identified containing a text in the digital table image, wherein the text comprises a reading number and a physical unit.

Wherein, the physical unit closest to the index in the digital table image is the physical unit corresponding to the index. Of course, the text may also include content other than the number and physical units, such as the number of the dial and the like.

Step S130, recognizing the text in the region to be recognized to obtain a number recognition result and a physical unit recognition result corresponding to the number recognition result.

By identifying the region to be identified including the index and the physical unit in the digital table image, the index identification result and the physical unit identification result corresponding to the index identification result can be obtained. Therefore, when the system identifies and reads the reading of the digital meter, the reading including the decimal point displayed by the meter can be read, the physical unit corresponding to the reading, namely the physical meaning represented by the reading, can be read, the reading result is more complete through the reading with the physical unit, and finally the more complete reading of the digital meter is read.

In the embodiment of the application, the meter reading is read in a deep learning artificial intelligence mode, a novel digital meter reading method is provided, the meter reading is performed by using the artificial intelligence deep learning method to replace an artificial mode, and the method is more robust and higher in accuracy.

The above steps are described in detail below.

In some embodiments, the digital table image comprises a plurality of digital table pictures acquired in succession; the method may further comprise the steps of:

step a), acquiring a plurality of reading identification results corresponding to the same physical unit from a plurality of digital table pictures;

and b), determining a final reading result corresponding to the same physical unit based on the plurality of reading identification results.

For example, after the meter recognition algorithm reads the readings in one frame of the meter picture, the meter image can be continuously read, and the readings and the physical units in the meter image can be read. As shown in fig. 2, if there is only one physical unit in the meter reading mechanism and after a period of time, assuming that there is only one physical unit after 12 seconds, the reading of the meter is considered to be completed; and if the read meter structure has at least two physical units, and the currently read physical unit is the same as one of the previously read physical units of the meter reading, all the readings of the meter are considered to be finished, and the meter reading is finished and all the readings are returned.

For the specific process of the reading traversal judgment, for example, as shown in fig. 2, the meter identification network can continuously read the clear meter pictures taken by the camera and store the reading (including the reading and the physical unit) read each time. If only one read physical unit is found and reading of the readings continues for a period of time, all the readings of the current meter are read, and reading results of the readings are returned; if it is found that at least two different physical units have been read previously and the physical unit of the current reading is the same as one of the physical units previously read, then all readings of the current meter are considered to have been read. At this point, reading of the readings is stopped, and all reading results of the readings are returned.

The prior art does not read the number of the number table of the cycle display number. The digital meter identification method provided by the application is different from a method for reading only the meter reading number during shooting, and can read the number and the physical unit thereof aiming at the digital meter which circularly displays the number in reality, thereby improving the reading efficiency of the digital meter.

In some embodiments, step b) above may include a variety of implementations. As an example, the step b) may include the steps of:

step c), filtering the registration identification results exceeding the preset numerical range from the plurality of registration identification results to obtain normal registration identification results;

and d), calculating the average value of the plurality of normal reading recognition results to obtain a final reading result.

By calculating the average value of the filtered normal reading recognition results, the accuracy and precision of the finally obtained final reading result corresponding to the same physical unit can be improved.

As another example, the step b) may include the steps of:

step e), determining the last acquired target reading identification result in a plurality of reading identification results corresponding to the same physical unit according to the acquisition time sequence;

and f), determining the target reading identification result as a final reading result corresponding to the same physical unit.

By determining the last acquired target index identification result as the final index result corresponding to the physical unit, the numerical value detected by the digital table when the previous data state is unstable can be automatically ignored, and the accurate numerical value after the final stable state can be directly obtained.

In some embodiments, the step S110 may include the following steps:

step g), obtaining an original image, wherein the original image comprises a digital table;

and h), detecting the position and the size of the digital table in the original image by using the target detection neural network model to obtain a digital table image.

In the embodiment of the application, a clear digital meter picture is assumed to be obtained, at the moment, the algorithm calls the target detection algorithm to frame out the meter, then the algorithm calls the meter reading number detection and identification network to read out the reading number and the unit in the meter, and the reading result is returned to the robot.

Illustratively, as shown in fig. 3, the digital meter recognition network intercepts a meter region picture from an original picture of a camera video stream based on meter position information output from a meter detection network, and then reads the reading in the meter picture. After the meter detection network detects the meter, the picture of the meter is extracted, and the meter identification network is called. For the extracted meter picture, for example, as shown in fig. 4, the display contents are: 283.28 is KPA; 28.324 for Range.

