CN112861861A - Method and device for identifying nixie tube text and electronic equipment - Google Patents
Method and device for identifying nixie tube text and electronic equipment Download PDFInfo
- Publication number
- CN112861861A CN112861861A CN202110053318.6A CN202110053318A CN112861861A CN 112861861 A CN112861861 A CN 112861861A CN 202110053318 A CN202110053318 A CN 202110053318A CN 112861861 A CN112861861 A CN 112861861A
- Authority
- CN
- China
- Prior art keywords
- text
- image
- model
- equipment
- nixie tube
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000010586 diagram Methods 0.000 claims abstract description 56
- 238000012545 processing Methods 0.000 claims abstract description 31
- 238000009432 framing Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 238000004806 packaging method and process Methods 0.000 claims description 4
- 230000003628 erosive effect Effects 0.000 claims description 3
- 230000010339 dilation Effects 0.000 claims description 2
- 238000012015 optical character recognition Methods 0.000 description 19
- 230000000694 effects Effects 0.000 description 6
- 238000012549 training Methods 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 102100032202 Cornulin Human genes 0.000 description 1
- 101000920981 Homo sapiens Cornulin Proteins 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000007639 printing Methods 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
Abstract
The application relates to a method, a device and electronic equipment for identifying a nixie tube text, belonging to the technical field of character identification, wherein the method for identifying the nixie tube text comprises the steps of acquiring a field acquisition image of target equipment; selecting an equipment model diagram matched with the target equipment from a pre-constructed model diagram library, and identifying and deducting the field collected image according to the equipment model diagram to obtain an image of an area to be identified; based on the image of the area to be recognized, adopting a pre-constructed and trained nixie tube text recognition model to perform text recognition to obtain a text recognition result; and performing matching combination processing according to the text attribute configuration information associated with the equipment model diagram and the text recognition result, and combining to obtain structured data as a final recognition result. The method and the device better realize the identification of the nixie tube display.
Description
Technical Field
The application belongs to the technical field of character recognition, and particularly relates to a method and a device for recognizing a nixie tube text and electronic equipment.
Background
Optical character recognition OCR generally refers to a process of inspecting characters printed on paper using an electronic device (scanner, digital camera, etc.), and translating them into characters by detecting brightness and shape using a character recognition technique. Tesseract-OCR is widely used in conventional OCR recognition applications, and the software was developed by Hewlett-packard company Bristol laboratories in 1984 and 1994, and was originally used as a text recognition engine for Hewlett-packard flat panel scanners. The first one got a great deal of attention in the 1995 UNLV OCR character recognition accuracy test. OCR markets were abandoned by hewlett packard after 1994 to stop development. Hewlett-packard contributed Tesseract-OCR to the open source community in 2005. Google obtains the source code and starts to perform function expansion and optimization on the source code.
In a complex scene, OCR recognition (such as commodity photo brand recognition, webpage information recognition, automatic driving guideboard recognition, standard certificate recognition, license plate recognition and the like) is carried out, and a core algorithm of the OCR recognition mainly comprises three parts, namely text detection, character segmentation and character recognition (part of a neural network does not need character segmentation).
The file detection method comprises graphic imaging positioning and machine learning positioning, wherein the graphic positioning comprises color positioning, texture positioning, edge detection and mathematical morphology, but the graphic imaging positioning method is easily interfered by external interference information to cause positioning failure. For example, in a positioning method of color analysis, if the background color of the license plate is similar to the color of the license plate, the license plate is difficult to extract from the background; in the edge detection method, the target edge is easy to be contaminated, so that the positioning is easy to fail. The positioning algorithm can be deceived by the interference of the external interference information, so that the positioning algorithm generates excessive wrong target candidate areas to be identified, and the system load is increased. The text character segmentation license plate and the standard certificate generally adopt a vertical projection method, because the projection of characters in the vertical direction is bound to obtain the vicinity of a local minimum value at a gap between characters or in the characters, and the position should meet the character writing format, characters, size limitation and other conditions of the license plate, the vertical projection method has a good effect on character segmentation in the automobile image in a complex environment. The character recognition method mainly comprises a template matching algorithm and an artificial neural network algorithm, wherein the template matching algorithm firstly matches the segmented characters with all templates, and finally selects the best matching as a result. There are two algorithms for artificial neuron networks: firstly, splitting a text character into single characters, and taking the characters as input training neural network distributors to realize recognition; the other method is that the whole text characters are directly transmitted into a trained neural network, and the network realizes the fast recognition of the whole text through feature extraction, the method has wide application, and the network structure is as follows: CRNN, CNN + ctcorcr, DenseNet + CTC, and the like.
