A method and system for recognizing layered characters in optical distribution boxes.

Abstract

This invention discloses a method and system for recognizing layered characters in optical distribution boxes. The method includes: correcting an acquired image of an optical distribution box to obtain a first image at a preset angle; determining the port positions of the optical distribution box in the first image according to a preset image segmentation algorithm, and determining the coordinates of the rightmost port of the optical distribution box; using the coordinates of the rightmost port as a reference, segmenting the character region to be recognized in the first image into multiple sample images to be detected; recognizing the characters in the multiple sample images to be detected, and corresponding the recognized characters with the ports of different layers of optical distribution boxes. Even when the optical distribution box is not properly positioned, the coordinates of a boundary point can be determined, and then the ports can be recognized layer by layer. The recognized characters are then matched with the ports, achieving type recognition and position detection of characters with box structure information markings on the right side of the optical distribution box, ultimately realizing the visualization of the box structure information.

Description

Technical Field

[0001] This invention relates to the field of recognition technology, and in particular to a method and system for recognizing layered characters in optical distribution boxes. Background Technology

[0002] Optical distribution boxes, also commonly known as street-side cabinets, are typically placed on the main optical cable for cable branching. They are passive devices that use ports for data transmission and employ hierarchical characters to identify the specific location of the data's destination. The speed and accuracy of hierarchical character recognition directly determine the validity of inspection results.

[0003] There are many methods for character recognition, but due to the diverse forms of characters on optical distribution boxes, such as printed fonts, handwritten fonts, and logos combining printed and handwritten characters, it is necessary to determine which optical distribution box port layer the character belongs to. Existing recognition technologies cannot identify these characters directly. Summary of the Invention

[0004] This invention provides a method and system for recognizing layered characters on an optical distribution box, in order to solve the problem of characters being unrecognizable due to the diverse character shapes on the optical distribution box.

[0005] To solve the above-mentioned technical problems, the present invention is implemented as follows:

[0006] Firstly, a method for recognizing layered characters in optical junction boxes is provided, the method comprising:

[0007] The acquired optical distribution box image is corrected to obtain a first image at a preset angle;

[0008] Based on the preset image segmentation algorithm, the position of the optical distribution box port in the first image is determined, and the coordinates of the rightmost port of the optical distribution box are determined.

[0009] Using the coordinates of the rightmost port as a reference, the character region to be identified in the first image is divided into multiple sample images to be detected;

[0010] Characters in the multiple sample images to be detected are identified, and the identified characters are matched with the ports of different optical distribution boxes.

[0011] Secondly, a system for recognizing layered characters in optical junction boxes is provided, the system comprising:

[0012] The correction module is used to correct the acquired optical distribution box image to obtain a first image at a preset angle;

[0013] The coordinate determination module is used to determine the position of the optical distribution box port in the first image according to a preset image segmentation algorithm, and to determine the coordinates of the rightmost port of the optical distribution box body;

[0014] The segmentation module is used to segment the character region to be identified in the first image into multiple sample images to be detected, based on the coordinates of the rightmost port.

[0015] The recognition module is used to identify characters in the multiple sample images to be detected and to match the identified characters with the ports of different optical distribution boxes.

[0016] Thirdly, a terminal device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method described in the first aspect.

[0017] Fourthly, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0018] In this embodiment of the invention, the acquired optical distribution box image is first corrected to obtain a first image at a preset angle. Then, according to the image segmentation method, the position of the optical distribution box port in the first image is determined, and the coordinates of the rightmost port of the optical distribution box are determined. Using these coordinates as a reference, the character region to be identified in the first image is segmented into multiple sample images to be detected. Finally, the characters in the multiple sample images to be detected are identified, and the identified characters are matched with the ports of different layers of optical distribution boxes. Even when the optical distribution box is not placed correctly, the coordinates of a certain boundary point can be determined, and then the ports can be identified layer by layer. The identified characters are then matched with the ports, realizing type identification and position detection of characters with box structure information on the right side of the optical distribution box, and finally realizing visualization of the box structure information. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0020] Figure 1 This is a flowchart of a method for recognizing layered characters in an optical distribution box, provided by an embodiment of the present invention;

[0021] Figure 2 This is a schematic diagram of a layered character recognition system for optical distribution boxes provided in an embodiment of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] This invention provides a method and system for recognizing layered characters in optical distribution boxes. Taking optical fiber distribution boxes as the research object, a series of image preprocessing and image segmentation are performed on the optical distribution boxes using image processing technology to obtain the layered character regions to be detected in each optical distribution box. Then, an improved target detection algorithm is used to detect the character regions to be detected, thereby obtaining the character type and the position information of the corresponding port layer on the right side of each optical distribution box.