Of course, the neural network model for meter detection may be a neural network model for target detection in various forms to realize the function of meter detection and detect the position information and the size information of the meter in the picture.

The detection and the identification of the meter are realized by depending on a deep learning artificial neural network method, the steps of preprocessing an image by using a traditional machine vision processing method and the like are not involved in the process, the complex preprocessing process can be saved to improve the speed of the identification process, and the method has strong adaptability to the inclination of the meter and the change of the environment and improves the identification efficiency of the digital meter.

Based on the step h), the target detection neural network model is a Yolo neural network model.

In practical application, the target detection neural network model may be based on a Yolo network in a deep learning convolutional neural network, as shown in fig. 3, based on a meter detection network formed by improving and perfecting the Yolo network. The YOLO is an object recognition and positioning algorithm based on a deep neural network.

For the training process of the target detection neural network model, a data set used by a training improved network is adopted as a meter picture which is automatically collected on site. Data acquisition is required for all meters with meter reading requirements. After the acquisition is completed, the data needs to be labeled to be used as a data set for subsequent training.

Then, aiming at the training Yolo network, the labeled data set is used for training, parameters (such as learning rate, the number of samples selected by one training and the like) required by the training network are set, and after a certain number of iterations, the model converges, so that the training of the target detection neural network model is successful.

By using the Yolo network, the trained network model is integrated into the meter recognition algorithm, a program calling interface is compiled, and when the position of the meter needs to be detected, the model only needs to be called to carry out reasoning every time. Compared with the original network, the inference speed is higher when the network scale is smaller.

In some embodiments, the text further includes dial characters of the digital watch, the dial characters being used to represent information of the digital watch; the step S130 may include the following steps:

step i), recognizing the text in the region to be recognized to obtain a text recognition result;

and j), screening the text recognition results, taking the text recognition results in the number indication form as the number indication recognition results, determining the text recognition results in the physical unit form closest to the number indication recognition results as the physical unit recognition results corresponding to the number indication recognition results, and determining the text recognition results except the number indication recognition results and the physical unit recognition results as the dial character recognition results.

In practical application, besides reading the number of the meter, other texts on the digital dial plate can be read, such as text information of manufacturer, model, type and the like.

For a meter recognition network model executing a digital meter recognition process, the network model training process is similar to that of a target detection neural network model, a data set is also a meter sample acquired on site, and then the number and text information on the meter are labeled. In the training process of the model, the model can be trained by using marked data containing the position of the text region on the digital meter and the content of the text region.

And for the use process of the meter identification network model, calling the meter identification model to perform text detection and identification after the meter detection network detects the meter picture meeting the conditions each time. The recognition result may include text region information and text content information. It should be noted that, for the meter number identification network, the common text detection and identification network may also achieve the effect described in the embodiment of the present application.

As shown in fig. 3, a meter identification network may be invoked, which internally includes two convolutional neural network modules: a text detection network and a text recognition network. The meter identification network can firstly detect the meter picture, detect text areas (including numbers, units and other texts on the dial) in the meter picture, and then identify the content of the text areas as a preliminary identification result by the text identification network.

Next, after obtaining preliminary recognition results, these results are filtered by an algorithm as shown in fig. 3. Firstly, extracting the display number, then searching the text which represents the physical unit near the display area, and selecting the text which is closest to the display number as the unit of the display number. Text other than the units of the reading and the reading may be omitted or used as a specific identifier for the meter.

For the screening process of the identification result, the result output by the meter identification network is only the text and the position thereof on the meter panel, and no information about the meaning represented by the text is available. It is necessary to screen these identifications to obtain the physical meaning represented by the representation.

Of course, the result output by the meter reading recognition network does not necessarily need to screen the reading before searching for the physical unit when matching the reading with the physical unit, and conversely, the physical unit may be searched first and then the meter reading is deduced.

The method provided by the embodiment of the application can optimize the network structure, and for one meter, the processing time for reading the network reasoning time and the identification result by the number is within 300 milliseconds. By using the deep learning convolutional neural network, even in a variable field environment, such as influence factors as weather change, shooting angle change, dial plate stain and the like, higher identification accuracy can be maintained, such as accuracy of more than 98%.

