CN114882204A - Automatic ship name recognition method - Google Patents
Automatic ship name recognition method Download PDFInfo
- Publication number
- CN114882204A CN114882204A CN202210215789.7A CN202210215789A CN114882204A CN 114882204 A CN114882204 A CN 114882204A CN 202210215789 A CN202210215789 A CN 202210215789A CN 114882204 A CN114882204 A CN 114882204A
- Authority
- CN
- China
- Prior art keywords
- ship
- image
- name
- training
- model
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
Abstract
The invention discloses a method and a medium for automatically identifying ship names, wherein the method comprises the following steps: acquiring ship video data, and preprocessing the ship video data to extract ship image data; inputting the ship image into a character region detection model to extract an interested region where a ship name corresponding to the ship image is located, marking the interested region where the ship name is located, and performing data expansion according to the ship image marked with the interested region where the ship name is located to generate a training sample set; training the model according to the training sample set to obtain a final ship name automatic identification model; acquiring a ship plate image to be recognized, inputting the ship plate image to be recognized into a final ship name automatic recognition model, and automatically recognizing the ship name of the ship plate image to be recognized through the final ship name automatic recognition model; the collected data can be automatically labeled, the training efficiency of the recognition model is improved, and the labor cost required to be consumed in the training process of the recognition model is reduced.
Description
Technical Field
The invention relates to the technical field of ship management, in particular to an automatic ship name identification method and a computer readable storage medium.
Background
In the related art, in the process of extracting the name of the ship, a training recognition model is mostly adopted, so that the image is recognized through the recognition model. However, in the process of training the recognition model, the collected data needs to be manually screened and labeled, so that the training efficiency of the recognition model is low, and a great labor cost needs to be consumed.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the art described above. Therefore, an object of the present invention is to provide an automatic ship name recognition method, which can automatically label acquired data, improve training efficiency of a recognition model, and reduce labor cost required in a training process of the recognition model.
A second object of the invention is to propose a computer-readable storage medium.
In order to achieve the above object, an embodiment of a first aspect of the present invention provides an automatic ship name identification method, including the following steps: acquiring ship video data, and preprocessing the ship video data to extract ship image data; inputting a ship image in the ship image data into a character region detection model to extract an interested region where a ship name corresponding to the ship image is located, labeling the interested region where the ship name is located, and performing data expansion according to the ship image labeled with the interested region where the ship name is located to generate a training sample set; training a model according to the training sample set to obtain a final ship name automatic identification model; and acquiring a ship plate image to be recognized, inputting the ship plate image to be recognized into the final ship name automatic recognition model, and automatically recognizing the ship name of the ship plate image to be recognized through the final ship name automatic recognition model.
According to the automatic ship name identification method provided by the embodiment of the invention, firstly, ship video data are obtained and preprocessed to extract ship image data; then, inputting a ship image in the ship image data into the character region detection model to extract an interested region where a ship name corresponding to the ship image is located, labeling the interested region where the ship name is located, and performing data expansion according to the ship image labeled with the interested region where the ship name is located to generate a training sample set; then, training the model according to the training sample set to obtain a final ship name automatic identification model; then, acquiring a ship plate image to be recognized, inputting the ship plate image to be recognized into a final ship name automatic recognition model, and automatically recognizing the ship name of the ship plate image to be recognized through the final ship name automatic recognition model; therefore, the collected data are automatically labeled, the training efficiency of the recognition model is improved, and the labor cost required to be consumed in the training process of the recognition model is reduced.
In addition, the automatic ship name identification method provided by the above embodiment of the present invention may further have the following additional technical features:
optionally, the preprocessing the ship video data includes: extracting the ship video data at preset frame intervals to obtain initial ship video frames; inputting an initial ship video frame into a target classification pre-training model, and screening the initial ship video frame through the target classification pre-training model to obtain ship image data containing ship information.
Optionally, the data expansion is performed according to the ship image labeled with the region of interest where the ship name is located, and the data expansion includes: and performing data enhancement on the ship image marked with the interesting region where the ship name is located, performing background replacement on the ship image marked with the interesting region where the ship name is located according to an edge detection algorithm, and automatically generating a ship plate image and a corresponding label according to a ship plate automatic generation algorithm.
