CN113920308A - Identification method and identification system for steel coil number - Google Patents
Identification method and identification system for steel coil number Download PDFInfo
- Publication number
- CN113920308A CN113920308A CN202111514135.6A CN202111514135A CN113920308A CN 113920308 A CN113920308 A CN 113920308A CN 202111514135 A CN202111514135 A CN 202111514135A CN 113920308 A CN113920308 A CN 113920308A
- Authority
- CN
- China
- Prior art keywords
- steel coil
- information area
- information
- coil number
- sample
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/018—Certifying business or products
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
- General Factory Administration (AREA)
Abstract
The application provides a method and a system for identifying a steel coil number, which relate to the field of information identification, and the method comprises the following steps: acquiring video data corresponding to the steel coil; analyzing the video data to obtain a video frame of the steel coil; determining an information area containing a coil number from the video frame; inputting the information area into a preset identification model, and fitting the information area in the preset identification model by utilizing a Bezier curve to obtain a bent sample; and utilizing an alignment layer to straighten the bent sample, and extracting information of the small target sample in the information area to obtain the coil number. This application is through carrying out analysis and information extraction to the video data of coil of strip number, obtains the information area who contains the coil of strip number to utilize and predetermine the identification model, combine the Bessel curve to carry out you and discernment to the information area, thereby obtain the coil of strip number, can automatic real-time identification coil of strip number information, provide the information support for the logistics management of coil of strip.
Description
Technical Field
The application relates to the field of information identification, in particular to a method and a system for identifying a steel coil number.
Background
With the development of economy, the transfer amount of the steel coil in a railway freight yard is increased sharply, so that the management, monitoring, statistics and tracking of the circulation condition of the steel coil are particularly important, and the identification of the steel coil number as the unique identity of the steel coil is an important step. The traditional operation mode in a railway freight yard is to manually carry out on-site transcription before loading/unloading, so that the recognition rate cannot be guaranteed while certain potential safety hazards exist. Therefore, the automatic identification of the steel coil number in the railway freight yard becomes a problem which needs to be solved urgently at present.
Disclosure of Invention
The application aims to provide a steel coil number identification method and a steel coil number identification system, which can automatically identify steel coil number information and improve the identification efficiency of the steel coil number.
In order to solve the technical problem, the application provides a method for identifying a steel coil number, which has the following specific technical scheme:
acquiring video data corresponding to the steel coil;
analyzing the video data to obtain a video frame of the steel coil;
determining an information area containing a coil number from the video frame;
inputting the information area into a preset identification model, and fitting the information area in the preset identification model by utilizing a Bezier curve to obtain a bent sample;
and utilizing an alignment layer to straighten the bent sample, and extracting information of the small target sample in the information area to obtain the coil number.
Optionally, fitting the information region by using a bezier curve in the preset identification model to obtain a curved sample includes:
in the preset identification model, the steel coil image is used as a learning sample;
marking the learning samples one by one, utilizing a Bezier curve to define the number of the steel coil, and taking the area containing the number of the steel coil as an information area;
and identifying the information area to obtain a bending sample.
Optionally, fitting the information region by using a bezier curve in the preset identification model to obtain a curved sample, further comprising:
straightening the bent text of the text box corresponding to the information area, and marking the straightened text one by one;
and discarding the video frames with the definition lower than the preset value after the characters are marked.
Optionally, the step of determining the number of the steel coil by using the bezier curve, and the step of using the area containing the number of the steel coil as the information area includes:
the method comprises the following steps of determining the number of a steel coil by utilizing a cubic Bezier curve, and simplifying a region containing the number of the steel coil into a bounding box regression of eight control points;
and taking the region corresponding to the regression of the bounding box as an information region according to the coordinates of the control points.
Optionally, before taking a region corresponding to the regression of the bounding box as an information region according to the coordinates of the control point, the method further includes:
optionally, the information extraction of the small target sample in the information area to obtain the steel coil number includes:
extracting information of the small target sample in the information area by using a preset identification model to obtain the steel coil number; the preset recognition model is a model which comprises a plurality of convolution layers, 1 bidirectional LSTM layer and 1 fully-connected layer and is used for aligning text strings by using CTC loss based on output classification scores.