[0024] like Figure 1 The diagram shown is a flowchart of a method for recognizing layered characters in an optical distribution box according to an embodiment of the present invention. Figure 1 As shown, the method for recognizing the layered characters of the optical distribution box may include the contents shown in steps S101 to S104.

[0025] In step S101, the acquired optical distribution box image is corrected to obtain a first image at a preset angle.

[0026] In other words, the general outline of the image is first obtained by using a clustering algorithm based on the image color. Then, the corner points of the obtained outline are extracted. The optical cross-section box is corrected into a vertically placed rectangle by using an affine transformation algorithm on the corner points of the box based on the image tilt features of different optical cross-section boxes, so as to facilitate the segmentation in subsequent steps.

[0027] In step S102, the position of the optical distribution box port in the first image is determined according to the preset image segmentation algorithm, and the coordinates of the rightmost port of the optical distribution box are determined.

[0028] That is, the port position is determined according to the logical relationship between the ports by a preset image segmentation algorithm, and then a coordinate axis relative to the box is established to determine the coordinates of the rightmost port of the optical distribution box.

[0029] In step S103, the character region to be identified in the first image is segmented into multiple sample images to be detected, based on the coordinates of the rightmost port.

[0030] In other words, based on the logical relationship between the optical distribution box port and the character region, and using the coordinates of the rightmost port of the optical distribution box as a reference, the character region to be identified is segmented into image samples of a preset shape. Preferably, the image samples are segmented into images of uniform size for subsequent recognition processing.

[0031] In step S104, characters in multiple sample images to be detected are identified, and the identified characters are matched with the ports of different optical distribution boxes.

[0032] In other words, the detection and recognition results are matched with the ports of different optical distribution boxes by using the logical relationships of each character saved when the sample is obtained, thus completing the entire recognition process.

[0033] In this embodiment, the acquired optical distribution box image is first corrected to obtain a first image at a preset angle. Then, based on the image segmentation method, the port position of the optical distribution box in the first image is determined, and the coordinates of the rightmost port of the optical distribution box are determined. Using these coordinates as a reference, the character region to be identified in the first image is segmented into multiple sample images to be detected. Finally, the characters in the multiple sample images to be detected are identified, and the identified characters are matched with the ports of different layers of optical distribution boxes. Even when the optical distribution box is not properly positioned, the coordinates of a certain boundary point can be determined, and then the ports can be identified layer by layer. The identified characters are then matched with the ports, realizing type identification and position detection of characters with box structure information on the right side of the optical distribution box, and ultimately realizing the visualization of the box structure information.

[0034] In one possible implementation of this application, the acquired optical distribution box image is corrected to obtain a first image at a preset angle, which may include the following steps.

[0035] Obtain the box outline from the optical cross-section image; extract the box corner points of the box outline; correct the box corner points using an affine transformation algorithm to obtain a first image, wherein the first image includes a vertically placed rectangular box image.

[0036] In this embodiment, the correction of the optical distribution box image is mainly used to address situations where, during the optical distribution box shooting process, the box is not positioned correctly, resulting in the inability to accurately identify the coordinates of the rightmost port of the optical distribution box, thus failing to obtain a good region of characters to be identified. This invention first uses clustering to obtain the box portion in the image, and then combines affine transformation in image processing technology to correct the image tilt, obtaining an image of the optical distribution box to be processed where the box's length and width are on horizontal and vertical lines, respectively.

[0037] In one possible implementation of this application, obtaining the coordinates of the rightmost port of the optical distribution box mainly involves obtaining the left boundary reference of the sample image of the layered character to be detected. This invention segments the image using the YCrCb color channel of the image to be processed, and obtains the unconnected ports with red caps through threshold segmentation. Simultaneously, considering that the use of optical distribution box ports is generally from left to right, and that the majority of unconnected ports in the optical distribution box contain red caps, the relative coordinates of the rightmost port of the optical distribution box are obtained. Since an affine transformation has been performed, the image's length and width are on the horizontal and vertical lines respectively; therefore, the left boundary of the layered character to be detected is obtained using the relative horizontal coordinate of the port within the optical distribution box.

[0038] In one possible implementation of this application, the character region to be identified in the first image is segmented into multiple sample images to be detected, based on the coordinates of the rightmost port. This may include the following steps.

[0039] Using the coordinates of the rightmost port as a reference, and the port of the first preset layer number as the length and the port of the preset layer number as the width, a preset character region to be recognized is obtained; all character regions to be recognized in the first image are divided into multiple preset character regions to be recognized, and the multiple preset character regions to be recognized are multiple sample images to be detected.

[0040] In this embodiment, the segmentation of the character region to be detected on the right side of the image optical cross-section is mainly aimed at obtaining the region image of the specific layered character to be detected. This invention uses the left boundary of the aforementioned layered character to be detected as a reference, and uses the port of every three layers as the width, obtaining the right boundary of the region to be identified according to a square relationship with equal length and width, thereby obtaining an image sample of the layered character region to be detected.

[0041] In one possible implementation of this application, a fiber optic box character detection algorithm that meets engineering requirements in terms of both time response and recognition accuracy can be obtained by combining a deep learning image recognition network for feature extraction and a target detection algorithm for classification and localization.

[0042] Specifically, the character detection and recognition algorithm is designed primarily for the efficient identification and detection of layered character regions. This invention uses the layered character region image samples obtained above as the research object. It trains the samples by combining a deep learning image recognition network for feature extraction and a target detection algorithm for classification and localization. Then, the trained network structure model is used to perform target detection on the layered character regions of subsequent optical distribution box images, obtaining the character type of each target and its relative position within the optical distribution box.

[0043] In one possible implementation of this application, identifying characters in multiple sample images to be detected and corresponding the identified characters with ports of different optical distribution boxes may include the following steps.

[0044] The system identifies characters in multiple sample images to be detected; the identified characters are then matched with the ports of different optical distribution boxes in the first image to obtain the identified optical distribution box layered characters.

[0045] Specifically, the detection and output of the character region to be identified are mainly used to obtain the visual identification results of the characters of the corresponding port layer relative to the structure of the optical distribution box. This invention matches each port layer with the characters of the corresponding port layer by comparing the coordinates of the outermost red hat port and the position of the detected layered characters, and then visualizes them according to the structural information of the box.

[0046] This invention also provides a system for recognizing layered characters in optical distribution boxes. For example... Figure 2 The diagram shown is a schematic representation of a layered character recognition system for optical distribution boxes provided in an embodiment of the present invention. Figure 2 As shown, the optical cross-section layered character recognition system may include: a correction module 201, a coordinate determination module 202, a segmentation module 203, and a recognition module 204.

[0047] Specifically, the correction module 201 is used to correct the acquired optical distribution box image to obtain a first image at a preset angle; the coordinate determination module 202 is used to determine the position of the optical distribution box port in the first image according to a preset image segmentation algorithm, and to determine the coordinates of the rightmost port of the optical distribution box; the segmentation module 203 is used to segment the character region to be identified in the first image into multiple sample images to be detected based on the coordinates of the rightmost port; and the recognition module 204 is used to recognize the characters in the multiple sample images to be detected, and to match the recognized characters with the ports of different layers of optical distribution boxes.

[0048] In this embodiment, the correction module 201 first corrects the acquired optical distribution box image to obtain a first image at a preset angle. Then, the coordinate determination module 202 determines the port position of the optical distribution box in the first image according to the image segmentation method, and determines the coordinates of the rightmost port of the optical distribution box. The segmentation module 203 uses these coordinates as a reference to segment the character region to be identified in the first image into multiple sample images to be detected. Finally, the recognition module 204 identifies the characters in the multiple sample images to be detected and corresponds the identified characters to the ports of different layers of optical distribution boxes. Even when the optical distribution box is not placed correctly, the coordinates of a certain boundary point can be determined, and then the ports can be identified layer by layer. The identified characters are then matched with the ports, realizing the type recognition and position detection of characters with box structure information on the right side of the optical distribution box, and finally realizing the visualization of the box structure information.

[0049] Optionally, the correction module 201 can be used to: acquire the box outline in the optical cross-section image; extract the box corner points of the box outline; and correct the box corner points using an affine transformation algorithm to obtain a first image, wherein the first image includes a vertically placed rectangular box image.

[0050] Optionally, the segmentation module 203 can be used to: obtain a preset character region to be recognized based on the coordinates of the rightmost port, with the length of the port of the first preset layer and the width of the port of the preset layer; and segment all the character regions to be recognized in the first image into multiple preset character regions to be recognized, wherein the multiple preset character regions to be recognized are multiple sample images to be detected.

[0051] Optionally, the recognition module 204 can be used to: recognize characters in multiple sample images to be detected; and match the recognized characters with the ports of different optical distribution boxes in the first image to obtain the recognized optical distribution box layered characters.