Optionally, the data enhancement mode includes zooming and/or translating the ship image marked with the region of interest where the ship name is located.
Optionally, training a model according to the training sample set includes: adjusting the loss function to improve the loss weight ratio of Chinese character learning, wherein in the calculation process of the loss function, the type of the prediction result is judged, and the corresponding weight value of the prediction result is determined according to the type of the prediction result; randomly extracting a preset amount of training data in a training sample set from an iterator in each iteration step for training; and performing a comparison experiment to determine a loss weight value corresponding to the type of each prediction result.
Optionally, the type of the prediction result includes chinese characters, letters, and arabic numbers.
Optionally, the loss function is expressed by the following formula:
wherein, loss (X) represents a loss function, X represents an input sequence, pi represents a path of an output sequence, M represents a mapping function, T' represents a characteristic diagram length extracted by a ship name recognition network, P (pi | X) represents the probability that the path of the output sequence is pi, and alpha represents the weight of each type of character.
Optionally, acquiring an image of the ship board to be identified includes: acquiring existing ship data, and training a ship tracking model according to the existing ship data; the method comprises the steps of obtaining images of the same ship at different spatial positions through the ship tracking model, detecting the images at different spatial positions through the character region detection model to obtain ship plate images of multiple angles corresponding to the ship, calculating the ship plate image of each angle according to a preset activation function to obtain confidence degrees corresponding to the ship plate image of each angle, and taking the ship plate image of the confidence degrees as a ship plate image to be recognized.
Optionally, the preset activation function is expressed by the following formula:
I＝Max(P 0 ,P 1 ,P 2 ,P 3 ,...P k )
Score＝Max(I 0 ,I 1 ,I 2 ,...I n ,)
wherein k represents the number of all character classes in the training sample set, v i Indicating the output value of class i, P i One confidence belonging to class I is represented, I represents the boat image with the highest confidence, and Score represents the confidence output Score of the boat image with the highest confidence.
In order to achieve the above object, a second embodiment of the present invention provides a computer-readable storage medium, on which a ship name automatic identification program is stored, which, when executed by a processor, implements the ship name automatic identification method as described above.
According to the computer-readable storage medium of the embodiment of the invention, the automatic ship name recognition program is stored, so that the processor can realize the automatic ship name recognition method when executing the automatic ship name recognition program, thereby automatically marking the acquired data, improving the training efficiency of the recognition model, and reducing the labor cost required to be consumed in the training process of the recognition model
Drawings
Fig. 1 is a schematic flow chart of a method for automatically identifying a ship name according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of ship board image data expansion according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
In the related technology, in the process of training the recognition model, the collected data needs to be screened and labeled manually, so that the training efficiency of the recognition model is low, and great labor cost needs to be consumed; according to the automatic ship name identification method provided by the embodiment of the invention, firstly, ship video data are obtained and preprocessed to extract ship image data; then, inputting a ship image in the ship image data into the character region detection model to extract an interested region where a ship name corresponding to the ship image is located, labeling the interested region where the ship name is located, and performing data expansion according to the ship image labeled with the interested region where the ship name is located to generate a training sample set; then, training the model according to the training sample set to obtain a final ship name automatic identification model; then, acquiring a ship plate image to be recognized, inputting the ship plate image to be recognized into a final ship name automatic recognition model, and automatically recognizing the ship name of the ship plate image to be recognized through the final ship name automatic recognition model; therefore, the collected data are automatically labeled, the training efficiency of the recognition model is improved, and the labor cost required to be consumed in the training process of the recognition model is reduced.
In order to better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
In order to better understand the technical scheme, the technical scheme is described in detail in the following with reference to the attached drawings of the specification and specific embodiments.
Fig. 1 is a schematic flow chart of a ship name automatic identification method according to an embodiment of the present invention, and as shown in fig. 1, the ship name automatic identification method includes the following steps:
s101, acquiring ship video data, and preprocessing the ship video data to extract ship image data.