Fig. 5 provides a schematic structural diagram of a digital table identifying device. As shown in fig. 5, the number table recognition apparatus 500 includes:

a first obtaining module 501, configured to obtain a digital table image;

a first determining module 502, configured to determine, in the digital table image, an area to be identified that includes a text, where the text includes a reading and a physical unit; wherein, the physical unit closest to the index in the digital table image is the physical unit corresponding to the index;

the identifying module 503 is configured to identify the text in the area to be identified, so as to obtain a number identification result and a physical unit identification result corresponding to the number identification result.

In some embodiments, the digital table image comprises a plurality of digital table pictures acquired in succession; the device also includes:

the second acquisition module is used for acquiring a plurality of reading identification results corresponding to the same physical unit from the plurality of digital table pictures;

and the second determining module is used for determining a final reading result corresponding to the same physical unit based on the plurality of reading identification results.

In some embodiments, the second determining module is specifically configured to:

filtering the registration identification results exceeding a preset numerical range from the plurality of registration identification results to obtain normal registration identification results;

determining the last acquired target reading identification result in a plurality of reading identification results corresponding to the same physical unit according to the acquisition time sequence;

In some embodiments, the first obtaining module 501 is specifically configured to:

and detecting the position and the size of the digital table in the original image by using the target detection neural network model to obtain a digital table image.

In some embodiments, the target detection neural network model is a Yolo neural network model.

In some embodiments, the text further includes dial characters of the digital watch, the dial characters being used to represent information of the digital watch; the identification module 503 is specifically configured to:

The digital table identification device provided by the embodiment of the application has the same technical characteristics as the digital table identification method provided by the embodiment, so that the same technical problems can be solved, and the same technical effects can be achieved.

As shown in fig. 6, an electronic device 600 provided in an embodiment of the present application includes: a processor 601, a memory 602 and a bus, wherein the memory 602 stores machine-readable instructions executable by the processor 601, when the electronic device is operated, the processor 601 and the memory 602 communicate with each other through the bus, and the processor 601 executes the machine-readable instructions to execute the steps of the digital table identification method.

Specifically, the memory 602 and the processor 601 can be general-purpose memories and processors, which are not specifically limited herein, and the digital table identifying method can be performed when the processor 601 runs a computer program stored in the memory 602.

The processor 601 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware or instructions in the form of software in the processor 601. The Processor 601 may be a general-purpose Processor, and includes a Central Processing Unit (CPU), a Network Processor (NP), and the like; the device can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, or a discrete hardware component. The various methods, steps, and logic blocks disclosed in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The storage medium is located in the memory 602, and the processor 601 reads the information in the memory 602 and completes the steps of the method in combination with the hardware thereof.

Corresponding to the digital table identification method, the embodiment of the application also provides a computer readable storage medium, wherein the computer readable storage medium stores machine executable instructions, and when the computer executable instructions are called and executed by a processor, the computer executable instructions cause the processor to execute the steps of the digital table identification method.

The digital table identifying device provided by the embodiment of the application can be specific hardware on the device or software or firmware installed on the device. The device provided by the embodiment of the present application has the same implementation principle and technical effect as the foregoing method embodiments, and for the sake of brief description, reference may be made to the corresponding contents in the foregoing method embodiments where no part of the device embodiments is mentioned. It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the foregoing systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.

For another example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments provided in the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the digital table identifying method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus once an item is defined in one figure, it need not be further defined and explained in subsequent figures, and moreover, the terms "first", "second", "third", etc. are used merely to distinguish one description from another and are not to be construed as indicating or implying relative importance.

Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the scope of the embodiments of the present application. Are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for digital table identification, the method comprising:

acquiring a digital table image;

2. The method of claim 1, wherein the digital table image comprises a plurality of digital table pictures acquired in succession; the method further comprises the following steps:

3. The method of claim 2, wherein the step of determining a final reading result corresponding to the same physical unit based on a plurality of the reading recognition results comprises:

4. The method of claim 2, wherein the step of determining a final reading result corresponding to the same physical unit based on a plurality of the reading recognition results comprises:

5. The method of claim 1, wherein the step of obtaining the digital table image comprises:

6. The method of claim 5, wherein the target detection neural network model is a Yolo neural network model.

7. The method of claim 1, wherein the text further comprises a dial word of the digital watch, the dial word being used to represent information of the digital watch;

8. A digital table identifying apparatus, comprising:

the acquisition module is used for acquiring a digital table image;

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program operable on the processor, and wherein the processor implements the steps of the method of any of claims 1 to 7 when executing the computer program.

10. A computer readable storage medium having stored thereon machine executable instructions which, when invoked and executed by a processor, cause the processor to execute the method of any of claims 1 to 7.