CN111639513A - Ship shielding identification method and device and electronic equipment - Google Patents
Ship shielding identification method and device and electronic equipment Download PDFInfo
- Publication number
- CN111639513A CN111639513A CN201911261163.4A CN201911261163A CN111639513A CN 111639513 A CN111639513 A CN 111639513A CN 201911261163 A CN201911261163 A CN 201911261163A CN 111639513 A CN111639513 A CN 111639513A
- Authority
- CN
- China
- Prior art keywords
- ship
- feature
- image
- features
- shielding
- 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
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of image recognition, in particular to a ship shielding recognition method and device and electronic equipment. The method comprises the following steps: acquiring an image to be identified, preprocessing the acquired image to be identified, and outputting a preprocessed image; performing region selection on the preprocessed image to generate a preliminary selection region; performing feature extraction on the preliminarily selected area through a convolutional neural network to obtain first ship features and performing feature extraction on the preliminarily selected area according to preset features to obtain second ship features; performing feature fusion on the first ship feature and the second ship feature to form a feature fusion network; training a feature fusion network to obtain a ship shielding recognition model; and identifying the ship shielding by using the ship shielding model, and outputting an identification result. Through first ship characteristic and second ship characteristic to construct ship recognition model and carry out quick ship to the ship that traveles and shelter from discernment, and then reduce the wrong report probability of electron purse seine, improve maritime affairs supervision efficiency.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a ship shielding recognition method and device and electronic equipment.
Background
In the existing electronic purse net, maritime personnel can monitor ships running in a sea area range through video monitoring, but the problem of ship shielding always exists in the monitoring process; for example: a situation where the vessel is obscured by stationary objects in the background, or by other moving vessels. The false alarm rate of the electronic purse net is increased due to the ship shielding problem, and the maritime supervision efficiency is also influenced.
Disclosure of Invention
In view of this, embodiments of the present invention provide a ship occlusion identification method, apparatus, and electronic device, so as to solve the problem of identifying whether a ship is occluded.
According to a first aspect, an embodiment of the present invention provides a ship occlusion identification method, including:
acquiring an image to be identified, preprocessing the acquired image to be identified, and outputting a preprocessed image;
according to the area selection of the preprocessed image, generating a preliminary selection area;
performing feature extraction on the preliminary selection area through a convolutional neural network to obtain first ship features, and performing feature extraction on the preliminary selection area according to preset features to obtain second ship features;
performing feature fusion on the first ship feature and the second ship feature to form a feature fusion network;
training a feature fusion network to obtain a ship occlusion recognition model;
and carrying out ship shielding identification on the image to be identified by utilizing the ship shielding model, and outputting an identification result.
The first ship feature and the second ship feature with the image to be recognized are extracted through the convolutional neural network, ship shielding recognition is carried out on a running ship through building a ship recognition model, then the false alarm probability of the electronic purse net is reduced, and the maritime supervision efficiency is improved.
With reference to the first aspect, in a first implementation manner of the first aspect, acquiring an image to be recognized, preprocessing the acquired image to be recognized, and outputting a preprocessed image includes: acquiring multi-frame image data from a video database, and sequentially carrying out graying, image data enhancement and image data adjustment on the acquired image data.
In extracting the image from video monitoring in the past, the model of guaranteeing to train can accord with actual demand, and carry out graying, image data reinforcing and image data adjustment to the picture data who obtains before sheltering from the discernment to the ship thereby it can guarantee to discern the ship condition of sheltering from to make things convenient for subsequent step to carry out the feature extraction.
With reference to the first aspect, in a second implementation manner of the first aspect, generating a preliminary selection region by performing region selection on the preprocessed image includes: and processing the preprocessed image by using an RPN network to generate a rectangular preliminary selection area.
The preprocessed image is selected through the RPN, so that unnecessary environmental data is eliminated, and only data related to ships are extracted in subsequent ship shielding identification, so that identification pictures are reduced, and the ship shielding identification efficiency is improved.
With reference to the first aspect, in a third implementation manner of the first aspect, performing feature extraction on the preliminarily selected area according to a convolutional neural network to obtain a first ship feature includes: and sending the image data of the preliminarily selected area into a convolutional neural network, and outputting the first ship characteristic after passing through a convolutional layer, an activation layer and a normalization layer.
And the convolution neural network is utilized to carry out feature selection on the preliminarily selected area, so that ships with shielding overlapping can be identified quickly and efficiently.