As mentioned above, conventional and complex scenario OCR have been well solved and implemented in practical applications, but in a specific scenario, the recognition of nixie tube display text (including text displayed in a nixie tube font) on a display screen of a device in the related art is not ideal.
Specifically, the traditional OCR mainly aims at the character recognition of the type of the printing paper, and has the advantages that for a simple scene, the difference between characters and a background is large, and the effect of a binaryzation scene is obvious; in practical application scenes, interference factors are many, an original picture contains the whole equipment and an operation environment, the original picture is influenced by various factors such as brightness, angle and color during identification, irrelevant information is often identified as useful information, a character target cannot be effectively extracted, and an existing model does not support seven-segment nixie tube character identification.
The OCR recognition function of the complex scene supports seven-segment nixie tube target positioning, but no algorithm model support exists, a large amount of time is needed for designing, training and optimizing the network, a large amount of full equipment types are needed for training, effective marking pictures of scenes with different light, shade, angles, colors and the like are needed, a large amount of manpower and material resources are needed for the part of work, the work cannot be provided on site, the trained network needs to be continuously verified and optimized, when new types of equipment are added, the network needs to be retrained and verified and optimized, and when the new types of equipment need to be rapidly developed, the part of work cannot be estimated. Meanwhile, the method has the defect that the existing model does not support the recognition of the seven-segment nixie tube text, so that the OCR function in the complex scene is supported, but a large amount of work is required from the design realization to the optimization, the recognition effect cannot be ensured, the recognition result is a segment of text, and no data attribute service information exists.
The above is only for the purpose of assisting understanding of the technical aspects of the present invention, and does not represent an admission that the above is prior art.
Disclosure of Invention
In order to overcome the problems in the related art at least to a certain extent, the application provides a method, a device and electronic equipment for identifying a nixie tube text, which are beneficial to avoiding the defects in the prior art and better realize the identification of nixie tube display.
In order to achieve the purpose, the following technical scheme is adopted in the application:
in a first aspect,
the application provides a method for identifying a nixie tube text, which comprises the following steps:
acquiring a field acquisition image of target equipment;
selecting an equipment model diagram matched with the target equipment from a pre-constructed model diagram library, and identifying and deducting the field collected image according to the equipment model diagram to obtain an image of an area to be identified;
based on the image of the area to be recognized, adopting a pre-constructed and trained nixie tube text recognition model to perform text recognition to obtain a text recognition result;
and performing matching combination processing according to the text attribute configuration information associated with the device model diagram and the text recognition result, and combining to obtain structured data as a final recognition result.
Optionally, the selecting a device model diagram matched with the target device from a pre-constructed model diagram library specifically includes:
and selecting an equipment model diagram matched with the target equipment from a pre-constructed model diagram library according to the model information of the target equipment contained in the field acquisition image.
Optionally, the identifying and deducting the field collected image according to the device model map to obtain an image of an area to be identified specifically includes:
and extracting feature points from the equipment model map by adopting a model map matching algorithm, identifying and deducting the field collected image according to the feature points, and correcting the size and the angle of the identified and deducted area image to obtain the area image to be identified.
Optionally, the feature points include corner points, edge points, and bright points of dark areas.
Optionally, the performing text recognition by using a pre-constructed and trained nixie tube text recognition model based on the image of the area to be recognized to obtain a text recognition result includes:
determining and extracting a text unit image from the region image to be identified according to text position configuration information associated with the device model diagram;
and preprocessing the text unit image, and performing text recognition by adopting the nixie tube text recognition module according to the processed image to obtain the text recognition result.
Optionally, the preprocessing includes performing binarization processing, dilation processing, and erosion optimization processing on the image.
Optionally, the process of pre-building the model gallery includes:
collecting equipment pictures, and capturing pictures which contain screen display content areas and simultaneously extend a certain area outwards from the equipment pictures as equipment model pictures;
marking each equipment model diagram by a display unit, framing an image area capable of displaying the maximum value, and generating a configuration file associated with the corresponding equipment model diagram according to the position information obtained by framing and the attribute information corresponding to the framed area;
and taking the model number of the equipment as an identification field, and warehousing each equipment model diagram and the corresponding associated configuration file thereof to obtain the model diagram library.