That is, the camera is used for shooting the ship to obtain the ship video data corresponding to the ship, and it can be understood that all video frames in the ship video data do not contain effective ship information; therefore, the ship video data needs to be preprocessed to extract valid video frames therein, so as to obtain ship image data.
In some embodiments, pre-processing the vessel video data comprises: extracting ship video data at preset frame number intervals to obtain initial ship video frames; and inputting the initial ship video frame into a target classification pre-training model, and screening the initial ship video frame through the target classification pre-training model to obtain ship image data containing ship information.
As an example, firstly, ship video data is collected to obtain ship video images of a ship entering and exiting a port; then, reading the ship video data by utilizing OpenCV; then, extracting video frames in the ship video data at preset frame number intervals to obtain initial ship video frames; then, inputting the initial ship video frame into a target classification pre-training model, screening the initial ship video frame through the target classification pre-training model, and eliminating images which do not contain effective ship information; thereby obtaining ship image data containing ship information; it can be understood that the ship video data is preprocessed in such a way, manual participation is not needed, the data processing efficiency can be effectively improved, and the labor cost is reduced.
S102, inputting the ship image in the ship image data into the character region detection model to extract the interesting region of the ship name corresponding to the ship image, labeling the interesting region of the ship name, and performing data expansion according to the ship image labeled with the interesting region of the ship name to generate a training sample set.
That is to say, the ship image in the ship image data is input into the character region detection model, so as to detect the character region in the ship image through the character region detection model, and further, the interested region where the ship name corresponding to the ship image is located is extracted; and then, the region of interest where the ship name is located is automatically marked, so that the image is not required to be marked manually, the labor cost is greatly saved, and the forming efficiency of the training sample set is improved. Then, performing data expansion according to the ship image marked with the interesting region of the ship name to generate a training sample set; therefore, by performing data expansion on the original data, a large number of training samples can be generated under the condition that the original samples are limited, and the generation difficulty of the training sample set is reduced.
In some embodiments, the data expansion is performed according to the ship image marked with the interested area of the ship name, and the data expansion comprises the following steps: and performing data enhancement on the ship image marked with the interesting region where the ship name is located, performing background replacement on the ship image marked with the interesting region where the ship name is located according to an edge detection algorithm, and automatically generating a ship plate image and a corresponding label according to a ship plate automatic generation algorithm.
As an example, the data enhancement includes zooming and/or panning the ship image that marks the region of interest where the ship name is located.
As shown in fig. 2, fig. 2a is an initial ship name area image; fig. 2b is the image after data enhancement, fig. 2c is the image after background replacement, and fig. 2d is the image of the final generated ship plate.
And S103, training the model according to the training sample set to obtain a final ship name automatic identification model.
In some embodiments, the training of the model is performed according to a training sample set, comprising: adjusting the loss function to improve the loss weight ratio of Chinese character learning, wherein in the calculation process of the loss function, the type of the prediction result is judged, and the corresponding weight value of the prediction result is determined according to the type of the prediction result; randomly extracting a preset amount of training data in a training sample set from an iterator in each iteration step for training; and performing a comparison experiment to determine a loss weight value corresponding to the type of each prediction result.
It can be understood that, in the model training stage, the recognition of chinese (simplified and traditional) characters is a big difficulty, so that the loss weight ratio of the learning of the chinese characters is improved by adjusting the loss function, and the model can give more attention to the recognition of chinese characters in the learning process, so as to improve the recognition efficiency and accuracy of the final model for the chinese characters. In addition, by using the iterator, the training data is continuously updated in the iteration process, so that overfitting of the model can be effectively prevented, and the identification precision is further improved.
The type of the prediction result may be set in various ways.
As an example, the types of prediction results include Chinese characters, letters, and Arabic numerals.
As an example, the loss function is expressed by the following formula:
wherein loss (X) represents a loss function, X represents an input sequence, pi represents a path of an output sequence, M represents a mapping function, T' represents a characteristic diagram length extracted by a ship name recognition network, P (pi | X) represents the probability that the path of the output sequence is pi, and alpha represents the weight of each type of character.