[0052] The optical cross-connect box layered character recognition system described in this invention has already demonstrated its functionality. Figure 1 The method embodiments shown are described in detail. Therefore, for any parts not covered in detail in this embodiment, please refer to the relevant descriptions in the foregoing embodiments, which will not be repeated here.

[0053] Optionally, embodiments of the present invention also provide a terminal device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the various processes of the optical cross-section layered character recognition method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0054] Optionally, embodiments of the present invention also provide a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the above-described method for recognizing layered characters in optical distribution boxes, and achieves the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0055] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0056] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0057] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

Claims

1. A method for recognizing layered characters in an optical distribution box, characterized in that, include: The acquired optical distribution box image is corrected to obtain a first image at a preset angle, wherein the length and width of the optical distribution box are respectively on the horizontal line and the vertical line of the first image; Based on the preset image segmentation algorithm, the position of the optical distribution box port in the first image is determined, and the coordinates of the rightmost port of the optical distribution box are determined. Using the coordinates of the rightmost port as a reference, the character region to be identified in the first image is divided into multiple sample images of layered characters to be detected; Identify characters in sample images of the multiple layers of characters to be detected, and associate the identified characters with the ports of different layers of optical distribution boxes; The step of segmenting the character region to be identified in the first image into multiple sample images of layered characters to be detected, based on the coordinates of the rightmost port, includes: Using the coordinates of the rightmost port as a reference, and the length of the port at the first preset layer number as the length and the width of the port at the preset layer number as the width, a preset character region to be recognized is obtained; The first image is divided into multiple preset character regions to be identified, and the multiple preset character regions to be identified are sample images of multiple layered characters to be detected; The coordinates of the rightmost port of the optical distribution box are obtained by obtaining the left boundary reference of the sample image of the layered character to be detected, and the left boundary of the layered character to be detected is determined by the relative horizontal coordinate of the port in the optical distribution box.

2. The method according to claim 1, characterized in that, The step of correcting the acquired optical distribution box image to obtain a first image at a preset angle includes: Obtain the box outline from the optical distribution box image; Extract the corner points of the box body outline; The corner points of the box are corrected using an affine transformation algorithm to obtain a first image, wherein the first image includes an image of a vertically placed rectangular box.

3. The method according to claim 1, characterized in that, The process of identifying characters in sample images of the plurality of layered characters to be detected, and corresponding the identified characters with the ports of different optical distribution boxes, includes: Identify characters in sample images of the multiple hierarchical characters to be detected; The characters in the sample image are matched with the ports of different optical distribution boxes in the first image to obtain the recognized optical distribution box layered characters.

4. A system for recognizing layered characters in an optical distribution box, characterized in that, include: The correction module is used to correct the acquired optical distribution box image to obtain a first image at a preset angle, wherein the length and width of the optical distribution box are respectively on the horizontal line and the vertical line of the first image; The coordinate determination module is used to determine the position of the optical distribution box port in the first image according to a preset image segmentation algorithm, and to determine the coordinates of the rightmost port of the optical distribution box body; The segmentation module is used to segment the character region to be identified in the first image into multiple sample images of layered characters to be detected, based on the coordinates of the rightmost port. The recognition module is used to recognize the characters in the sample images of the multiple layered characters to be detected, and to match the recognized characters with the ports of different optical distribution boxes; The segmentation module is used for: Using the coordinates of the rightmost port as a reference, and the length of the port at the first preset layer number as the length and the width of the port at the preset layer number as the width, a preset character region to be recognized is obtained; The first image is divided into multiple preset character regions to be identified, and the multiple preset character regions to be identified are sample images of multiple layered characters to be detected; The coordinates of the rightmost port of the optical distribution box are obtained by obtaining the left boundary reference of the sample image of the layered character to be detected, and the left boundary of the layered character to be detected is determined by the relative horizontal coordinate of the port in the optical distribution box.

5. The system according to claim 4, characterized in that, The correction module is used for: Obtain the box outline from the optical distribution box image; Extract the corner points of the box body outline; The corner points of the box are corrected using an affine transformation algorithm to obtain a first image, wherein the first image includes an image of a vertically placed rectangular box.

6. The system according to claim 4, characterized in that, The identification module is used for: Identify characters in sample images of the multiple hierarchical characters to be detected; The characters in the sample image are matched with the ports of different optical distribution boxes in the first image to obtain the recognized optical distribution box layered characters.

7. A terminal device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method as described in any one of claims 1-3.

8. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the steps of the method as described in any one of claims 1-3.

Images