Optionally, based on a text recognition model in Tesseract-OCR, performing nixie tube font custom packaging on the model to realize construction of the nixie tube text recognition model.
In a second aspect of the present invention,
the application provides a device for identifying a nixie tube text, which comprises:
the acquisition module is used for acquiring a field acquisition image of the target equipment;
the first identification processing module is used for selecting an equipment model diagram matched with the target equipment from a pre-constructed model diagram library, and identifying and deducting the field collected image according to the equipment model diagram to obtain an image of an area to be identified;
the second recognition processing module is used for recognizing the text by adopting a pre-constructed and trained nixie tube text recognition model based on the image of the area to be recognized to obtain a text recognition result;
and the combined processing module is used for carrying out matching and combining processing according to the text attribute configuration information associated with the equipment model diagram and the text recognition result, and combining the text attribute configuration information and the text recognition result to obtain structured data serving as a final recognition result.
In a third aspect,
the application provides an electronic device, including:
a memory having an executable program stored thereon;
a processor for executing the executable program in the memory to implement the steps of the method described above.
This application adopts above technical scheme, possesses following beneficial effect at least:
according to the technical scheme, in the text recognition process, the model image is firstly adopted to match and recognize the field collected image to determine the image of the area to be recognized, and then subsequent recognition processing is carried out, so that the realization difficulty of text position recognition in the whole realization is reduced. The text attribute configuration associated with the model diagram is established to structure the identification text data, so that the usability of the data is ensured. And the user-defined identification model is adopted, so that the identification accuracy of the display text of the nixie tube is guaranteed.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
The accompanying drawings are included to provide a further understanding of the technology or prior art of the present application and are incorporated in and constitute a part of this specification. The drawings expressing the embodiments of the present application are used for explaining the technical solutions of the present application, and should not be construed as limiting the technical solutions of the present application.
FIG. 1 is a schematic flow chart diagram illustrating a method for identifying nixie tube text according to an embodiment of the present application;
FIG. 2 is a schematic illustration of a device model diagram according to an embodiment of the present application;
FIG. 3 is a schematic illustration of the labeling of a display unit of a device model map during the construction of a model gallery according to an embodiment of the present application;
FIG. 4 is a schematic structural diagram of an apparatus for identifying nixie tube text according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail below. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the examples given herein without making any creative effort, shall fall within the protection scope of the present application.
As described in the background art, in the related art, the effect of identifying the nixie tube display text on the device display screen is not ideal, and for this reason, the present application provides a method for identifying the nixie tube text, which is helpful to better identify the nixie tube display.
In one embodiment, as shown in fig. 1, the method for identifying a nixie tube text provided by the present application includes the following steps:
and step S110, acquiring a field acquisition image of the target equipment.
For example, the application scenario of this embodiment is equipment inspection, and the field captured image is an image obtained by an inspector taking a picture of a target equipment (e.g., a power cabinet of a certain device) through an image capturing device (e.g., a PDA).
And step S120, selecting an equipment model diagram matched with the target equipment from a pre-constructed model diagram library, and identifying and deducting the on-site acquired image according to the equipment model diagram to obtain an image of the area to be identified.
Specifically, in step S120, an equipment model diagram matched with the target equipment is selected from a pre-constructed model diagram library according to the model information of the target equipment contained in the field acquired image; for example, the name of an image file bearing the field collected image information contains the model code of the target device, and the model information of the target device is obtained through analyzing the name of the image file, and then the device model map corresponding to the target device is retrieved and selected from the model map library. It should be noted that the device model map in this application refers to a picture (as an example shown in fig. 2) that includes a content area displayed on a device screen and is expanded outward by a certain area.
After the equipment model image is selected, the on-site collected image is matched and identified and deducted according to the equipment model image, and the position positioning of the target to be identified is favorably and efficiently realized.
In step S120 of this embodiment, a model map matching algorithm (for example, an SIFT matching algorithm) is used, feature points (generally, the feature points include corner points, edge points, bright points in dark regions, and the like) are extracted from an equipment model map, an image collected in the field is identified and deducted according to the feature points, and the size and the angle of the image in the identified and deducted region are corrected, so as to obtain an image of the region to be identified.