Optionally, after obtaining the steel coil number, the method further includes:
and uploading the steel coil number and the corresponding video data to a corresponding upper computer so as to be convenient for the upper computer to check the historical database data and output a steel coil number information table.
The application also provides a steel coil number identification system, including:
the video acquisition module is used for acquiring video data of the steel coil by using a high-definition camera when the steel coil is detected;
the video analysis module is used for analyzing the video data to obtain a video frame of the steel coil;
the information area identification module is used for determining an information area containing a coil number from the video frame;
the model processing module is used for inputting the information area into a preset identification model, and fitting the information area in the preset identification model by using a Bezier curve to obtain a bent sample;
and the steel coil number acquisition module is used for straightening the bent sample by utilizing the alignment layer and extracting information of the small target sample in the information area to obtain the steel coil number.
The application provides a method for identifying a steel coil number, which comprises the following steps: acquiring video data corresponding to the steel coil; analyzing the video data to obtain a video frame of the steel coil; determining an information area containing a coil number from the video frame; inputting the information area into a preset identification model, and fitting the information area in the preset identification model by utilizing a Bezier curve to obtain a bent sample; and utilizing an alignment layer to straighten the bent sample, and extracting information of the small target sample in the information area to obtain the coil number.
This application is through carrying out analysis and information extraction to the video data of coil of strip number, obtains the information area who contains the coil of strip number to utilize and predetermine the identification model, combine the Bessel curve to carry out you and discernment to the information area, thereby obtain the coil of strip number, can automatic real-time identification coil of strip number information, provide the information support for the logistics management of coil of strip.
This application still provides a steel coil number's identification system, has above-mentioned beneficial effect, and here is no longer repeated.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart of a method for identifying a steel coil number according to an embodiment of the present disclosure;
fig. 2 is a schematic diagram illustrating a process of identifying a steel coil number according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a system for identifying a steel coil number according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. 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.
Referring to fig. 1, fig. 1 is a flowchart of a method for identifying a steel coil number provided in an embodiment of the present application, and a specific scheme is as follows:
s101: acquiring video data corresponding to the steel coil;
s102: analyzing the video data to obtain a video frame of the steel coil;
s103: determining an information area containing a coil number from the video frame;
s104: inputting the information area into a preset identification model, and fitting the information area in the preset identification model by utilizing a Bezier curve to obtain a bent sample;
s105: and utilizing an alignment layer to straighten the bent sample, and extracting information of the small target sample in the information area to obtain the coil number.
The embodiment can be applied to any scene needing steel coil number identification. Specifically, in the identification process, video data of the steel coil number is shot by a high-definition camera. It should be noted that the video data may be a video shot by a high-definition camera, or may also be a continuous image shot by the high-definition camera, and both may be applied to the steel coil number identification process in this embodiment. In other words, the image or video including the content of the steel coil number captured by the high-definition camera can be processed and identified by the embodiment. It is easy to understand that, when the present embodiment is implemented, the real-time video data corresponding to the steel coil may be acquired in S101. In other embodiments of the present application, the video data obtained in other manners may be directly processed, and is not limited to the source and the shooting manner of the video data.
In step S102, the video data is analyzed, that is, the video data is analyzed, so as to obtain a video frame of the steel coil. It should be noted that, if the video data is to be separated frame by frame, the video frames containing the steel coils are identified, and the video frames not containing the steel coils are removed.
After that, an information region containing the coil number needs to be determined from the video frame, and the process may use a deep learning model for feature recognition to determine the information region containing the coil number, or may directly use a later preset recognition model for recognition, which is not limited specifically herein. Of course, the size of the information area should be set in association with the setting of the model to be used. It is easily understood that, in general, the smaller the information area determined in this step is, the more efficient the identification of the coil number in the subsequent execution information area is.
After that, the identification process of the preset identification model to the coil number in the information area is executed, and the specific process can be divided into two stages, namely obtaining the bent sample firstly and identifying the bent sample. Because the steel coil is usually in a curled shape, the steel coil number on the surface of the steel coil also presents certain surface radian along with the curling of the steel coil, and certain difficulty is caused for information identification. When the information area is specifically identified, in the first stage, the information area is firstly input into a preset identification model, and a Bezier curve is used for fitting the information area in the preset identification model to obtain a bent sample.