With reference to the first aspect, in a fourth embodiment of the first aspect, the preset features include: conventional invariant moment features and LOMO features.
By utilizing the traditional characteristics and the LOMO characteristics, enough characteristic points can be ensured, the characteristic points are easy to distinguish, and the identification efficiency for identifying ship shielding cannot be influenced in the calculation process when the characteristics are calculated.
With reference to the first aspect, in a fifth embodiment of the first aspect, a feature fusion network is formed by feature fusion of a first ship feature and a second ship feature, and includes:
and respectively inputting the first ship feature and the second ship feature into a full connection layer, connecting the input of the two full connection layers into a feature connection layer, connecting the output of the feature connection layer with a full connection layer, and outputting a calculation result after calculating a loss function.
The ship feature fusion network comprises 3 full connection layers and 1 feature connection layer, wherein the first two full connection layers are used for integrating multi-feature data in the same feature space, connecting different features through the middle feature connection layer and transmitting the different features to the last full connection layer in a forward mode, and a final result is output after calculation. By means of feature fusion, backward propagation gradient in the network is adjusted, so that the network can be restrained and adjusted, and ship detection results under the shielding condition can be optimized.
With reference to the first aspect, in a sixth implementation manner of the first aspect, before performing feature extraction on the first ship feature and the second ship feature, the method further includes: and performing dimensionality reduction on the preliminarily selected area.
According to a second aspect, an embodiment of the present invention provides a ship shelter recognition device, including:
the acquisition module is used for acquiring an image to be recognized, preprocessing the acquired image to be recognized and outputting a preprocessed image;
the region selection module is used for performing region selection on the preprocessed image to generate a preliminary selected region;
the feature extraction module is used for extracting features of the preliminary selection area by the convolutional neural network to obtain first ship features, and extracting the features of the preliminary selection area according to preset features to obtain second ship features;
the characteristic fusion module is used for carrying out characteristic fusion on the first ship characteristic and the second ship characteristic to form a characteristic fusion network;
the training module is used for training the feature fusion network to obtain a ship shielding recognition model;
and the identification module is used for carrying out ship shielding identification on the image to be identified by the ship shielding model.
Through the data interaction between the modules and the execution of the preset control command, ship shielding recognition is carried out on a running ship, the false alarm probability of the electronic purse net is reduced, and the maritime supervision efficiency is improved.
According to a third aspect, an embodiment of the present invention provides an electronic device, including: a memory and a processor, the memory and the processor are communicatively connected with each other, the memory stores computer instructions, and the processor executes the computer instructions to execute the ship occlusion identification method described in the first aspect or any one of the embodiments of the first aspect.
According to a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores computer instructions for causing a computer to execute the ship occlusion identification method of the first aspect or any one of the implementation manners of the first aspect.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a vessel occlusion identification method according to an embodiment of the invention;
fig. 2 is a block diagram of a ship block recognition apparatus according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a hardware structure of a ship occlusion identification electronic device according to an embodiment of the present invention.
Reference numerals
10-an acquisition module; 20-a selection module; 30-a feature extraction module; 40-a feature fusion module; 50-a training module; 60-an identification module;
301-a processor; 302-bus; 303-a communication interface; 304-memory.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention provides a ship shielding identification method, as shown in fig. 1, comprising the following steps:
s10, acquiring an image to be recognized, preprocessing the acquired image to be recognized and outputting a preprocessed image;
specifically, acquiring an image to be recognized, preprocessing the acquired image to be recognized, and outputting a preprocessed image includes: acquiring multi-frame image data from a video database, and sequentially carrying out graying, image data enhancement and image data adjustment on the acquired image data. In extracting the image from video monitoring in the past, the model of guaranteeing to train can compound actual demand, and carry out graying, image data reinforcing and image data adjustment to the picture data who obtains before sheltering from the discernment to the ship in order to make things convenient for the characteristic of follow-up step to draw, thereby guarantee to shelter from the condition to discern the ship.
For example: extracting frame image data from a marine monitoring area of the electronic purse net, wherein the acquired data is coastal area monitoring video frame data under an illumination condition, and preprocessing the image data, wherein the preprocessing needs to sequentially perform image graying, image enhancement and image adjustment, wherein the image enhancement can be image filtering, for example: mean filtering, median filtering, etc. Image details are highlighted through image filtering, and feature extraction is facilitated.
S20, performing region selection on the preprocessed image to generate a preliminary selection region;
specifically, the generating of the preliminary selection region by performing region selection on the preprocessed image includes: and processing the preprocessed image by using an RPN network to generate a rectangular preliminary selection area. The preprocessed image is selected through the RPN, so that unnecessary environmental data is eliminated, and only data related to ships are extracted in subsequent ship shielding identification, so that identification pictures are reduced, and the ship shielding identification efficiency is improved.
For example: generating 1000 candidate regions for each picture, selecting the candidate regions with the region coincidence degree with the real target region in the preprocessed image being more than 0.8 as positive samples, selecting the candidate regions with the IOU being less than 0.2 as negative samples, and providing the positive and negative samples for a subsequent network for feature extraction, wherein the number of the positive samples is 64.
S30, performing feature extraction on the preliminary selection area through a convolutional neural network to obtain first ship features, and performing feature extraction on the preliminary selection area according to preset features to obtain second ship features; the method comprises the steps of extracting a first ship feature and a second ship feature with an image to be recognized through a convolutional neural network, carrying out quick ship shielding recognition on a running ship by constructing a ship recognition model, further reducing the false alarm probability of the electronic purse net, and improving the monitoring efficiency of the electronic purse net.
Specifically, the convolutional neural network performs feature extraction on the preliminarily selected region to obtain a first ship feature, and the method includes: and sending the image data of the preliminarily selected area into a convolutional neural network, and outputting the first ship characteristic after passing through a convolutional layer, an activation layer and a normalization layer. And the convolution neural network is utilized to carry out feature selection on the preliminarily selected area, so that ships with shielding overlapping can be identified quickly and efficiently.
For example: and (3) performing feature extraction on the preliminary candidate region:
extracting convolution characteristics, inputting the convolution characteristics into an original picture by utilizing a convolutional neural network, wherein the original picture comprises a convolutional layer 1, an active layer 1, a normalization layer 1, a pooling layer 1, a convolutional layer 2, an active layer 2, a normalization layer 2, a pooling layer 2, a convolutional layer 3, an active layer 3, a convolutional layer 4, an active layer 4, a convolutional layer 5, an active layer 5 and an interested region pooling layer ROI, and outputting a characteristic diagram of the candidate region by accessing the output of the active layer 5 and a candidate region extracted by RPN into the interested region pooling layer ROI.
Extracting preset features, for example, extracting SIFT features, generating a scale space and extreme points, accurately positioning key points and deleting unstable points after the extreme points are determined; determining the direction of the key point; and finally, calculating the feature vector and outputting the SIFT feature vector.
S40, performing feature fusion on the first ship feature and the second ship feature to form a feature fusion network;
specifically, a feature fusion network is formed by performing feature fusion on a first ship feature and a second ship feature, and the feature fusion network includes: and respectively inputting the first ship feature and the second ship feature into a full connection layer, connecting the input of the two full connection layers into a feature connection layer, connecting the output of the feature connection layer with a full connection layer, and outputting a calculation result after calculating a loss function. The ship feature fusion network comprises 3 full connection layers and 1 feature connection layer, wherein the first two full connection layers are used for integrating multi-feature data in the same feature space, connecting different features through the middle feature connection layer and transmitting the different features to the last full connection layer in a forward mode, and a final result is output after calculation. By means of feature fusion, backward propagation gradient in the network is adjusted, so that the network can be restrained and adjusted, and ship detection results under the shielding condition can be optimized.
S50, training the feature fusion network to obtain a ship occlusion recognition model;
and S60, carrying out ship occlusion recognition on the image to be recognized by utilizing the ship occlusion model, and outputting a recognition result. For example: the output model may be validated and tested using the test data. After the verification is passed, inputting the image to be detected to the trained feature fusion network, and obtaining the ship target detection result.
Optionally, the preset features include: conventional invariant moment features and LOMO features. By utilizing the traditional characteristics and the LOMO characteristics, enough characteristic points can be ensured, the characteristic points are easy to distinguish, and the identification efficiency for identifying ship shielding cannot be influenced in the calculation process when the characteristics are calculated. The traditional invariant moment feature and the LOMO feature represent local details of a target, the features keep invariance to translation, rotation, scale scaling and brightness change, and keep stability to view angle change, affine change and noise to a certain degree.
Optionally, before the feature extraction is performed on the first ship feature and the second ship feature, the method further includes: and performing dimensionality reduction on the preliminarily selected area.
Specifically, after the extracted features are obtained, dimension reduction processing is also required, and for each candidate region, the features of the image block in the vertical direction are extracted in the form of a sliding window. The generated candidate area is traversed from top to bottom and from left to right by the sliding window, so that the matching degree of the features is improved.
Optionally, besides monitoring by using the existing monitoring device, the monitoring device can also use an unmanned aerial vehicle or an aerial photography device to shoot monitoring videos.
An embodiment of the present invention provides a ship shielding recognition apparatus, as shown in fig. 2, including:
the acquisition module 10 is configured to acquire an image to be recognized, preprocess the acquired image to be recognized, and output a preprocessed image;
a selecting module 20, configured to perform region selection on the preprocessed image to generate a preliminary selected region;
a feature extraction module 30, configured to perform feature extraction on the preliminary selected region by using a convolutional neural network to obtain a first ship feature, and perform feature extraction on the preliminary selected region according to a preset feature to obtain a second ship feature;
the feature fusion module 40 is used for performing feature fusion on the first ship features and the second ship features to form a feature fusion network;
the training module 50 is used for training the feature fusion network to obtain a ship occlusion recognition model;
and the identification module 60 is used for carrying out ship occlusion identification on the image to be identified by the ship occlusion model.
The image data is acquired by the acquisition module 10, the image data selection module is used for carrying out area framing on the image data, then the characteristic extraction module is used for extracting characteristic data from the framed area, and the extracted characteristic data is fused by the characteristic fusion module to obtain a fusion characteristic network; training the fusion characteristic network through a training module, wherein the training result is a training model, and finally, putting data to be identified into the training model for identification so as to obtain an identification result; utilize the device to carry out ship to the ship of traveling and shelter from discernment, and then reduce the wrong report probability of electron purse seine, improve maritime affairs supervision efficiency.
An embodiment of the present invention further provides an electronic device, as shown in fig. 3, the device may include a processor 301 and a memory 302, where the processor 301 and the memory 302 may be connected through a bus 303 or in another manner, and fig. 3 takes the connection through the bus 303 as an example.
The memory 302 is a non-transitory computer-readable storage medium, and can be used for storing non-transitory software programs, non-transitory computer-executable programs, and modules, such as program instructions/modules corresponding to the ship occlusion recognition method in the embodiment of the present invention (for example, the acquisition module 10, the selection module 20, the feature extraction module 30, the feature fusion module 40, the training module 50, and the recognition module 60 shown in fig. 2). The processor 301 executes various functional applications and data processing of the processor by running non-transitory software programs, instructions and modules stored in the memory 302, namely, implements the ship occlusion identification method in the above method embodiment.
The memory 302 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created by the processor 301, and the like. Further, the memory 302 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 302 may optionally include memory located remotely from the processor 301, which may be connected to the processor 301 via 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 one or more modules are stored in the memory 302 and when executed by the processor 301 perform the vessel occlusion identification method as in the embodiment shown in fig. 1.
The details of the electronic device may be understood with reference to the corresponding descriptions and effects in the embodiments shown in fig. 1 to fig. 2, and are not described herein again.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD) or a Solid State Drive (SSD), etc.; the storage medium may also comprise a combination of memories of the kind described above.
Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope defined by the appended claims.
Claims (10)
1. A ship occlusion identification method is characterized by comprising the following steps:
acquiring an image to be identified, preprocessing the acquired image to be identified, and outputting a preprocessed image;
according to the area selection of the preprocessed image, generating a preliminary selection area;
performing feature extraction on the preliminary selection area through a convolutional neural network to obtain first ship features, and performing feature extraction on the preliminary selection area according to preset features to obtain second ship features;
performing feature fusion on the first ship feature and the second ship feature to form a feature fusion network;
training a feature fusion network to obtain a ship occlusion recognition model;
and carrying out ship shielding identification on the image to be identified by utilizing the ship shielding model, and outputting an identification result.
2. The method according to claim 1, wherein the acquiring the image to be recognized, preprocessing the acquired image to be recognized, and outputting a preprocessed image; the method comprises the following steps: acquiring multi-frame image data from a video database, and sequentially carrying out graying, image data enhancement and image data adjustment on the acquired image data.
3. The method of claim 1, wherein generating a preliminary selection region by performing region selection on the preprocessed image comprises: and processing the preprocessed image by using an RPN network to generate a rectangular preliminary selection area.
4. The method of claim 1, wherein said performing feature extraction on said preliminarily selected region according to a convolutional neural network to obtain first vessel features comprises:
and sending the image data of the preliminarily selected area into a convolutional neural network, and outputting the first ship characteristic after passing through a convolutional layer, an activation layer and a normalization layer.
5. The method of claim 1, wherein the preset features comprise: conventional invariant moment features and LOMO features.
6. The method of claim 1, wherein said forming a feature fusion network by feature fusing a first ship feature and a second ship feature comprises:
and respectively inputting the first ship feature and the second ship feature into a full connection layer, connecting the input of the two full connection layers into a feature connection layer, connecting the output of the feature connection layer with a full connection layer, and outputting a calculation result after calculating a loss function.
7. The method of claim 1, further comprising, prior to said feature extracting the first vessel feature and the second vessel feature: and performing dimensionality reduction on the preliminarily selected area.
8. A ship shelter recognition device, comprising:
the acquisition module is used for acquiring an image to be recognized, preprocessing the acquired image to be recognized and outputting a preprocessed image;
the region selection module is used for performing region selection on the preprocessed image to generate a preliminary selected region;
the feature extraction module is used for extracting features of the preliminary selection area by the convolutional neural network to obtain first ship features, and extracting the features of the preliminary selection area according to preset features to obtain second ship features;
the characteristic fusion module is used for carrying out characteristic fusion on the first ship characteristic and the second ship characteristic to form a characteristic fusion network;
the training module is used for training the feature fusion network to obtain a ship shielding recognition model;
and the identification module is used for carrying out ship shielding identification on the image to be identified by the ship shielding model.
9. An electronic device, comprising:
a memory and a processor, the memory and the processor being communicatively connected to each other, the memory having stored therein computer instructions, the processor executing the computer instructions to perform the ship occlusion identification method of any of claims 1-7.
10. A computer-readable storage medium storing computer instructions for causing a computer to perform the vessel occlusion recognition method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911261163.4A CN111639513A (en) | 2019-12-10 | 2019-12-10 | Ship shielding identification method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911261163.4A CN111639513A (en) | 2019-12-10 | 2019-12-10 | Ship shielding identification method and device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111639513A true CN111639513A (en) | 2020-09-08 |
Family
ID=72332430
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911261163.4A Pending CN111639513A (en) | 2019-12-10 | 2019-12-10 | Ship shielding identification method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111639513A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111612028A (en) * | 2019-12-13 | 2020-09-01 | 珠海大横琴科技发展有限公司 | Ship feature optimization method and device based on deep learning and electronic equipment |
| CN112148723A (en) * | 2020-10-15 | 2020-12-29 | 珠海大横琴科技发展有限公司 | Abnormal data optimization method and device based on electronic purse net and electronic equipment |
| CN112270326A (en) * | 2020-11-18 | 2021-01-26 | 珠海大横琴科技发展有限公司 | Detection optimization method and device for ship sheltering and electronic equipment |
| CN112308010A (en) * | 2020-11-12 | 2021-02-02 | 珠海大横琴科技发展有限公司 | Ship shielding detection method and device based on YOLO-V3 algorithm |
| CN112395993A (en) * | 2020-11-18 | 2021-02-23 | 珠海大横琴科技发展有限公司 | Method and device for detecting ship sheltered based on monitoring video data and electronic equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109376591A (en) * | 2018-09-10 | 2019-02-22 | 武汉大学 | The ship object detection method of deep learning feature and visual signature joint training |
| CN109657533A (en) * | 2018-10-27 | 2019-04-19 | 深圳市华尊科技股份有限公司 | Pedestrian recognition methods and Related product again |
| CN109740665A (en) * | 2018-12-29 | 2019-05-10 | 珠海大横琴科技发展有限公司 | Shielded image ship object detection method and system based on expertise constraint |
| CN109815967A (en) * | 2019-02-28 | 2019-05-28 | 北京环境特性研究所 | CNN ship seakeeping system and method based on Fusion Features |
| CN109902618A (en) * | 2019-02-26 | 2019-06-18 | 青岛海之声科技有限公司 | A kind of sea ship recognition methods and device |
| WO2019126755A1 (en) * | 2017-12-21 | 2019-06-27 | Fugro N.V. | Generating and classifying training data for machine learning functions |
| CN110414414A (en) * | 2019-07-25 | 2019-11-05 | 合肥工业大学 | SAR image Ship Target discrimination method based on the fusion of multi-layer depths of features |