The method and the device have the advantages that the characteristic points extracted from the model map are used as references for matching and matting, heavy marked data preparation is avoided, new equipment types are identified, new equipment model maps are added, and secondary training is not needed.
Then, step S130 is carried out, based on the image of the area to be recognized obtained in step S120, a pre-constructed and trained nixie tube text recognition model is adopted for text recognition, and a text recognition result is obtained;
in the embodiment, the model is subjected to nixie tube font self-defining packaging based on a text recognition model in Tesseract-OCR to realize the construction of the nixie tube text recognition model.
Firstly, the Tesseract-OCR recognition principle is introduced, the Tesseract-OCR recognition step is roughly divided into four steps,
the first step is as follows: analyzing the connected region, detecting a character region (contour outline) and a sub-contour, and integrating contour lines into a block region at the stage;
the second step is that: text lines (text lines) are derived from the character outline and the block region. There are two methods of analyzing text lines, fixed scene and scaled scene. The fixed scene is divided into single characters through character units, and the Proportional scene (rendering text) is divided through clear spaces and fuzzy spaces (fuzzy spaces):
the third step: each character is analyzed and recognized in sequence, a self-adaptive classifier is used, the classifier has learning capacity, the characters meeting the conditions are analyzed and simultaneously used as training samples, so that the characters (such as page tails) which are farther back are recognized more accurately, the character recognition accuracy of page heads is lower, and the realization algorithm can perform secondary recognition on the characters which are not well recognized again so as to improve the recognition accuracy, so that the step has two times of processing;
the fourth step: resolving ambiguous spaces, examining x-height, positioned (small-cap) text, and using other methods for recognition.
In the prior art, the Tesseract-OCR officially provides a text recognition model of Chinese, English, number and the like, but does not have a recognition model of seven-segment nixie tube fonts. The method comprises the following steps of performing custom packaging based on a text recognition model in Tesseract-OCR to obtain a required model:
specifically, the optimized picture (nixie tube font display) is packaged through a tool, a single text position and a text value are marked, the text position marking requirement comprises a complete text, the distance between a frame and the text picture is as small as possible, unnecessary abnormal information is reduced, and the identification accuracy is guaranteed.
And after the text recognition model is packaged, the test can be used. And subsequently, according to the actual use effect, an abnormal recognition text marking object is added in the model, so that the model recognition effect is enhanced.
Preferably, in step S130, a text unit image is determined and extracted from the image of the region to be recognized according to text position configuration information associated with the device model map (the configuration information is generated when the model map library is constructed, and relevant contents are described in detail later), where the text unit image refers to a character display region image (e.g., a frame selection region shown in fig. 3) in the image of the region to be recognized;
and then, preprocessing the text unit image, for example, performing binarization processing, expansion processing, erosion optimization processing and the like on the image, and performing text recognition by adopting a nixie tube text recognition module according to the processed image to obtain a text recognition result.
By adding the text unit extraction function on the basis of the text position, the object identification is further accurate, the identification information amount is reduced, the result is prevented from being influenced by abnormal information, the subsequent establishment of the data and business relationship is facilitated, the result data structuring is facilitated, and the logic processing flow is simplified. And through the preprocessing of the text unit image, the different colors and background pictures are converted into black and white pictures, so that the influence of noise point information on the identification precision is further reduced.
And continuing to return to fig. 1, and after the step S130, performing a step S140, performing matching and combining processing according to the text attribute configuration information and the text recognition result associated with the device model diagram, and combining the result to obtain structured data as a final recognition result.
The text recognition result obtained in step S130 is only the text number symbol displayed by the nixie tube, and the specific meaning thereof is unknown from the data processing perspective. In step S140, based on the text attribute configuration information associated with the device model map, the text recognition result and the corresponding data attribute are combined, and the combined structure is used as the final recognition result.
For example, if the text recognition result for a certain text unit image is "56", the text unit image corresponds to the first line of real-time temperature display item in the model map, and the combined structured data is the real-time temperature 56 degrees.