The process can comprise the following steps:
firstly, in a preset identification model, taking a steel coil image as a learning sample;
secondly, labeling the learning samples one by one, utilizing a Bezier curve to define the number of the steel coil, and taking the area containing the number of the steel coil as an information area;
and thirdly, identifying the information area to obtain a bending sample.
It is easily understood that the present embodiment defaults to acquiring the preset recognition model before executing S104, and no specific limitation is made on how to obtain the preset recognition model. The present embodiment provides a preferred generation process of the preset recognition model:
(1) acquiring an end face image of a steel coil, and taking the steel coil image as a learning sample;
the number of samples is not particularly limited, and for example, the number of samples may be not less than 200000;
(2) further, labeling the learning samples one by one, using a Bezier curve to define a steel coil number, straightening the bent text of the text box, and then labeling the straightened text content character by character;
(3) and further, inputting the marked sample into a deep learning model for learning training to obtain a preset identification model.
Similarly, when step S104 is executed, the bent text of the text box corresponding to the information area may be straightened, the straightened text is labeled one by one, and the video frame with the definition lower than the preset value after the character labeling is discarded. The definition criterion may refer to the above image quality scoring rule, or may adopt other definition determining manners, which are not specifically described herein. It can be understood that, in the process of identifying the steel coil number, the identification efficiency of the steel coil number can be improved by discarding the image frame with lower definition.
Specifically, in order to obtain the information area, the steel coil number may be defined by using a cubic bezier curve, the area including the steel coil number is simplified into a bounding box regression of eight control points, and then the area corresponding to the bounding box regression is used as the information area according to the coordinates of the control points. To determine the coordinates of the control points, Bezier curve true labels may be generated and the coordinates of the control points may be learned using a regression formula.
The FPN is a characteristic pyramid network structure formed by different convolutional neural networks in deep learning, parameters in the network structure can be trained, and parameters of the Bezier curve can be trained by means of the network structure of the FPN. Then based on the FPN, the Bezier curve is a parameterized curve that can be used
It is shown that the curve uses a bernstein polynomial as its bias. The definition is as follows:
wherein the content of the first and second substances,
indicating the degree;
is shown as
A control point;
representing a Bernstein polynomial, t is a sub-parameter, is a variable in a Bernstein polynomial function, and 0<t<1 is the normalized interval range which is defined as follows:
wherein
Are binomial coefficients. Using cubic bezier curves (i.e.
3) to mark the curved text. The plurality of control points form a control polygon, and the curve is a cubic bezier curve.
Based on cubic bezier curves, arbitrarily curved text detection is simplified to a bounding box regression with a total of eight control points. In order to learn the coordinates of the control points, Bezier curve real labels are generated first, and then a regression method is used to learn the target. For each text sample:
in the formula
And
the minimum values of x and y in the four vertexes are respectively, and x and y are coordinate values in the two-dimensional image. The advantage of predicting the relative distance is independent of whether the bezier curve control point exceeds the image boundary. Under inspectionInside the measuring head, because Bezier is eight control points, each point has two coordinate values, namely 16 coordinate values, and each channel corresponds to one coordinate value, only a convolution layer with 16 output channels is needed to learn
And
and (4) finishing.
And (3) providing a Bezier alignment method for feature sampling by using the parameterization characteristic of the compact Bezier curve bounding box. Each column of the bezier-aligned sampling grid is orthogonal to the bezier curve boundaries of the text. The width and height of the sampling points are respectively at equal intervals, and bilinear interpolation is carried out on the coordinates. The sampling point refers to sampling of a block region on an image, and is sampling of a plurality of pixel points, so that the width is actually the width of the region.
Given the input feature map and the Bezier curve control points, the output pixel size is processed simultaneously to be
The rectangular feature map of (1). Output the position in the graph by the feature
Is formed by a plurality of pixels
For example, it is calculated from the following formula:
in the above equation, T is a corresponding coefficient for calculating a position on the feature map to a coordinate position on the original image.
And then calculating the points of the upper Bezier curve boundary tp and the lower Bezier curve boundary bp by using the definition formula of t and the Bezier curve. Using tp and bp, the sampling points op can be linearly indexed by:
and calculating the corresponding coordinate value in the image by using the position of the sampling point op and applying bilinear interpolation.
And then, identifying the information area, straightening the bent sample by using the alignment layer, and extracting information of the small target sample in the information area to obtain the coil number.
The preset recognition model is a model which comprises a plurality of convolution layers, 1 bidirectional LSTM layer and 1 fully connected layer and is used for aligning text strings by utilizing CTC loss based on output classification scores. Based on the output classification score, text string (GT) alignment is performed using classical CTC loss. Since the generated Bezier curve GT is directly used to extract the RoI features during the training process. The detection branch does not affect the identification branch. In the inference phase, the RoI region is replaced by a detected bezier curve.
The part describes a function module related to a deep learning model identification principle, and the building of a deep learning network model requires the combination of different convolutional neural networks, and the identification module of LSTM and ROI, which is formed by the combination of different convolutional neural networks, is as follows: a convolutional layer, a pooling layer, an activation layer and other basic neurons. In the deep model training stage, a loss function is needed to calculate the loss value, so that the model weight parameters (the loss value: the difference value between the predicted value and the real value) are updated through back propagation. CTC loss requires the output of every possible output and its conditional probability, a loss function for deep learning model training. ROI (region of interest): in the region of interest to be identified in the image (i.e. the region of the image to be detected), the model needs to find the ROI in the image to identify the specific content first.
This application embodiment is through carrying out analysis and information extraction to the video data of coil of strip number, obtains the information area who contains the coil of strip number to utilize and predetermine the identification model, combine the Bessel curve to carry out you and discernment to the information area, thereby obtain the coil of strip number, can automatic real-time identification coil of strip number information, provide the information support for the logistics management of coil of strip.
In addition, after the steel coil number is obtained, the steel coil number and corresponding video data can be uploaded to a corresponding upper computer, so that the upper computer can conveniently correct the data of the historical library and output a steel coil number information table. Through outputting the steel coil number information table, the information of the steel coil number required to be identified can be quickly acquired, so that a user can timely count and store the information in a warehouse and the like.
A specific application process of the present application is described below, referring to fig. 2, fig. 2 is a schematic diagram of a specific identification process of a steel coil number provided in an embodiment of the present application, and fig. 2 includes a gantry crane weight sensor 1, a high definition camera 2, a switch 3, a steel coil number identification device 4, an upper computer 5 and an output unit 6, specifically:
the gantry crane weight sensor 1 is used for triggering the camera to shoot, and as shown in fig. 2, when the gantry crane lifts a steel coil, and the data of the weight sensor exceeds a set threshold, a trigger signal is sent to the camera.
The coil number recognition device 4 is used for analyzing and recognizing the received video signals to obtain the serial number information of the coil. The method comprises the steps of identifying the characteristics of a steel coil by using a deep learning model calculation engine, then carrying out large-target labeling on a steel coil number information area, filtering out video frames without information marks, scoring the information frames, identifying key frames with the highest discrimination scores as information identification frames, reconstructing a curved scene text in parameterized Bezier curve information identification frames, marking a regular Bezier curve bounding box, accurately calculating and extracting curve sequence characteristics of a curved shape text example by using a cubic Bezier characteristic alignment layer, straightening the curved text of a text box, finally identifying information of each character by using an identification layer, sequencing all identified characters by using a circulating network, finally obtaining a complete identification result, and sending the complete identification result to an upper computer 5.
The upper computer 5 is used for integrating the serial number information of the steel coil, summarizing the serial number information into a steel coil serial number summary table, and providing the identification result for a service application system by the output unit 6.
In the following, the identification system of the steel coil number provided by the embodiment of the present application is introduced, and the identification system of the steel coil number described below and the identification method of the steel coil number described above may be referred to correspondingly.
Referring to fig. 3, fig. 3 is a schematic structural diagram of a system for identifying a steel coil number provided in an embodiment of the present application, and the present application further provides a system for identifying a steel coil number, including:
the video acquisition module is used for acquiring video data of the steel coil by using a high-definition camera when the steel coil is detected;
the video analysis module is used for analyzing the video data to obtain a video frame of the steel coil;
the information area identification module is used for determining an information area containing a coil number from the video frame;
the model processing module is used for inputting the information area into a preset identification model, and fitting the information area in the preset identification model by using a Bezier curve to obtain a bent sample;
and the steel coil number acquisition module is used for straightening the bent sample by utilizing the alignment layer and extracting information of the small target sample in the information area to obtain the steel coil number.
Based on the above embodiment, as a preferred embodiment, the method further includes:
the marking module is used for straightening the bent text of the text box corresponding to the information area and marking the straightened text one by one; and discarding the video frames with the definition lower than the preset value after the characters are marked.
Based on the above embodiment, as a preferred embodiment, the method further includes:
and the information table output module is used for uploading the steel coil number and the corresponding video data to a corresponding upper computer so as to be convenient for the upper computer to correct the data of the historical library and output a steel coil number information table.
The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system provided by the embodiment, the description is relatively simple because the system corresponds to the method provided by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present application are explained herein using specific examples, which are provided only to help understand the method and the core idea of the present application. It should be noted that, for those skilled in the art, it is possible to make several improvements and modifications to the present application without departing from the principle of the present application, and such improvements and modifications also fall within the scope of the claims of the present application.
It is further noted that, in the present specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (8)
1. A method for identifying a steel coil number is characterized by comprising the following steps:
acquiring video data corresponding to the steel coil;
analyzing the video data to obtain a video frame of the steel coil;
determining an information area containing a coil number from the video frame;
inputting the information area into a preset identification model, and fitting the information area in the preset identification model by utilizing a Bezier curve to obtain a bent sample;
and utilizing an alignment layer to straighten the bent sample, and extracting information of the small target sample in the information area to obtain the coil number.
2. The method for identifying the steel coil number according to claim 1, wherein fitting the information area by using a Bezier curve in the preset identification model to obtain a bending sample comprises:
in the preset identification model, the steel coil image is used as a learning sample;
marking the learning samples one by one, utilizing a Bezier curve to define the number of the steel coil, and taking the area containing the number of the steel coil as an information area;
and identifying the information area to obtain a bending sample.
3. The method for identifying the steel coil number according to claim 1, wherein when the information area is fitted with a bezier curve in the preset identification model to obtain a bending sample, the method further comprises:
straightening the bent text of the text box corresponding to the information area, and marking the straightened text one by one;
and discarding the video frames with the definition lower than the preset value after the characters are marked.
4. The method for identifying a steel coil number according to claim 1, wherein the step of delineating the steel coil number using a Bezier curve and the step of defining a region including the steel coil number as an information region includes:
the method comprises the following steps of determining the number of a steel coil by utilizing a cubic Bezier curve, and simplifying a region containing the number of the steel coil into a bounding box regression of eight control points;
and taking the region corresponding to the regression of the bounding box as an information region according to the coordinates of the control points.
5. The method for identifying a steel coil number according to claim 4, wherein before the region corresponding to the bounding box regression is used as an information region according to the coordinates of the control point, the method further comprises:
and generating a Bezier curve real label, and learning the coordinates of the control points by using a regression formula.
6. The method for identifying the steel coil number according to claim 1, wherein extracting information of the small target sample in the information area to obtain the steel coil number comprises:
extracting information of the small target sample in the information area by using a preset identification model to obtain the steel coil number; the preset recognition model is a model which comprises a plurality of convolution layers, 1 bidirectional LSTM layer and 1 fully-connected layer and is used for aligning text strings by using CTC loss based on output classification scores.
7. The method for identifying the steel coil number according to claim 1, wherein after obtaining the steel coil number, the method further comprises:
and uploading the steel coil number and the corresponding video data to a corresponding upper computer so as to be convenient for the upper computer to check the historical database data and output a steel coil number information table.
8. The utility model provides a steel coil number's identification system which characterized in that includes:
the video acquisition module is used for acquiring video data of the steel coil by using a high-definition camera when the steel coil is detected;
the video analysis module is used for analyzing the video data to obtain a video frame of the steel coil;
the information area identification module is used for determining an information area containing a coil number from the video frame;
the model processing module is used for inputting the information area into a preset identification model, and fitting the information area in the preset identification model by using a Bezier curve to obtain a bent sample;
and the steel coil number acquisition module is used for straightening the bent sample by utilizing the alignment layer and extracting information of the small target sample in the information area to obtain the steel coil number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111514135.6A CN113920308A (en) | 2021-12-13 | 2021-12-13 | Identification method and identification system for steel coil number |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111514135.6A CN113920308A (en) | 2021-12-13 | 2021-12-13 | Identification method and identification system for steel coil number |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113920308A true CN113920308A (en) | 2022-01-11 |
Family
ID=79248604
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111514135.6A Pending CN113920308A (en) | 2021-12-13 | 2021-12-13 | Identification method and identification system for steel coil number |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113920308A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111259908A (en) * | 2020-03-24 | 2020-06-09 | 中冶赛迪重庆信息技术有限公司 | Machine vision-based steel coil number identification method, system, equipment and storage medium |
| CN112906801A (en) * | 2021-02-26 | 2021-06-04 | 北京房江湖科技有限公司 | Image processing method, device and storage medium |
| CN113516124A (en) * | 2021-05-29 | 2021-10-19 | 大连民族大学 | Electric energy meter electricity consumption information identification algorithm based on computer vision technology |
-
2021
- 2021-12-13 CN CN202111514135.6A patent/CN113920308A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111259908A (en) * | 2020-03-24 | 2020-06-09 | 中冶赛迪重庆信息技术有限公司 | Machine vision-based steel coil number identification method, system, equipment and storage medium |
| CN112906801A (en) * | 2021-02-26 | 2021-06-04 | 北京房江湖科技有限公司 | Image processing method, device and storage medium |
| CN113516124A (en) * | 2021-05-29 | 2021-10-19 | 大连民族大学 | Electric energy meter electricity consumption information identification algorithm based on computer vision technology |
Non-Patent Citations (2)
| Title |
|---|
| YULIANG LIU ET AL.: "ABCNet: Real-Time Scene Text Spotting With Adaptive Bezier-Curve Network", 《2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)》 * |
| 刘禹良: "基于深度学习的自然场景文本检测及端到端识别的研究", 《中国博士学位论文全文数据库 (信息科技辑)》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110738101B (en) | Behavior recognition method, behavior recognition device and computer-readable storage medium | |
| CN108229509B (en) | Method and device for identifying object class and electronic equipment | |
| CN113705478B (en) | Mangrove single wood target detection method based on improved YOLOv5 | |
| CN110807422A (en) | Natural scene text detection method based on deep learning | |
| CN110348439B (en) | Method, computer readable medium and system for automatically identifying price tags | |
| CN107992819B (en) | Method and device for determining vehicle attribute structural features | |
| CN111639629B (en) | Pig weight measurement method and device based on image processing and storage medium | |
| CN110287952B (en) | Method and system for recognizing characters of dimension picture | |
| CN111368766A (en) | Cattle face detection and identification method based on deep learning | |
| CN111178405A (en) | Similar object identification method fusing multiple neural networks | |
| CN116740758A (en) | Bird image recognition method and system for preventing misjudgment | |
| CN113378764B (en) | Video face acquisition method, device, equipment and medium based on clustering algorithm | |
| CN111241905A (en) | Power transmission line nest detection method based on improved SSD algorithm | |
| CN112364687A (en) | Improved Faster R-CNN gas station electrostatic sign identification method and system | |
| CN113920308A (en) | Identification method and identification system for steel coil number | |
| Calefati et al. | Reading meter numbers in the wild | |
| CN115116074A (en) | Handwritten character recognition and model training method and device | |
| CN115953744A (en) | Vehicle identification tracking method based on deep learning | |
| CN114758139A (en) | Foundation pit accumulated water detection method | |
| CN115049600A (en) | Intelligent identification system and method for small sample pipeline defects | |
| CN110472583B (en) | Human face micro-expression recognition system based on deep learning | |
| CN109583428B (en) | Court sign recognition method based on video | |
| CN114155475B (en) | Method, device and medium for identifying end-to-end personnel actions under view angle of unmanned aerial vehicle | |
| CN113361503B (en) | Intelligent garden arbor quantity detection method and system based on unmanned aerial vehicle aerial photography | |
| CN115810020B (en) | Semantic guidance-based coarse-to-fine remote sensing image segmentation 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220111 |