CN115527209A - Method, device and system for identifying shore bridge box number and computer equipment - Google Patents
Method, device and system for identifying shore bridge box number and computer equipment Download PDFInfo
- Publication number
- CN115527209A CN115527209A CN202211155058.4A CN202211155058A CN115527209A CN 115527209 A CN115527209 A CN 115527209A CN 202211155058 A CN202211155058 A CN 202211155058A CN 115527209 A CN115527209 A CN 115527209A
- Authority
- CN
- China
- Prior art keywords
- identification
- box number
- image
- recognition
- container
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/146—Aligning or centring of the image pick-up or image-field
- G06V30/147—Determination of region of interest
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/15—Cutting or merging image elements, e.g. region growing, watershed or clustering-based techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The application relates to a method, a device, a system and a computer device for identifying a shore bridge box number, wherein the method for identifying the shore bridge box number comprises the following steps: the method comprises the steps of obtaining an image of a container through a container number recognition system, carrying out first recognition on the image based on a well-trained model to obtain a first recognition result, processing the first recognition result according to a container number combination algorithm, obtaining a target container number when the first recognition result after the judgment processing is successful, carrying out second recognition on the image based on a secondary container number recognition algorithm when the first recognition result after the judgment processing is failed, obtaining a second recognition result, processing the second recognition result according to the container number combination algorithm, and obtaining the target container number when the second recognition result after the judgment processing is successful.
Description
Technical Field
The application relates to the technical field of container number identification, in particular to a method, a device, a system and computer equipment for identifying a quayside container number.
Background
With the advance of the global economy integration process, the resource flow in the global world is accelerated, and the container trade volume is rapidly increased and becomes an important part of the port trade. The container number of the container is used as the unique identifier of the container, and the port operation efficiency can be obviously improved by quickly and accurately identifying the container number of the container, so that the effective management of the container is usually realized based on the identification of the container number.
In the existing box number identification method, characters are extracted from an acquired box number image through an open source framework Tesseract, and the box number of a container is further acquired from the extracted characters. However, under the condition that a large amount of noise exists in the image or the image is not subjected to proper preprocessing, the text recognition rate of Tesseract is low, and the high-intensity transportation makes the surface of the container easily have oil stain or rust and the like, and is influenced by weather and shooting angles, so that the imaging quality of the box number image is often poor, and the box number recognition method cannot effectively extract the text in the box number image.
Aiming at the problem that the characters in the box number image cannot be effectively identified in the related technology, no effective solution is provided at present.
Disclosure of Invention
The embodiment provides a method, a device, a system and computer equipment for identifying a quayside container number, so as to solve the problem that characters in a container number image cannot be effectively identified in the related art.
In a first aspect, the present embodiment provides a method for identifying a quayside container number, which is suitable for a container number identification system; the box number identification system includes: the identification ball machine is mounted on a cross beam of a bridge crane, the control host is used for acquiring operation signals of the bridge crane and controlling the identification ball machine, and the identification ball machine is connected with the control host; the method comprises the following steps:
acquiring an image of the container through the container number identification system;
based on a completely trained model, carrying out first recognition on the image to obtain a first recognition result, processing the first recognition result according to a box number combination algorithm, and obtaining a target box number when the first recognition result after the processing is judged to be successful;
and when the first identification result after the judgment processing fails, performing secondary identification on the image based on a secondary box number identification algorithm to obtain a second identification result, processing the second identification result according to the box number combination algorithm, and when the second identification result after the judgment processing succeeds, obtaining the target box number.
In some embodiments, the acquiring, by the box number identification system, an image of the container comprises:
acquiring an operation signal of the bridge crane through the box number identification system;
presetting the identification dome camera according to the operation signal and a judgment mechanism of a preset check point;
and capturing images of the container through the preset identification ball machine to obtain the images.
In some embodiments, the presetting the identification ball machine according to the working signal and the determination mechanism of the preset check point includes:
and determining the image capturing parameters of the identification dome camera according to the operation signal and the judgment mechanism of the preset check point, and calling the identification dome camera to a preset position.
In some embodiments, the performing, based on the fully trained model, a first recognition on the image to obtain a first recognition result includes:
detecting a key target of the image to obtain a key box number area;
performing text detection on the key box number area through a text detection network for deep learning to obtain an accurate box number area;
and performing text recognition on the accurate box number area through a text recognition network for deep learning to obtain a first recognition result.
In some embodiments, the processing the first recognition result according to the box number combination algorithm includes:
analyzing the spatial position relationship of the first identification result to obtain a first box number;
analyzing the number of characters of the first recognition result to obtain a second box number;
and combining the first box number and the second box number according to a box number rule.
In some embodiments, the performing a second recognition on the image based on the secondary box number recognition algorithm to obtain a second recognition result includes:
image preprocessing is carried out on the key box number area;
according to an image threshold segmentation algorithm, segmenting the preprocessed key box number area to obtain a target box number area;
and classifying and identifying the target box number area according to a classifier algorithm to obtain the second identification result.