According to the technical scheme, in the text recognition process, the model image is firstly adopted to match and recognize the field collected image to determine the image of the area to be recognized, and then subsequent recognition processing is carried out, so that the realization difficulty of text position recognition in the whole realization is reduced. The text attribute configuration associated with the model diagram is established to structure the identification text data, so that the usability of the data is ensured. And the user-defined identification model is adopted, so that the identification accuracy of the display text of the nixie tube is guaranteed.
Next, how to construct the model gallery in advance in the technical solution of the present application will be described.
Firstly, collecting related device pictures, wherein each device has at least one picture, and intercepting a region to be identified in the picture by using picture processing software to be used as a device model picture (the model picture is required to contain a complete region of screen display content, and simultaneously, a certain region is outwards expanded, and exclusive characteristic information is contained except the screen display region so as to obtain more characteristic information positioning targets by using an actual application algorithm).
Then, marking each equipment model diagram by a display unit, framing an image area capable of displaying the maximum value, and generating a configuration file associated with the corresponding equipment model diagram according to the position information obtained by framing and the attribute information corresponding to the framed area;
the display unit is marked because in practice, if the picture extracted from the text target is directly identified, the text identification is directly carried out because the picture contains all texts of the equipment, the equipment is different and the layout is different, so that the abnormal information is more and the identification difficulty is high; and the recognized result is a section of text, which lacks data attributes, and if judged according to code logic, standard specifications cannot be defined.
Based on the above, in order to reduce the recognition difficulty and construct a structured recognition result, in the process of constructing the model gallery, text unit information labeling is performed on each model map, self-developed marking software is used for framing and selecting the position information (shown in fig. 3) of the recognition object, the framing and selecting range must include the range capable of displaying the maximum image, when the extreme value is displayed, the recognition range is not lost, the attribute information of the object name and the data type is created at the same time, and the structured data model (which can be realized based on the configuration file) of the device display information is established, so that the subsequent business processing is facilitated.
After marking is finished, storing the model diagram and the configuration file according to a certain format, establishing an association relationship through code logic, and extracting text units, names and data type information through position information in the configuration file when the type of equipment is requested to be identified, so that a structured text identification result is conveniently established;
for example, the model of the device may be used as the identification field, and each device model diagram and its corresponding associated configuration file are stored in a library to obtain a model diagram library.
Fig. 4 is a schematic structural diagram of an apparatus 400 for identifying a nixie tube text according to an embodiment of the present application, where as shown in fig. 4, the apparatus 400 for identifying a nixie tube text includes:
an obtaining module 401, configured to obtain a field acquisition image of a target device;
the first identification processing module 402 is configured to select an equipment model map matched with the target equipment from a pre-constructed model map library, and perform identification deduction on the field acquired image according to the equipment model map to obtain an image of the area to be identified;
the second recognition processing module 403 is configured to perform text recognition by using a pre-constructed and trained nixie tube text recognition model based on the image of the area to be recognized, so as to obtain a text recognition result;
and the combination processing module 404 is configured to perform matching combination processing according to the text attribute configuration information and the text recognition result associated with the device model diagram, and obtain structured data as a final recognition result by combination.
With respect to the apparatus 400 for identifying the nixie tube text in the above-described related embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be described in detail here.
Fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application, and as shown in fig. 5, the electronic device 500 includes:
a memory 501 on which an executable program is stored;
a processor 502 for executing the executable program in the memory 501 to implement the steps of the above-described method.
With respect to the electronic device 500 in the above embodiment, the specific manner of executing the program in the memory 501 by the processor 502 thereof has been described in detail in the embodiment related to the method, and will not be elaborated herein.
The above description is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A method of identifying a nixie tube text, comprising:
acquiring a field acquisition image of target equipment;
selecting an equipment model diagram matched with the target equipment from a pre-constructed model diagram library, and identifying and deducting the field collected image according to the equipment model diagram to obtain an image of an area to be identified;
based on the image of the area to be recognized, adopting a pre-constructed and trained nixie tube text recognition model to perform text recognition to obtain a text recognition result;
and performing matching combination processing according to the text attribute configuration information associated with the device model diagram and the text recognition result, and combining to obtain structured data as a final recognition result.
2. The method according to claim 1, wherein the device model map matching the target device is selected from a pre-constructed model map library, specifically:
and selecting an equipment model diagram matched with the target equipment from a pre-constructed model diagram library according to the model information of the target equipment contained in the field acquisition image.