And S104, acquiring the ship plate image to be recognized, inputting the ship plate image to be recognized into the final ship name automatic recognition model, and automatically recognizing the ship name of the ship plate image to be recognized through the final ship name automatic recognition model.
That is to say, after the training of the final ship name automatic identification model is completed, the ship image of the ship name to be identified is acquired, the ship plate image to be identified is input into the final ship name automatic identification model, the ship plate image to be identified is identified through the final ship name automatic identification model, the ship name corresponding to the image is obtained, and the automatic identification process is completed.
In some embodiments, obtaining the ship board image to be identified comprises: acquiring the existing ship data, and training a ship tracking model according to the existing ship data; the method comprises the steps of obtaining images of the same ship at different spatial positions through a ship tracking model, detecting the images at different spatial positions through a character region detection model to obtain ship plate images of multiple angles corresponding to the ship, calculating the ship plate image of each angle according to a preset activation function to obtain confidence corresponding to the ship plate image of each angle, and taking the ship plate image of the confidence as a ship plate image to be identified.
That is, firstly, based on the existing ship data, a ship tracking model is trained, so as to obtain ship images through the ship tracking model, and obtain images of the same ship at different spatial positions; then, detecting the images at different spatial positions by using a character region detection model to obtain ship plate images at a plurality of angles corresponding to the ship; then, inputting the ship plate images of a plurality of angles into a final ship name automatic identification model so as to identify the ship name through the final ship name automatic identification model; and in the identification process, the confidence corresponding to the ship plate image of each angle is output according to the activation function, and the identification result with the highest confidence is finally selected.
The setting mode of the activation function can be various.
As an example, the preset activation function is expressed by the following formula:
I＝Max(P 0 ,P 1 ,P 2 ,P 3 ,...P k )
Score＝Max(I 0 ,I 1 ,I 2 ,...I n ,)
wherein k represents the number of all character classes in the training sample set, v i Indicating the output value of class i, P i One confidence belonging to class I is represented, I represents the boat image with the highest confidence, and Score represents the confidence output Score of the boat image with the highest confidence.
In summary, according to the automatic ship name identification method provided by the embodiment of the invention, firstly, the ship video data is obtained, and the ship video data is preprocessed to extract the ship image data; then, inputting a ship image in the ship image data into the character region detection model to extract an interested region where a ship name corresponding to the ship image is located, labeling the interested region where the ship name is located, and performing data expansion according to the ship image labeled with the interested region where the ship name is located to generate a training sample set; then, training the model according to the training sample set to obtain a final ship name automatic identification model; then, acquiring a ship plate image to be recognized, inputting the ship plate image to be recognized into a final ship name automatic recognition model, and automatically recognizing the ship name of the ship plate image to be recognized through the final ship name automatic recognition model; therefore, the collected data are automatically labeled, the training efficiency of the recognition model is improved, and the labor cost required to be consumed in the training process of the recognition model is reduced.
In order to achieve the above embodiments, an embodiment of the present invention proposes a computer-readable storage medium having stored thereon a ship name automatic identification program that, when executed by a processor, implements a ship name automatic identification method as described above.
According to the computer-readable storage medium of the embodiment of the invention, the automatic ship name recognition program is stored, so that the processor can realize the automatic ship name recognition method when executing the automatic ship name recognition program, thereby automatically marking the acquired data, improving the training efficiency of the recognition model, and reducing the labor cost required to be consumed in the training process of the recognition model
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It should be noted that in the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
In the description of the present invention, it is to be understood that the terms "first", "second" and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implying any number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
In the present invention, unless otherwise expressly stated or limited, the terms "mounted," "connected," "secured," and the like are to be construed broadly and can, for example, be fixedly connected, detachably connected, or integrally formed; can be mechanically or electrically connected; either directly or indirectly through intervening media, either internally or in any other relationship. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
In the present invention, unless otherwise expressly stated or limited, the first feature "on" or "under" the second feature may be directly contacting the first and second features or indirectly contacting the first and second features through an intermediate. Also, a first feature "on," "over," and "above" a second feature may be directly or diagonally above the second feature, or may simply indicate that the first feature is at a higher level than the second feature. A first feature being "under," "below," and "beneath" a second feature may be directly under or obliquely under the first feature, or may simply mean that the first feature is at a lesser elevation than the second feature.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above should not be understood to necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (10)
1. A ship name automatic identification method is characterized by comprising the following steps:
acquiring ship video data, and preprocessing the ship video data to extract ship image data;
inputting a ship image in the ship image data into a character region detection model to extract an interested region where a ship name corresponding to the ship image is located, labeling the interested region where the ship name is located, and performing data expansion according to the ship image labeled with the interested region where the ship name is located to generate a training sample set;
training a model according to the training sample set to obtain a final ship name automatic identification model;
and acquiring a ship plate image to be recognized, inputting the ship plate image to be recognized into the final ship name automatic recognition model, and automatically recognizing the ship name of the ship plate image to be recognized through the final ship name automatic recognition model.
2. The method of claim 1, wherein preprocessing the vessel video data comprises:
extracting the ship video data at preset frame number intervals to obtain initial ship video frames;
inputting an initial ship video frame into a target classification pre-training model, and screening the initial ship video frame through the target classification pre-training model to obtain ship image data containing ship information.
3. The method for automatically identifying the ship name according to claim 1, wherein the data expansion is performed according to the ship image marked with the interested area of the ship name, and the method comprises the following steps:
and performing data enhancement on the ship image marked with the interesting region where the ship name is located, performing background replacement on the ship image marked with the interesting region where the ship name is located according to an edge detection algorithm, and automatically generating a ship plate image and a corresponding label according to a ship plate automatic generation algorithm.
4. The automatic ship name recognition method according to claim 3, characterized in that the data enhancement mode comprises zooming and/or translating the ship image marked with the area of interest of the ship name; the background replacement mode comprises the steps of carrying out pixel segmentation on a character area, and extracting foreground and background images, so that ship name characters are written in different backgrounds; the automatic generation algorithm of the ship plate comprises the steps of intercepting contour backgrounds of a plurality of ships and taking the contour backgrounds as ship character base images, designing the ship plate images by combining an image preprocessing technology, and automatically generating the ship plate images and labels by introducing environmental noise, Gaussian blur, saturation adjustment, random conversion of character sizes in a set range and the like.
5. The method for automatically identifying the ship name according to claim 1, wherein the training of the model according to the training sample set comprises:
adjusting the loss function to improve the loss weight ratio of Chinese character learning, wherein in the calculation process of the loss function, the type of the prediction result is judged, and the corresponding weight value of the prediction result is determined according to the type of the prediction result;
randomly extracting a preset amount of training data in a training sample set from an iterator in each iteration step for training;
and performing a comparison experiment to determine a loss weight value corresponding to the type of each prediction result.
6. The method of claim 5, wherein the type of the prediction result comprises Chinese characters, letters and Arabic numerals.
7. The automatic ship name recognition method according to claim 5, wherein the loss function is expressed by the following formula:
wherein loss (X) represents a loss function, X represents an input sequence, pi represents a path of an output sequence, M represents a mapping function, T' represents a characteristic diagram length extracted by a ship name recognition network, P (pi | X) represents the probability that the path of the output sequence is pi, and alpha represents the weight of each type of character.
8. The automatic ship name recognition method of claim 1, wherein obtaining the image of the ship plate to be recognized comprises:
acquiring existing ship data, and training a ship tracking model according to the existing ship data;
the images of the same ship at different spatial positions are obtained through the ship tracking model, the images at different spatial positions are detected through the character region detection model to obtain ship plate images at a plurality of angles corresponding to the ship,
and calculating the ship plate image of each angle according to a preset activation function to obtain the corresponding confidence coefficient of the ship plate image of each angle, and taking the ship plate image of the confidence coefficient as the ship plate image to be recognized.
9. The automatic ship name recognition method according to claim 8, wherein the preset activation function is expressed by the following formula:
I＝Max(P 0 ,P 1 ,P 2 ,P 3 ,...P k )
Score＝Max(I 0 ,I 1 ,I 2 ,...I n ,)
wherein k represents the number of all character classes in the training sample set, v i Indicating the output value of class i, P i One confidence belonging to class I is represented, I represents the boat image with the highest confidence, and Score represents the confidence output Score of the boat image with the highest confidence.
10. A computer-readable storage medium, characterized in that a ship name automatic recognition program is stored thereon, which when executed by a processor, implements the ship name automatic recognition method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210215789.7A CN114882204A (en) | 2022-03-07 | 2022-03-07 | Automatic ship name recognition method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210215789.7A CN114882204A (en) | 2022-03-07 | 2022-03-07 | Automatic ship name recognition method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114882204A true CN114882204A (en) | 2022-08-09 |
Family
ID=82667138
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210215789.7A Pending CN114882204A (en) | 2022-03-07 | 2022-03-07 | Automatic ship name recognition method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114882204A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116543398A (en) * | 2023-07-06 | 2023-08-04 | 浙江华是科技股份有限公司 | Ship name and violation identification method and system |
| CN116977435A (en) * | 2023-09-15 | 2023-10-31 | 广州志正电气有限公司 | Shore power system for automatic identification of ship on shore |
-
2022
- 2022-03-07 CN CN202210215789.7A patent/CN114882204A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116543398A (en) * | 2023-07-06 | 2023-08-04 | 浙江华是科技股份有限公司 | Ship name and violation identification method and system |
| CN116543398B (en) * | 2023-07-06 | 2023-09-08 | 浙江华是科技股份有限公司 | Ship name and violation identification method and system |
| CN116977435A (en) * | 2023-09-15 | 2023-10-31 | 广州志正电气有限公司 | Shore power system for automatic identification of ship on shore |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111325203B (en) | American license plate recognition method and system based on image correction | |
| TWI744283B (en) | Method and device for word segmentation | |
| CN111027493B (en) | Pedestrian detection method based on deep learning multi-network soft fusion | |
| CN111414906A (en) | Data synthesis and text recognition method for paper bill picture | |
| CN107480585B (en) | Target detection method based on DPM algorithm | |
| CN114882204A (en) | Automatic ship name recognition method | |
| CN112528862B (en) | Remote sensing image target detection method based on improved cross entropy loss function | |
| CN112307919B (en) | Improved YOLOv 3-based digital information area identification method in document image | |
| CN110008899B (en) | Method for extracting and classifying candidate targets of visible light remote sensing image | |
| CN110443235B (en) | Intelligent paper test paper total score identification method and system | |
| CN113516116B (en) | Text detection method, system and medium suitable for complex natural scene | |
| CN112287941A (en) | License plate recognition method based on automatic character region perception | |
| CN110827312A (en) | Learning method based on cooperative visual attention neural network | |
| CN112215190A (en) | Illegal building detection method based on YOLOV4 model | |
| CN112348028A (en) | Scene text detection method, correction method, device, electronic equipment and medium | |
| CN116385374A (en) | Cell counting method based on convolutional neural network | |
| CN110458019B (en) | Water surface target detection method for eliminating reflection interference under scarce cognitive sample condition | |
| CN111027538A (en) | Container detection method based on instance segmentation model | |
| CN112364687A (en) | Improved Faster R-CNN gas station electrostatic sign identification method and system | |
| CN116740758A (en) | Bird image recognition method and system for preventing misjudgment | |
| CN114612738B (en) | Training method of cell electron microscope image segmentation model and organelle interaction analysis method | |
| CN114898290A (en) | Real-time detection method and system for marine ship | |
| CN115937095A (en) | Printing defect detection method and system integrating image processing algorithm and deep learning | |
| CN115082551A (en) | Multi-target detection method based on unmanned aerial vehicle aerial video | |
| CN114927236A (en) | Detection method and system for multiple target images |
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 |