In a second aspect, the present embodiment provides an apparatus for identifying a quayside container number, which is suitable for a container number identification system; the box number identification system includes: the identification ball machine is mounted on a cross beam of a bridge crane, the control host is used for acquiring operation signals of the bridge crane and controlling the identification ball machine, and the identification ball machine is connected with the control host; the device comprises:
the acquisition module acquires an image of the container through the box number identification system;
the first recognition module is used for carrying out first recognition on the image based on a model with complete training to obtain a first recognition result, processing the first recognition result according to a box number combination algorithm, and obtaining a target box number when the first recognition result after processing is judged to be successful;
and the second identification module is used for carrying out second identification on the image based on a secondary box number identification algorithm to obtain a second identification result when the first identification result after the judgment processing fails, processing the second identification result according to the box number combination algorithm, and obtaining the target box number when the second identification result after the judgment processing succeeds.
In a third aspect, in this embodiment, a bin number identifying system is provided, including: the identification ball machine is installed on a cross beam of the bridge crane, the control host is used for acquiring operation signals of the bridge crane and controlling the identification ball machine, and the identification ball machine is connected with the control host.
In a fourth aspect, in the present embodiment, there is provided a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the method for identifying a shore connection box number according to the first aspect.
In a fifth aspect, in the present embodiment, there is provided a storage medium having stored thereon a computer program which, when executed by a processor, implements the method for identification of a shore bridge box number as described in the first aspect above.
Compared with the related art, the method, the device, the system and the computer equipment for identifying the quayside container number, which are provided by the embodiment, have the advantages that the image of the container is obtained through the container number identification system, the image is identified for the first time based on the well-trained model to obtain the first identification result, the first identification result is processed according to the container number combination algorithm, the target container number is obtained when the first identification result after the processing is judged to be successful, the image is identified for the second time based on the secondary container number identification algorithm when the first identification result after the processing is judged to be failed, the second identification result is obtained, the second identification result is processed according to the container number combination algorithm, the target container number is obtained when the second identification result after the processing is judged to be successful, the problem that characters in the container number image cannot be effectively identified is solved, and the beneficial effect of accurately obtaining the container number is realized.
The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below to provide a more thorough understanding of the application.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a block diagram of a hardware structure of a terminal device for a method for identifying a shore bridge box number according to an embodiment of the present disclosure;
FIG. 2 is a flow chart of a method for identification of a shore bridge box number provided by an embodiment of the present application;
FIG. 3 is a preferred flow chart of a method for identification of a shore bridge box number provided by an embodiment of the present application;
FIG. 4 is a schematic structural diagram of a system for identifying a quayside container number according to an embodiment of the present application;
fig. 5 is a schematic flowchart of image recognition in a method for identifying a quayside container number according to an embodiment of the present application;
fig. 6 is a schematic flowchart of a box number combination process in the method for identifying a box number of a shore bridge according to an embodiment of the present application;
fig. 7 is a flowchart of ship-unloading operation triggering in the method for identifying a shore bridge box number according to an embodiment of the present application;
fig. 8 is a block diagram of an apparatus for identifying a quayside container number according to an embodiment of the present application.
In the figure: 10. an acquisition module; 20. a first identification module; 30. and a second identification module.
Detailed Description
For a clearer understanding of the objects, technical solutions and advantages of the present application, reference is made to the following description and accompanying drawings.
Unless defined otherwise, technical or scientific terms referred to herein shall have the same general meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The use of the terms "a" and "an" and "the" and similar referents in the context of this application do not denote a limitation of quantity, either in the singular or the plural. The terms "comprises," "comprising," "has," "having," and any variations thereof, as referred to in this application, are intended to cover non-exclusive inclusions; for example, a process, method, and system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to the listed steps or modules, but may include other steps or modules (elements) not listed or inherent to such process, method, article, or apparatus. Reference throughout this application to "connected," "coupled," and the like is not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. Reference to "a plurality" in this application means two or more. "and/or" describes an association relationship of associated objects, meaning that three relationships may exist, for example, "A and/or B" may mean: a exists alone, A and B exist simultaneously, and B exists alone. In general, the character "/" indicates a relationship in which the objects associated before and after are an "or". The terms "first," "second," "third," and the like in this application are used for distinguishing between similar items and not necessarily for describing a particular sequential or chronological order.
The method embodiments provided in the present embodiment may be executed in a terminal, a computer, or a similar computing device. For example, the method is executed on a terminal, and fig. 1 is a hardware configuration block diagram of the terminal for the method for identifying a bank bridge box number according to the embodiment. As shown in fig. 1, the terminal may include one or more processors 102 (only one shown in fig. 1) and a memory 104 for storing data, wherein the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA. The terminal may also include a transmission device 106 for communication functions and an input-output device 108. It will be understood by those of ordinary skill in the art that the structure shown in fig. 1 is merely an illustration and is not intended to limit the structure of the terminal described above. For example, the terminal may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.
The memory 104 can be used for storing computer programs, for example, software programs and modules of application software, such as a computer program corresponding to the method for identifying a shore bridge box number in the present embodiment, and the processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, so as to implement the above-mentioned method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to the terminal over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission device 106 is used to receive or transmit data via a network. The network described above includes a wireless network provided by a communication provider of the terminal. In one example, the transmission device 106 includes a Network adapter (NIC) that can be connected to other Network devices through a base station to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used to communicate with the internet in a wireless manner.
In the present embodiment, a method for identifying a quayside container number is provided, and fig. 2 is a flowchart of the method for identifying a quayside container number of the present embodiment, as shown in fig. 2, the flowchart includes the following steps:
step S210, acquiring an image of the container through the container number identification system.
Specifically, in the box number identification system, the operation signal of the bridge crane is obtained through the control host, the identification ball machine is preset based on a judgment mechanism of a preset check point, and the side face of the box body of the container is subjected to image capture through the preset identification ball machine to complete image acquisition.
It should be noted that the box number identification system comprises an identification ball machine and a control host machine, wherein the identification ball machine is installed on a cross beam of the bridge crane and used for capturing an image of a container, and the control host machine is connected with the identification ball machine and used for acquiring an operation signal of the bridge crane and controlling the identification ball machine.
Further, according to different container types, the acquired images are screened, the required images corresponding to the container types are selected and sent to a box number recognition model which is completely trained, for example, according to the size of the box type, different box face images are selected, and specifically, the conditions that the image faces of a single large box and a single small box for box number recognition have a sea side right, a land side right and a front and back face, and the image faces of double small boxes for box number recognition have a sea side left, a sea side right, a land side left, a land side right, a front and back face and the like.
Fig. 7 is a flowchart of trigger of ship unloading operation in the method for identifying the quayside container number according to the embodiment, as shown in fig. 7, when the sea-side suspension bridge is locked S710, the container passes through No. 1 check point, the sea-land-side dome camera and the tie-beam dome camera adjust preset position S720, and the container passes through No. 2 check point, the sea-land-side dome camera and the tie-beam dome camera shoot four sides, and the tie-beam dome camera adjusts preset position S730; when the land-side suspension bridge is unlocked, the connecting beam dome camera shoots the car number and adjusts the preset position S740, and further, the car number and the box number are identified S750; after the container passes through the No. 3 check point, the connecting beam ball machine adjusts a presetting bit S760; after a single job is completed, the variables are released and the parameters are zeroed S770.
And S220, performing first recognition on the image based on the completely trained model to obtain a first recognition result, processing the first recognition result according to a box number combination algorithm, and obtaining a target box number when the processed first recognition result is judged to be successful.
Specifically, in a deep learning neural network model, firstly, key target detection is performed on a received image to obtain a key target area, and secondly, box number recognition is performed on the key target area through a deep learning text detection network and a deep learning text recognition network to obtain a first recognition result of box number characters.
It needs to be known that, the image is sent to a key target detection network, the image is analyzed and a box number coarse positioning detection model is established, so that the image of the container is detected through the box number coarse positioning detection model, a key target area of the box number is obtained, and in the detection process, the model is optimized through self-adaptive learning; meanwhile, in a text detection network and a text recognition network, the images are analyzed, a box number recognition model is established, the content of box number characters is obtained through the model, and self-adaptive learning is carried out in the recognition process, so that the model is optimized.
And step S230, when the processed first identification result is judged to be failed, carrying out secondary identification on the image based on a secondary box number identification algorithm to obtain a second identification result, processing the second identification result according to a box number combination algorithm, and when the processed second identification result is judged to be successful, obtaining the target box number.
Specifically, if the box number obtained by the first recognition does not pass through box number verification, the image is subjected to second recognition through methods of image preprocessing, image threshold segmentation and classifier single character recognition in sequence, and a second recognition result of the box number character is obtained.
It is to be noted that, after the second recognition result of the box number character is obtained, box number combination and box number verification need to be performed again, and the target box number is obtained if the verification is qualified.
In the existing box number identification method, characters are extracted from an acquired box number image through an open source framework Tesseract, and the box number of a container is further acquired from the extracted characters. However, under the condition that a large amount of noise exists in the image or the image is not subjected to proper preprocessing, the text recognition rate of Tesseract is low, and the high-intensity transportation makes the surface of the container easily have oil stain or rust and the like, and is influenced by weather and shooting angles, so that the imaging quality of the box number image is often poor, and the box number recognition method cannot effectively extract the text in the box number image. In the embodiment, based on a detection network and an identification network in the prior art, a deep learning neural network model is established, the model is continuously optimized through analysis training and adaptive learning, the accuracy of box number identification is improved, further, secondary box number identification is performed on an image which does not pass through primary box number verification, and the accuracy of box number characters in an identification result is improved through image preprocessing, image threshold segmentation and classifier single character identification methods, so that a target box number of secondary identification is obtained, the problem that characters in a box number image cannot be effectively identified is solved, and the beneficial effect of accurately obtaining the box number of a container is achieved.
In some of these embodiments, acquiring an image of the container via the container number identification system comprises the steps of:
step S211, acquiring an operation signal of the bridge crane through a box number identification system;
step S212, presetting the identification dome camera according to the operation signal and a judgment mechanism of a preset check point;
and step S213, capturing images of the container through the preset identification ball machine to obtain images.
Specifically, in the box number identification system, the operation signals of the bridge crane are obtained through the control host, and the operation signals comprise the locking and unlocking signals, the horizontal and vertical displacement, the container size and the like of the bridge crane.
It should be noted that each preset check point has different attributes and each point is independent, so when the box body passes through each preset check point, the identification dome camera is preset based on the attribute information of the preset check point, and the box body is subjected to image capture by the identification dome camera, for example, if the operation type of the bridge crane is ship unloading operation and the sea side lifting appliance is locked, and meanwhile, the locking position in the attribute of No. 1 check point is sea side and the action is preset, when the box body passes through No. 1 check point, the identification dome camera is preset, and the box body is subjected to image capture by the identification dome camera.
Through this embodiment, through the case number identification system, acquire the operation signal of bridge crane to the judgment mechanism that combines the preset check point presets the identification ball machine, thereby carries out image capture to the container through the identification ball machine, obtains corresponding image, carries out multi-angle image capture to the box according to different operation types and check point attribute with this, effectively acquires the image of container.
In some embodiments, the presetting of the ball identification machine according to the working signal and the judgment mechanism of the preset check point includes the following steps:
and determining image capturing parameters of the identification dome camera according to the operation signal and a judgment mechanism of a preset check point, and calling the identification dome camera to a preset position.
Specifically, image capturing parameters of the recognition dome camera are set according to the job signal, including the image capturing range and the number and frequency of frames of image capturing, for example, 5 frames are continuously captured at a frequency of 60 milliseconds/frame.
It should be noted that calling the identification dome camera to the preset position refers to adjusting the identification dome camera to capture the orientation and parameter status of the preset key area.
Through the embodiment, the orientation state and the capturing parameters of the identification dome camera can be set quickly and accurately according to the operation signals and the judgment mechanism of the preset check point, so that the accuracy of capturing the container image is improved.
In some embodiments, the first recognition of the image based on the fully trained model to obtain the first recognition result includes the following steps:
step S221, detecting a key target of the image to obtain a key box number area;
step S222, performing text detection on the key box number area through a text detection network for deep learning to obtain an accurate box number area;
step S223, text recognition is carried out on the accurate box number area through a text recognition network of deep learning, and a first recognition result is obtained.
Specifically, the image is subjected to key target detection, and a key box number area is obtained by predicting the position of a box number character, and the key target detection method comprises an anchor point frame-based target detection method, a key point-based target detection method, an end-to-end prediction target detection method and the like.
It is to be noted that the text detection is performed on the key box number region through a text detection network for deep learning, where the deep learning method in the text detection field includes text detection based on candidate boxes, text detection based on segmentation, text detection based on mixture of candidate boxes and segmentation, and the like, where taking EAST network as an example, first sending the key box number region to a full convolution network, generating a single-channel pixel-level text fractional feature map and a multi-channel geometric feature map, and simultaneously adopting two geometric shapes, namely a rotating box and a horizontal box, to obtain an accurate box number region, where if the score of the accurate box number region exceeds a predetermined threshold, the accurate box number region is an effective accurate box number region.
Further, text recognition is performed on the accurate box number region through a deep learning text recognition Network, here, taking a Convolutional Recurrent Neural Network (CRNN Network for short) as an example, feature extraction is performed on the obtained accurate box number region through a Convolutional layer to obtain a feature map, a feature sequence is predicted in a cyclic layer, each feature vector in the sequence is learned to obtain a predicted label distribution, and then the predicted label distribution is converted into a label sequence in a transcription layer to obtain a corresponding box number character.
According to the method and the device, in the deep learning neural network model, the key target detection, the text detection and the text recognition are carried out on the image, so that an accurate box number recognition result is obtained, and the box number characters in the image can be effectively extracted.
In some embodiments, processing the first recognition result according to the box number combination algorithm includes the following steps:
analyzing the spatial position relation of the first identification result to obtain a first box number;
analyzing the number of characters of the first recognition result to obtain a second box number;
and combining the first box number and the second box number according to the box number rule.
Specifically, fig. 6 is a schematic flow diagram of box number combination processing in the method for identifying a box number of quayside container crane according to this embodiment, and as shown in fig. 6, after obtaining a first identification result, S610, performs spatial position relationship analysis on the first identification result, S620, adjusts an inclination angle and the like of a box number character in the identification result, and combines the box number character number obtained by the analysis with the box number character number S630, and combines the box number character according to a box number rule S640.
It is to be understood that the box number of a standard container consists of an 11-bit code, which, taking the box number FFAU1101750 as an example, includes three parts: the first part 'FFAU' is composed of 4-digit English letters; the second portion "110175" consists of 6-bit numbers; the third part "0" is a check code formed by 1 digit number, the check code is obtained by dividing the first four alphabets and 6 digit number by a modulus 11 through a weighting coefficient and a product, and the rest numbers are taken, and the specific formula is as follows:
wherein CheckNum is a check code C i Corresponding to the first 10 bits of the box number.
According to the method and the device, the space position and the number of the characters of the box number in the identification result are analyzed to obtain two sets of box number information, and further, the box number information is combined according to the box number rule to obtain the standard container number, so that the accuracy of the target box number is improved.
In some embodiments, the second recognition of the image based on the secondary box number recognition algorithm to obtain the second recognition result includes the following steps:
step S231, image preprocessing is carried out on the key box number area;
step S232, segmenting the preprocessed key box number region according to an image threshold segmentation algorithm to obtain a target box number region;
and step S233, classifying and identifying the target box number area according to a classifier algorithm to obtain a second identification result.
Specifically, the key box number area is preprocessed according to a homomorphic filtering image and processing algorithm, the algorithm separates incident light and reflected light to highlight key area information, firstly, logarithm operation is carried out on incident light components and reflected light components, the two components are decomposed in a time domain, and the two components are converted into a frequency domain from the time domain through Fourier transformation, wherein the incident components serve as low-frequency components, and the reflected components serve as high-frequency components, so that the low-frequency components and the high-frequency components of the Fourier transformation are controllably influenced through a limited filtering function, for example, the limited filtering function can approach to attenuate low frequency to enhance the contribution of high frequency, and then a high-pass filtering function of 3*3 is applied to increase character details.
It should be noted that, the preprocessed key box number region is segmented according to an image threshold segmentation algorithm, taking an OTSU method as an example, the foreground and the background of an image are set according to the gray scale of the image, and when the variance between the foreground and the background is larger, it indicates that the difference between two parts forming the image is larger, and thus it indicates that the variance between the two parts is larger, the probability of segmentation error is small, and a specific segmentation calculation formula is as follows:
Further, classification and recognition are carried out on the target box number region according to a classifier algorithm, a Support Vector machine (SVM for short) is mainly adopted for realizing, the SVM is a two-classification model, specifically, the actual image features of 0-9 and A-Z are obtained, a character recognition model is established, the character recognition model is trained through the characters of the target box number region, a corresponding training data set is obtained, then the training data set is correctly divided through the SVM algorithm, classification and recognition of the characters are completed, and a second recognition result is obtained.
According to the method and the device, image preprocessing, an image threshold segmentation algorithm and classification recognition are carried out on the key box number area to obtain a second recognition result, so that secondary recognition is carried out on the container image collected by the system, the box number character recognition rate of the image is improved, and the probability of obtaining an accurate box number is improved.
Fig. 5 is a schematic flowchart of image recognition in the method for identifying a quayside container number according to the embodiment, and as shown in fig. 5, the specific process of the method for identifying a quayside container number is as follows:
image capture S501 is carried out on the side surface of the container body of the container to obtain an image of the container, and the image is preprocessed S502; performing key target detection on the image S503, obtaining a key box number area by predicting the position of box number characters, performing box number character detection on the key box number area through a deep learning text detection network S504 to obtain an accurate box number area, and performing box number character recognition on the accurate box number area through a deep learning text recognition network S505 to obtain a first recognition result; further, the first recognition result is processed according to a box number combination algorithm S506, box number verification is carried out on the first recognition result S507, and when the verification of the first recognition result is successful, a target box number result S512 is obtained; if the first recognition result fails to be verified, performing image preprocessing on the key box number region S508, segmenting the preprocessed key box number region S509 according to an image threshold segmentation algorithm to obtain a target box number region, performing classification recognition on the target box number region S510 according to a classifier algorithm to obtain a second recognition result, and performing box number combination and box number verification on the second recognition result S511; if the verification is successful, a target box number result is obtained S512, and if the verification is failed, box number identification fails S513.
The present embodiment is described and illustrated below by means of preferred embodiments.
Fig. 3 is a preferred flowchart of the method for identifying a shore connection box number according to the present embodiment, and as shown in fig. 3, the method for identifying a shore connection box number includes the following steps:
and S310, acquiring an operation signal of the bridge crane through a box number identification system, presetting an identification ball machine according to the operation signal and a judgment mechanism of a preset check point, and capturing an image of the container through the preset identification ball machine to obtain the image.
Step S320, detecting a key target of the image to obtain a key box number area, performing text detection on the key box number area through a text detection network for deep learning to obtain an accurate box number area, and performing text recognition on the accurate box number area through a text recognition network for deep learning to obtain a first recognition result.
And step S330, processing the first identification result according to a box number combination algorithm, and obtaining the target box number when the processed first identification result is judged to be successful.
Step S340, when the processed first identification result is judged to be failed, image preprocessing is carried out on the key box number area, the preprocessed key box number area is segmented according to an image threshold segmentation algorithm to obtain a target box number area, and classification and identification are carried out on the target box number area according to a classifier algorithm to obtain a second identification result.
And step S350, processing the second identification result according to the box number combination algorithm, and obtaining the target box number when the processed second identification result is judged to be successful.
According to the embodiment, the operation signal of the bridge crane is obtained through the box number recognition system, the recognition dome camera is preset according to the operation signal and a judgment mechanism of a preset check point, the image of the container is captured through the preset recognition dome camera to obtain an image, the image is subjected to key target detection, text detection and text recognition based on a completely trained model to obtain a first recognition result, accurate box number characters are obtained through a deep learning neural network model, further, when the first recognition result after judgment and processing fails, image preprocessing, an image threshold segmentation algorithm and classification recognition are carried out on a key box number area to obtain a second recognition result, the accuracy of the box number characters in the recognition result is improved through secondary processing of the container image, and the recognition result is processed and checked according to the box number algorithm to obtain an effective target box number.
It should be noted that the steps illustrated in the above-described flow diagrams or in the flow diagrams of the figures may be performed in a computer system, such as a set of computer-executable instructions, and that, although a logical order is illustrated in the flow diagrams, in some cases, the steps illustrated or described may be performed in an order different than here.
In this embodiment, a device for identifying a quayside container number is further provided, and the device is used to implement the foregoing embodiment and preferred embodiments, and the description of the device that has been already made is omitted. The terms "module," "unit," "subunit," and the like as used below may implement a combination of software and/or hardware for a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware or a combination of software and hardware is also possible and contemplated.
Fig. 8 is a block diagram of the structure of the apparatus for identifying a shore bridge box number according to the present embodiment, and as shown in fig. 8, the apparatus includes: the acquisition module 10, the first recognition module 20 and the second recognition module 30;
the acquisition module acquires an image of the container through the container number identification system;
the first recognition module is used for carrying out first recognition on the image based on the model with complete training to obtain a first recognition result, processing the first recognition result according to a box number combination algorithm, and obtaining a target box number when the processed first recognition result is judged to be successful;
and the second identification module is used for carrying out second identification on the image based on a secondary box number identification algorithm to obtain a second identification result when the first identification result fails after the judgment processing, processing the second identification result according to a box number combination algorithm, and obtaining the target box number when the second identification result after the judgment processing is successful.
By the device provided by the embodiment, the image of the container is acquired through the box number identification system, the image is identified for the first time based on a well-trained model to obtain a first identification result, the first identification result is processed according to the box number combination algorithm, the target box number is obtained when the first identification result after the processing is judged to be successful, the image is identified for the second time based on the secondary box number identification algorithm when the first identification result after the processing is judged to be failed, the second identification result is obtained, the second identification result is processed according to the box number combination algorithm, and the target box number is obtained when the second identification result after the processing is judged to be successful, so that the problem that characters in the box number image cannot be effectively identified is solved, and the beneficial effect of accurately acquiring the box number of the container is realized.
In some embodiments, on the basis of fig. 8, the apparatus further comprises an image capturing module for acquiring a working signal of the bridge crane through a box number identification system; presetting the identification dome camera according to the operation signal and a judgment mechanism of a preset check point; and capturing images of the container through the preset identification ball machine to obtain images.
In some embodiments, on the basis of fig. 8, the apparatus further includes a presetting module, configured to determine an image capturing parameter of the identification dome camera according to the operation signal and a judgment mechanism of a preset check point, and call the identification dome camera to a preset position.
In some embodiments, on the basis of fig. 8, the apparatus further includes an accurate identification module, configured to perform key target detection on the image, so as to obtain a key box number region; performing text detection on the key box number area through a text detection network for deep learning to obtain an accurate box number area; and performing text recognition on the accurate box number area through a text recognition network for deep learning to obtain a first recognition result.
In some embodiments, on the basis of fig. 8, the apparatus further includes a combination module, configured to perform spatial position relationship analysis on the first recognition result, so as to obtain a first box number; analyzing the number of characters of the first recognition result to obtain a second box number; and combining the first box number and the second box number according to the box number rule.
In some embodiments, on the basis of fig. 8, the apparatus further includes an algorithm identification module for performing image preprocessing on the key box number region; segmenting the preprocessed key box number area according to an image threshold segmentation algorithm to obtain a target box number area; and classifying and identifying the target box number area according to a classifier algorithm to obtain a second identification result.
Fig. 4 is a schematic structural diagram of a system for identifying a shore bridge box number according to the present embodiment, and as shown in fig. 4, the system includes: the identification ball machine is arranged on a cross beam of the bridge crane, the control host is used for acquiring operation signals of the bridge crane and controlling the identification ball machine, and the identification ball machine is connected with the control host.
Specifically, the identification dome camera is installed on a cross beam of the bridge crane and used for capturing images of the container, and the control host is connected with the identification dome camera and used for acquiring operation signals of the bridge crane and controlling the identification dome camera.
It is required to know that, in the box number identification system, the operation signal of the bridge crane is obtained through the control host, the operation logic of the bridge crane is judged, the identification ball machine is preset based on a judgment mechanism of a preset check point, and the side face of the box body of the container is subjected to image capture through the preset identification ball machine to complete image acquisition.
Through the case number recognition system of this embodiment, obtain the operation signal of bridge crane through the control host computer to combine the judgment mechanism of presetting the checkpoint, preset the identification ball machine and carry out image acquisition, can accurately acquire the image of container, in order to carry out subsequent case number recognition.
It should be noted that the above modules may be functional modules or program modules, and may be implemented by software or hardware. For a module implemented by hardware, the modules may be located in the same processor; or the modules can be respectively positioned in different processors in any combination.
There is also provided in this embodiment a computer device comprising a memory having a computer program stored therein and a processor arranged to run the computer program to perform the steps of any of the method embodiments described above.
Optionally, the computer device may further include a transmission device and an input/output device, wherein the transmission device is connected to the processor, and the input/output device is connected to the processor.
It should be noted that, for specific examples in this embodiment, reference may be made to the examples described in the foregoing embodiments and optional implementations, and details are not described again in this embodiment.
In addition, in combination with the method for identifying a shore bridge box number provided in the above embodiment, a storage medium may also be provided to implement in this embodiment. The storage medium having stored thereon a computer program; the computer program, when executed by a processor, implements any of the above-described embodiments of a method for identification of a shore bridge box number.
It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to be limiting. All other embodiments, which can be derived by a person skilled in the art from the examples provided herein without any inventive step, shall fall within the scope of protection of the present application.
It is obvious that the drawings are only examples or embodiments of the present application, and it is obvious to those skilled in the art that the present application can be applied to other similar cases according to the drawings without creative efforts. Moreover, it should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another.
The term "embodiment" is used herein to mean that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is to be expressly or implicitly understood by one of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the patent protection. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.
Claims (10)
1. A method for identifying a quayside container number is characterized by being suitable for a container number identification system; the box number identification system includes: the identification ball machine is mounted on a cross beam of a bridge crane, the control host is used for acquiring operation signals of the bridge crane and controlling the identification ball machine, and the identification ball machine is connected with the control host; the method comprises the following steps:
acquiring an image of the container through the container number identification system;
based on a completely trained model, carrying out first recognition on the image to obtain a first recognition result, processing the first recognition result according to a box number combination algorithm, and obtaining a target box number when the first recognition result after the processing is judged to be successful;
and when the first identification result after the judgment processing fails, performing secondary identification on the image based on a secondary box number identification algorithm to obtain a second identification result, processing the second identification result according to the box number combination algorithm, and when the second identification result after the judgment processing succeeds, obtaining the target box number.
2. The method for shore bridge box number identification as claimed in claim 1, wherein said acquiring an image of a container by said box number identification system comprises:
acquiring an operation signal of the bridge crane through the box number identification system;
presetting the identification dome camera according to the operation signal and a judgment mechanism of a preset check point;
and capturing images of the container through the preset identification ball machine to obtain the images.
3. The method for identifying the quay crane box number according to claim 2, wherein the presetting of the identification ball machine according to the operation signal and a judgment mechanism of a preset check point comprises the following steps:
and determining the image capturing parameters of the identification dome camera according to the operation signal and the judgment mechanism of the preset check point, and calling the identification dome camera to a preset position.
4. The method for identifying the shore bridge box number according to claim 1, wherein the first identifying the image based on the well-trained model to obtain a first identification result comprises:
detecting a key target of the image to obtain a key box number area;
performing text detection on the key box number area through a text detection network for deep learning to obtain an accurate box number area;
and performing text recognition on the accurate box number area through a text recognition network for deep learning to obtain a first recognition result.
5. The method for identification of a shore bridge box number according to claim 1, wherein said processing said first identification result according to a box number combining algorithm comprises:
analyzing the spatial position relationship of the first identification result to obtain a first box number;
analyzing the number of characters of the first recognition result to obtain a second box number;
and combining the first box number and the second box number according to a box number rule.
6. The method for identifying the quayside container number according to claim 1, wherein the second identification of the image based on the secondary container number identification algorithm is carried out, and the obtaining of the second identification result comprises:
image preprocessing is carried out on the key box number area;
according to an image threshold segmentation algorithm, segmenting the preprocessed key box number area to obtain a target box number area;
and classifying and identifying the target box number area according to a classifier algorithm to obtain the second identification result.
7. A device for identifying a quayside container number is characterized by being suitable for a container number identification system; the box number identification system includes: the identification ball machine is mounted on a cross beam of a bridge crane, the control host is used for acquiring operation signals of the bridge crane and controlling the identification ball machine, and the identification ball machine is connected with the control host; the device comprises:
the acquisition module acquires an image of the container through the box number identification system;
the first recognition module is used for carrying out first recognition on the image based on a completely trained model to obtain a first recognition result, processing the first recognition result according to a box number combination algorithm, and obtaining a target box number when the first recognition result after processing is judged to be successful;
and the second identification module is used for carrying out secondary identification on the image based on a secondary box number identification algorithm to obtain a second identification result when the first identification result after judgment and processing fails, processing the second identification result according to the box number combination algorithm, and obtaining the target box number when the second identification result after judgment and processing succeeds.
8. A case number identification system, the system comprising: the identification ball machine is installed on a cross beam of the bridge crane, the control host is used for acquiring operation signals of the bridge crane and controlling the identification ball machine, and the identification ball machine is connected with the control host.
9. Computer arrangement comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the steps of the method for identification of a shore bridge box number according to any of claims 1 to 6.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for identification of a shore connection box number according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211155058.4A CN115527209A (en) | 2022-09-22 | 2022-09-22 | Method, device and system for identifying shore bridge box number and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211155058.4A CN115527209A (en) | 2022-09-22 | 2022-09-22 | Method, device and system for identifying shore bridge box number and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115527209A true CN115527209A (en) | 2022-12-27 |
Family
ID=84699332