3. The method according to claim 1, wherein the identifying and deducting the on-site collected image according to the device model map to obtain an image of an area to be identified specifically comprises:
and extracting feature points from the equipment model map by adopting a model map matching algorithm, identifying and deducting the field collected image according to the feature points, and correcting the size and the angle of the identified and deducted area image to obtain the area image to be identified.
4. The method of claim 3, wherein the feature points comprise corner points, edge points, and bright points of dark areas.
5. The method as claimed in claim 1, wherein the obtaining of the text recognition result by performing text recognition based on the image of the area to be recognized by using a pre-constructed and trained nixie tube text recognition model comprises:
determining and extracting a text unit image from the region image to be identified according to text position configuration information associated with the device model diagram;
and preprocessing the text unit image, and performing text recognition by adopting the nixie tube text recognition module according to the processed image to obtain the text recognition result.
6. The method according to claim 5, wherein the preprocessing comprises performing binarization processing, dilation processing, erosion optimization processing on the image.
7. The method according to any one of claims 1 to 6, wherein the process of pre-building a model gallery comprises:
collecting equipment pictures, and capturing pictures which contain screen display content areas and simultaneously extend a certain area outwards from the equipment pictures as equipment model pictures;
marking each equipment model diagram by a display unit, framing an image area capable of displaying the maximum value, and generating a configuration file associated with the corresponding equipment model diagram according to the position information obtained by framing and the attribute information corresponding to the framed area;
and taking the model number of the equipment as an identification field, and warehousing each equipment model diagram and the corresponding associated configuration file thereof to obtain the model diagram library.
8. The method as claimed in any one of claims 1 to 6, wherein the construction of the nixie tube text recognition model is realized by performing nixie tube font custom packaging on the model based on the text recognition model in Tesseract-OCR.
9. An apparatus for identifying nixie tube text, comprising:
the acquisition module is used for acquiring a field acquisition image of the target equipment;
the first identification processing module is used for selecting an equipment model diagram matched with the target equipment from a pre-constructed model diagram library, and identifying and deducting the field collected image according to the equipment model diagram to obtain an image of an area to be identified;
the second recognition processing module is used for recognizing the text by adopting a pre-constructed and trained nixie tube text recognition model based on the image of the area to be recognized to obtain a text recognition result;
and the combined processing module is used for carrying out matching and combining processing according to the text attribute configuration information associated with the equipment model diagram and the text recognition result, and combining the text attribute configuration information and the text recognition result to obtain structured data serving as a final recognition result.
10. An electronic device, comprising:
a memory having an executable program stored thereon;
a processor for executing the executable program in the memory to implement the steps of the method of any one of claims 1-8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110053318.6A CN112861861B (en) | 2021-01-15 | 2021-01-15 | Method and device for recognizing nixie tube text and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110053318.6A CN112861861B (en) | 2021-01-15 | 2021-01-15 | Method and device for recognizing nixie tube text and electronic equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112861861A true CN112861861A (en) | 2021-05-28 |
CN112861861B CN112861861B (en) | 2024-04-09 |
Family
ID=76006554
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110053318.6A Active CN112861861B (en) | 2021-01-15 | 2021-01-15 | Method and device for recognizing nixie tube text and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112861861B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113610082A (en) * | 2021-08-12 | 2021-11-05 | 北京有竹居网络技术有限公司 | Character recognition method and related equipment thereof |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE2146074A1 (en) * | 1970-09-21 | 1972-03-30 | Searle Medidata Inc | Data interpretation terminal |
CN103065146A (en) * | 2012-12-24 | 2013-04-24 | 广东电网公司电力调度控制中心 | Character recognition method for power communication machine room dumb equipment signboards |