-
2019
- 2019-12-10 CN CN201911261163.4A patent/CN111639513A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019126755A1 (en) * | 2017-12-21 | 2019-06-27 | Fugro N.V. | Generating and classifying training data for machine learning functions |
| CN109376591A (en) * | 2018-09-10 | 2019-02-22 | 武汉大学 | The ship object detection method of deep learning feature and visual signature joint training |
| CN109657533A (en) * | 2018-10-27 | 2019-04-19 | 深圳市华尊科技股份有限公司 | Pedestrian recognition methods and Related product again |
| CN109740665A (en) * | 2018-12-29 | 2019-05-10 | 珠海大横琴科技发展有限公司 | Shielded image ship object detection method and system based on expertise constraint |
| CN109902618A (en) * | 2019-02-26 | 2019-06-18 | 青岛海之声科技有限公司 | A kind of sea ship recognition methods and device |
| CN109815967A (en) * | 2019-02-28 | 2019-05-28 | 北京环境特性研究所 | CNN ship seakeeping system and method based on Fusion Features |
| CN110414414A (en) * | 2019-07-25 | 2019-11-05 | 合肥工业大学 | SAR image Ship Target discrimination method based on the fusion of multi-layer depths of features |
Non-Patent Citations (1)
| Title |
|---|
| 种衍杰等: "基于特征融合网络的行人重识别", 《计算机***应用》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111612028A (en) * | 2019-12-13 | 2020-09-01 | 珠海大横琴科技发展有限公司 | Ship feature optimization method and device based on deep learning and electronic equipment |
| CN112148723A (en) * | 2020-10-15 | 2020-12-29 | 珠海大横琴科技发展有限公司 | Abnormal data optimization method and device based on electronic purse net and electronic equipment |
| CN112308010A (en) * | 2020-11-12 | 2021-02-02 | 珠海大横琴科技发展有限公司 | Ship shielding detection method and device based on YOLO-V3 algorithm |
| CN112270326A (en) * | 2020-11-18 | 2021-01-26 | 珠海大横琴科技发展有限公司 | Detection optimization method and device for ship sheltering and electronic equipment |
| CN112395993A (en) * | 2020-11-18 | 2021-02-23 | 珠海大横琴科技发展有限公司 | Method and device for detecting ship sheltered based on monitoring video data and electronic equipment |
| CN112270326B (en) * | 2020-11-18 | 2022-03-22 | 珠海大横琴科技发展有限公司 | Detection optimization method and device for ship sheltering and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111639513A (en) | Ship shielding identification method and device and electronic equipment | |
| US11475681B2 (en) | Image processing method, apparatus, electronic device and computer readable storage medium | |
| US11244157B2 (en) | Image detection method, apparatus, device and storage medium | |
| CN111754394B (en) | Method and device for detecting object in fisheye image and storage medium | |
| CN111582021A (en) | Method and device for detecting text in scene image and computer equipment | |
| WO2023082453A1 (en) | Image processing method and device | |
| US20210201501A1 (en) | Motion-based object detection method, object detection apparatus and electronic device | |
| CN115375999B (en) | Target detection model, method and device applied to hazardous chemical vehicle detection | |
| CN115294332B (en) | Image processing method, device, equipment and storage medium | |
| WO2023174098A1 (en) | Real-time gesture detection method and apparatus | |
| CN110991385A (en) | Method and device for identifying ship driving track and electronic equipment | |
| CN115147648A (en) | Tea shoot identification method based on improved YOLOv5 target detection | |
| US9679218B2 (en) | Method and apparatus for image matching | |
| CN111523558A (en) | Ship shielding detection method and device based on electronic purse net and electronic equipment | |
| CN114170654A (en) | Training method of age identification model, face age identification method and related device | |
| WO2019100348A1 (en) | Image retrieval method and device, and image library generation method and device | |
| CN115546909A (en) | Living body detection method and device, access control system, equipment and storage medium | |
| CN113158773B (en) | Training method and training device for living body detection model | |
| CN114494990A (en) | Target detection method, system, terminal equipment and storage medium | |
| CN111435445A (en) | Training method and device of character recognition model and character recognition method and device | |
| CN112329729B (en) | Small target ship detection method and device and electronic equipment | |
| US20220122341A1 (en) | Target detection method and apparatus, electronic device, and computer storage medium | |
| CN112800932B (en) | Method for detecting remarkable ship target in offshore background and electronic equipment | |
| CN112052863B (en) | Image detection method and device, computer storage medium and electronic equipment | |
| CN112037262A (en) | Target tracking method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200908 |
|
| RJ01 | Rejection of invention patent application after publication |