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211155058.4A Pending CN115527209A (en) | 2022-09-22 | 2022-09-22 | Method, device and system for identifying shore bridge box number and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115527209A (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106203239A (en) * | 2015-05-04 | 2016-12-07 | 杭州海康威视数字技术股份有限公司 | Information processing method, device and system for container tally |
| CN106210616A (en) * | 2015-05-04 | 2016-12-07 | 杭州海康威视数字技术股份有限公司 | The acquisition method of container representation information, device and system |
| CN107368827A (en) * | 2017-04-01 | 2017-11-21 | 阿里巴巴集团控股有限公司 | Character identifying method and device, user equipment, server |
| CN107832767A (en) * | 2017-11-14 | 2018-03-23 | 深圳码隆科技有限公司 | Container number identification method, device and electronic equipment |
| CN108596166A (en) * | 2018-04-13 | 2018-09-28 | 华南师范大学 | A kind of container number identification method based on convolutional neural networks classification |
| CN109886077A (en) * | 2018-12-28 | 2019-06-14 | 北京旷视科技有限公司 | Image-recognizing method, device, computer equipment and storage medium |
| CN110070324A (en) * | 2019-04-24 | 2019-07-30 | 大连大学 | A kind of container terminal intelligence tallying system |
| CN110210433A (en) * | 2019-06-10 | 2019-09-06 | 南京大学 | A kind of container number detection and recognition methods based on deep learning |
| CN110555441A (en) * | 2019-09-10 | 2019-12-10 | 杭州橙鹰数据技术有限公司 | character recognition method and device |
| CN110852324A (en) * | 2019-08-23 | 2020-02-28 | 上海撬动网络科技有限公司 | Deep neural network-based container number detection method |
| CN110942057A (en) * | 2018-09-25 | 2020-03-31 | 杭州海康威视数字技术股份有限公司 | Container number identification method and device and computer equipment |
| CN113408512A (en) * | 2021-06-03 | 2021-09-17 | 云从科技集团股份有限公司 | Method, system, device and medium for checking container by using robot |
| CN113971811A (en) * | 2021-11-16 | 2022-01-25 | 北京国泰星云科技有限公司 | Intelligent container feature identification method based on machine vision and deep learning |
-
2022
- 2022-09-22 CN CN202211155058.4A patent/CN115527209A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106210616A (en) * | 2015-05-04 | 2016-12-07 | 杭州海康威视数字技术股份有限公司 | The acquisition method of container representation information, device and system |
| CN106203239A (en) * | 2015-05-04 | 2016-12-07 | 杭州海康威视数字技术股份有限公司 | Information processing method, device and system for container tally |
| CN107368827A (en) * | 2017-04-01 | 2017-11-21 | 阿里巴巴集团控股有限公司 | Character identifying method and device, user equipment, server |
| CN107832767A (en) * | 2017-11-14 | 2018-03-23 | 深圳码隆科技有限公司 | Container number identification method, device and electronic equipment |
| CN108596166A (en) * | 2018-04-13 | 2018-09-28 | 华南师范大学 | A kind of container number identification method based on convolutional neural networks classification |
| CN110942057A (en) * | 2018-09-25 | 2020-03-31 | 杭州海康威视数字技术股份有限公司 | Container number identification method and device and computer equipment |
| CN109886077A (en) * | 2018-12-28 | 2019-06-14 | 北京旷视科技有限公司 | Image-recognizing method, device, computer equipment and storage medium |
| CN110070324A (en) * | 2019-04-24 | 2019-07-30 | 大连大学 | A kind of container terminal intelligence tallying system |
| CN110210433A (en) * | 2019-06-10 | 2019-09-06 | 南京大学 | A kind of container number detection and recognition methods based on deep learning |
| CN110852324A (en) * | 2019-08-23 | 2020-02-28 | 上海撬动网络科技有限公司 | Deep neural network-based container number detection method |
| CN110555441A (en) * | 2019-09-10 | 2019-12-10 | 杭州橙鹰数据技术有限公司 | character recognition method and device |
| CN113408512A (en) * | 2021-06-03 | 2021-09-17 | 云从科技集团股份有限公司 | Method, system, device and medium for checking container by using robot |
| CN113971811A (en) * | 2021-11-16 | 2022-01-25 | 北京国泰星云科技有限公司 | Intelligent container feature identification method based on machine vision and deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tang et al. | Compressed-domain ship detection on spaceborne optical image using deep neural network and extreme learning machine | |
| US20080193020A1 (en) | Method for Facial Features Detection | |
| CN108806334A (en) | A kind of intelligent ship personal identification method based on image | |
| US9460367B2 (en) | Method and system for automating an image rejection process | |
| CN112560634B (en) | Power line insulator rapid detection and positioning method and system based on field image | |
| Ahamed et al. | HOG-CNN based real time face recognition | |
| CN113128481A (en) | Face living body detection method, device, equipment and storage medium | |
| CN112052830B (en) | Method, device and computer storage medium for face detection | |
| CN111539986A (en) | Target tracking method and device, computer equipment and storage medium | |
| Tueller et al. | Target detection using features for sonar images | |
| US8655079B2 (en) | Reducing false alarms in identifying whether a candidate image is from an object class | |
| CN111079626B (en) | Living body fingerprint identification method, electronic equipment and computer readable storage medium | |
| CN110679586B (en) | Bird repelling method and system for power transmission network and computer readable storage medium | |
| CN103810487A (en) | Method and system for target detection and identification of aerial ocean images | |
| CN115527209A (en) | Method, device and system for identifying shore bridge box number and computer equipment | |
| CN111476233A (en) | License plate number positioning method and device | |
| CN113361455B (en) | Training method of face counterfeit identification model, related device and computer program product | |
| CN115393743A (en) | Vehicle detection method based on double-branch encoding and decoding network, unmanned aerial vehicle and medium | |
| Zohra et al. | Adaptive face recognition based on image quality | |
| CN112532938B (en) | Video monitoring system based on big data technology | |
| CN113963178A (en) | Method, device, equipment and medium for detecting infrared dim and small target under ground-air background | |
| CN110717490A (en) | Bird nest identification method and system for power transmission line tower and storage medium | |
| Wei et al. | Petrochemical equipment detection by improved YOLOv5 with multiscale deep feature fusion and attention mechanism | |
| CN110969210A (en) | Small and slow target identification and classification method, device, equipment and storage medium | |
| Zhang et al. | [Retracted] Fast Target Detection Algorithm Based on CFAR and Target Variance Characteristics |
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 |