CN103984930A (en) * | 2014-05-21 | 2014-08-13 | 南京航空航天大学 | Digital meter recognition system and method based on vision |
CN109919014A (en) * | 2019-01-28 | 2019-06-21 | 平安科技(深圳)有限公司 | OCR recognition methods and its electronic equipment |
WO2019184524A1 (en) * | 2018-03-27 | 2019-10-03 | 杭州欧镭激光技术有限公司 | Detection system and detection method for detecting vehicle external environment information |
CN111582262A (en) * | 2020-05-07 | 2020-08-25 | 京源中科科技股份有限公司 | Segment type liquid crystal picture content identification method, device, equipment and storage medium |
CN111738264A (en) * | 2020-05-08 | 2020-10-02 | 上海允登信息科技有限公司 | Intelligent acquisition method for data of display panel of machine room equipment |
CN111832565A (en) * | 2020-07-24 | 2020-10-27 | 桂林电子科技大学 | Decision tree-based nixie tube identification method |
-
2021
- 2021-01-15 CN CN202110053318.6A patent/CN112861861B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE2146074A1 (en) * | 1970-09-21 | 1972-03-30 | Searle Medidata Inc | Data interpretation terminal |
CN103065146A (en) * | 2012-12-24 | 2013-04-24 | 广东电网公司电力调度控制中心 | Character recognition method for power communication machine room dumb equipment signboards |
CN103984930A (en) * | 2014-05-21 | 2014-08-13 | 南京航空航天大学 | Digital meter recognition system and method based on vision |
WO2019184524A1 (en) * | 2018-03-27 | 2019-10-03 | 杭州欧镭激光技术有限公司 | Detection system and detection method for detecting vehicle external environment information |
CN109919014A (en) * | 2019-01-28 | 2019-06-21 | 平安科技(深圳)有限公司 | OCR recognition methods and its electronic equipment |
CN111582262A (en) * | 2020-05-07 | 2020-08-25 | 京源中科科技股份有限公司 | Segment type liquid crystal picture content identification method, device, equipment and storage medium |
CN111738264A (en) * | 2020-05-08 | 2020-10-02 | 上海允登信息科技有限公司 | Intelligent acquisition method for data of display panel of machine room equipment |
CN111832565A (en) * | 2020-07-24 | 2020-10-27 | 桂林电子科技大学 | Decision tree-based nixie tube identification method |
Non-Patent Citations (2)
Title |
---|
卢卫娜;刘长荣;郑玉才;王海芳;: "一种基于模板匹配的数字仪表字符识别方法", 现代计算机(专业版), no. 03, pages 70 - 72 * |
郭爽;: "数码管数字仪表自动识别方法的研究", 通信技术, no. 08, pages 91 - 93 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113610082A (en) * | 2021-08-12 | 2021-11-05 | 北京有竹居网络技术有限公司 | Character recognition method and related equipment thereof |
Also Published As
Publication number | Publication date |
---|---|
CN112861861B (en) | 2024-04-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110766014B (en) | Bill information positioning method, system and computer readable storage medium | |
CA3027038C (en) | Document field detection and parsing | |
CN109948510B (en) | Document image instance segmentation method and device | |
US10817741B2 (en) | Word segmentation system, method and device | |
CN109753953B (en) | Method and device for positioning text in image, electronic equipment and storage medium | |
CN108764352B (en) | Method and device for detecting repeated page content | |
CN110874618B (en) | OCR template learning method and device based on small sample, electronic equipment and medium | |
CN111753692A (en) | Target object extraction method, product detection method, device, computer and medium | |
RU2640322C2 (en) | Methods and systems of effective automatic recognition of symbols | |
CN110598566A (en) | Image processing method, device, terminal and computer readable storage medium | |
CN114663904A (en) | PDF document layout detection method, device, equipment and medium | |
CN111915635A (en) | Test question analysis information generation method and system supporting self-examination paper marking | |
CN111259891A (en) | Method, device, equipment and medium for identifying identity card in natural scene | |
CN110533660B (en) | Method for detecting silk-screen defect of electronic product shell | |
CN115937887A (en) | Method and device for extracting document structured information, electronic equipment and storage medium | |
CN113392819B (en) | Batch academic image automatic segmentation and labeling device and method | |
CN111626145A (en) | Simple and effective incomplete form identification and page-crossing splicing method | |
CN114463767A (en) | Credit card identification method, device, computer equipment and storage medium | |
CN112861861B (en) | Method and device for recognizing nixie tube text and electronic equipment | |
CN111414889B (en) | Financial statement identification method and device based on character identification | |
CN115546219B (en) | Detection plate type generation method, plate card defect detection method, device and product | |
KR20170010753A (en) | Method for the optical detection of symbols | |
CN112200789A (en) | Image identification method and device, electronic equipment and storage medium | |
CN111832551A (en) | Text image processing method and device, electronic scanning equipment and storage medium | |
CN111008635A (en) | OCR-based multi-bill automatic identification method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |