CN111612028A - Ship feature optimization method and device based on deep learning and electronic equipment - Google Patents
Ship feature optimization method and device based on deep learning and electronic equipment Download PDFInfo
- Publication number
- CN111612028A CN111612028A CN201911282577.5A CN201911282577A CN111612028A CN 111612028 A CN111612028 A CN 111612028A CN 201911282577 A CN201911282577 A CN 201911282577A CN 111612028 A CN111612028 A CN 111612028A
- Authority
- CN
- China
- Prior art keywords
- feature
- ship
- deep learning
- features
- image data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (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 deep learning, in particular to a ship feature optimization method based on deep learning. The method comprises the following steps: acquiring a ship image dataset; carrying out feature extraction on the ship image data set through a convolution network to obtain a first feature; presetting ship characteristics according to the ship image data set to obtain second characteristics; wherein the second features include boat shape, texture, and color; combining the first and second features using a deep learning network to obtain optimized vessel features. The first feature and the second feature are combined, so that the first feature information extracted is complemented through the second feature information, the accuracy of feature extraction is guaranteed, optimization of ship features is achieved, and the accuracy of feature extraction under the shielding condition is improved.
Description
Technical Field
The invention relates to the technical field of deep learning, in particular to a ship feature optimization method based on deep learning.
Background
In prior art, people detect the sea defense through artifical and supervisory equipment and carry out the sea defense and patrol and examine, but in traditional sea defense patrols and examines, detect the ship of sea area within range through artifical and supervisory equipment, can take place to overlap and shelter from the condition because of the target ship when detecting the target ship, consequently can make people produce certain difficulty or the problem of unable discernment when discerning the target ship, also make the false retrieval phenomenon appear when carrying out target ship detection and discernment because of these problems.
Disclosure of Invention
In view of this, the embodiment of the invention provides a ship feature optimization method based on deep learning, so as to solve the problem that ship features can be accurately extracted under the condition that a target ship is overlapped and shielded.
According to a first aspect, an embodiment of the present invention provides a deep learning-based ship feature optimization method, including:
acquiring a ship image dataset;
carrying out feature extraction on the ship image data set through a convolution network to obtain a first feature;
presetting ship characteristics according to the ship image data set to obtain second characteristics; wherein the second features include boat shape, texture, and color;
combining the first and second features using a deep learning network to obtain optimized vessel features.
By combining the first characteristic and the second characteristic, the characteristic information which cannot be considered through the automatically extracted first characteristic is complemented through the second characteristic, the accuracy of characteristic extraction is guaranteed, and therefore optimization of ship characteristics is achieved, and the accuracy of characteristic extraction under the shielding condition is improved.
With reference to the first aspect, in a first embodiment of the first aspect, the acquired ship image data comprises overlapping, occluded ship pictures.
By overlapping. And ship features are extracted from the shielded ship pictures, so that the accuracy of the extracted features is ensured.
With reference to the first aspect, in a second implementation manner of the first aspect, feature extraction is performed on the ship image data through a convolutional network to obtain a first feature; the method comprises the following steps:
extracting an image of a preset target in the ship data set;
and performing feature extraction on the image of the preset target to obtain a first feature.
The method comprises the steps of extracting ship data by using a convolution network, so that the first characteristic information which can be obtained is the characteristic output through deep learning, and the function of automatically extracting the ship characteristic is realized.
With reference to the first aspect, in a third implementation manner of the first aspect, combining the first feature and the second feature through a deep learning network includes:
acquiring a first feature and a second feature, and extracting partial feature data from the first feature and the second feature to be training data;
inputting the training data into a feature training model for training to generate a feature model;
outputting a third feature by inputting the first feature and the second feature into a feature model; the third feature is an optimized watercraft feature.
And constructing a feature model by using the acquired first feature, the acquired second feature and the deep learning network, outputting a third feature, and rapidly and accurately extracting the sheltered ship image through the third feature.
With reference to the first aspect, in a fourth implementation manner of the first aspect, inputting the training data into a feature training model for training, and generating the feature model includes: conventional feature constrained convolutional neural networks are used for back propagation. .
The traditional feature constraint convolutional neural network is utilized to carry out backward propagation, so that the first feature and the second feature can be combined, the ship feature optimization is realized, and the ship feature can be accurately extracted under the shielding condition, so that the ship identification and detection efficiency under the shielding condition is improved.
According to a second aspect, an embodiment of the present invention provides a deep learning-based ship feature optimization apparatus, including:
the acquisition module is used for acquiring ship image data;
the first module is used for carrying out feature extraction on the ship image data through a convolution network to obtain a first feature;
the second module is used for presetting ship characteristics according to the ship image data to obtain second characteristics;
a third module for combining the first and second features using a deep learning network to obtain optimized vessel features.
The first module is used for acquiring data, the first module and the second module are used for extracting first features and second features, and the third module is used for combining the extracted feature information so as to optimize ship features and accurately identify a target ship under the shielding condition.
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 being communicatively connected to each other, the memory storing therein computer instructions, and the processor executing the computer instructions to perform the method for vessel feature optimization based on deep learning according to 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 storing computer instructions for causing a computer to execute the deep learning based ship feature optimization method described in the first aspect or any one of the implementation manners of the first aspect.
Drawings
The features and advantages of the present invention will be more clearly understood by reference to the accompanying drawings, which are illustrative and not to be construed as limiting the invention in any way, and in which:
FIG. 1 is a flow chart of a vessel feature optimization method based on deep learning according to an embodiment of the invention;
FIG. 2 is a feature extraction flow chart of a deep learning based ship feature optimization method according to an embodiment of the present invention;
FIG. 3 is a feature joint flow chart of a deep learning based ship feature optimization method according to an embodiment of the present invention;
FIG. 4 is a block diagram of a deep learning based ship feature optimization device according to an embodiment of the present invention;
FIG. 5 is a schematic structural diagram of an electronic device according to an embodiment of the invention;
reference numerals
10-an acquisition module; 20-a first module; 30-a second module; 40-a third module;
51-a processor; 52-a memory; 53-bus.
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 feature optimization method based on deep learning, which comprises the following steps of:
acquiring a ship image dataset; the acquired ship data set comprises: the method comprises the steps of collecting video data of the marine vessel within a preset range within a preset time, or using the existing open source marine vessel data, or self-making a video data set.
Carrying out feature extraction on the ship image data set through a convolution network to obtain a first feature; wherein the first characteristic is obtained by data convolution through a convolution network. The convolution network is a deep learning network, and the result obtained by the first characteristic can be changed by setting the convolution times.
Presetting ship characteristics according to the ship image data set to obtain second characteristics; wherein the second features include boat shape, texture, and color; the second characteristic is set by the user's demand.
Combining the first and second features using a deep learning network to obtain optimized vessel features. And combining the first characteristic and the second characteristic by using the convolutional network again, but before combining the first characteristic and the second characteristic, the dimension reduction processing needs to be carried out on the first characteristic and the second characteristic so that the first characteristic and the second characteristic can be subjected to convolutional combination, and the combined characteristic can better reflect target ship data.
The first feature and the second feature are combined, so that the first feature information extracted is complemented through the second feature information, the accuracy of feature extraction is guaranteed, optimization of ship features is achieved, and the accuracy of feature extraction under the shielding condition is improved.
As shown in fig. 2, performing feature extraction on the ship image data through a convolutional network to obtain a first feature, including:
extracting an image of a preset target in the ship data set; the pre-set target image needs to be marked before the image of the pre-set target in the ship data set is extracted, including by using a region selection tool to frame the pre-set target image. The preset target image may be a single image or a designated image among a plurality of images.
And performing feature extraction on the image of the preset target to obtain a first feature. By extracting the selected target image, unnecessary operations are reduced, and the efficiency of extracting the first feature is improved, wherein the first feature can be a feature histogram. In the feature extraction, the target image may be extracted from a plurality of images, or a single image may be extracted from a plurality of images.
And, as shown in fig. 3, the combining the first feature and the second feature through a deep learning network includes:
acquiring a first feature and a second feature, and extracting partial feature data from the first feature and the second feature to be training data; the first feature and the second feature form a feature set, partial feature data are extracted from the feature set for training, and before training the feature machine, the first feature and the second feature need to be standardized so that the first feature and the second feature can be combined.
Inputting the training data into a feature training model for training to generate a feature model; and carrying out convolution integration on the normalized first feature and the normalized second feature by utilizing a convolution neural network. The first feature and the second feature may also be assembled in an overlapping manner.
Outputting a third feature by inputting the first feature and the second feature into a feature model; the third feature is an optimized watercraft feature. The third characteristic is obtained by combination, the third characteristic is the optimized ship characteristic, the first characteristic is obtained by utilizing the convolution network, the characteristic data is automatically generated and is combined with the characteristic data customized by the user, and data are supplemented with each other by partnering superposition or convolution, so that the output third characteristic is closer to the real data, and the shielded target can be highlighted.
Optionally, inputting the training data into a feature training model for training, and generating the feature model, including: conventional feature constrained convolutional neural networks are used for back propagation. The existing convolutional network is utilized to carry out deep learning so as to ensure that the acquired data can be automatically detected or identified according to the requirements, and the characteristic data set by a user is added so as to realize accurate identification.
Optionally, the acquired ship image data comprises overlapping, occluded ship pictures.
The obtained first features and the second features are used for constructing a training network to generate a recognition network, the features obtained through deep learning and the features extracted in advance are combined to realize feature complementation, and the sheltered ship image can be extracted quickly and accurately in a comprehensive mode.
According to a second aspect, an embodiment of the present invention provides a deep learning based ship feature optimization apparatus, as shown in fig. 4, including:
an acquisition module 10 for acquiring ship image data; the image data can be an open source ship data set or a video data set for daily monitoring in a sea area in a detection range.
A first module 20, configured to perform feature extraction on the ship image data through a convolutional network to obtain a first feature;
the second module 30 presets ship characteristics according to the ship image data to obtain second characteristics;
a third module 40 for combining the first feature and the first feature using a deep learning network to obtain an optimized vessel feature.
The first module is used for acquiring data, the first module and the second module are used for extracting first features and second features, and the third module is used for combining the extracted feature information so as to optimize ship features and accurately identify a target ship under the shielding condition.
The acquisition module is used for preprocessing the acquired ship data and then sending the ship data to the first module, the first module is used for receiving the ship graph sent by the acquisition module and extracting the characteristics of the ship image by utilizing a deep learning network, the ship data is also sent to the second module, in the second module, the characteristics required to be extracted are changed according to different conditions and can be changed according to different conditions, so that different characteristics can be better adapted to extraction, the characteristics of the ship under the conditions of overlapping and shielding can be extracted more quickly, the characteristics can be extracted more quickly, finally, the acquired first characteristics and the acquired second characteristics are sent to the third module, and the third module is used for carrying out normalization dimension reduction processing on the first characteristics and the second characteristics so that the optimized ship characteristics can be obtained when the first characteristics and the second characteristics are fused subsequently.
When the ship characteristic after optimizing discerns the ship under the overlapping, sheltering from the condition, because the adjustability of second characteristic is in order to arrange in can adjust fast and can to overlapping, shelter from the condition under the ship discernment precision adjust to the reduction false retrieval phenomenon appears when carrying out target ship detection and discernment.
An embodiment of the present invention further provides an electronic device, as shown in fig. 5, the electronic device may include a processor 51 and a memory 52, where the processor 51 and the memory 52 may be connected by a bus 53 or in another manner, and fig. 5 takes the connection by the bus as an example.
The processor 51 may be a Central Processing Unit (CPU). The Processor 51 may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, or combinations thereof.
The memory 52, which is a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the deep learning based ship feature optimization apparatus in the embodiment of the present invention, which execute program instructions/modules corresponding to the method (e.g., the acquiring module 10, the first module 20, the second module 30, and the third module 40 shown in fig. 4). The processor 51 executes various functional applications and data processing of the processor by running non-transitory software programs, instructions and modules stored in the memory 52, namely, implementing the deep learning based ship feature optimization method in the above method embodiment.
The memory 52 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 51, and the like. Further, the memory 52 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 52 may optionally include memory located remotely from the processor 51, and these remote memories may be connected to the processor 51 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 52 and when executed by the processor 51 perform a deep learning based vessel feature optimization method as in the embodiment of fig. 1-3.
The specific details of the vehicle terminal may be understood by referring to the corresponding related descriptions and effects in the embodiments shown in fig. 1 to fig. 3, 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 (8)
1. A ship feature optimization method based on deep learning is characterized by comprising the following steps:
acquiring a ship image dataset;
carrying out feature extraction on the ship image data set through a convolutional neural network to obtain a first feature;
presetting ship characteristics according to the ship image data set to obtain second characteristics; wherein the second features include boat shape, texture, and color;
combining the first and second features using a deep learning network to obtain optimized vessel features.
2. The method of claim 1, wherein the acquiring vessel image data comprises overlapping, occluded vessel pictures.
3. The method of claim 1, wherein the feature extraction of the ship image data by a convolutional network results in a first feature; the method comprises the following steps:
extracting an image of a preset target in the ship data set;
and performing feature extraction on the image of the preset target to obtain a first feature.
4. The method of claim 3, wherein combining the first feature and the second feature using a deep learning network comprises:
acquiring a first feature and a second feature, and extracting partial feature data from the first feature and the second feature to be training data;
inputting the training data into a feature training model for training to generate a feature model;
outputting a third feature by inputting the first feature and the second feature into a feature model; the third feature is an optimized watercraft feature.
5. The method of claim 4, wherein the inputting the training data into a feature training model for training and generating a feature model comprises: conventional feature constrained convolutional neural networks are used for back propagation.
6. A ship feature optimization device based on deep learning is characterized by comprising:
the acquisition module is used for acquiring ship image data;
the first module is used for carrying out feature extraction on the ship image data through a convolution network to obtain a first feature;
the second module is used for presetting ship characteristics according to the ship image data to obtain second characteristics;
a third module for combining the first and second features using a deep learning network to obtain optimized vessel features.
7. An electronic device, comprising:
a memory and a processor, the memory and the processor are connected with each other in communication, the memory stores computer instructions, the processor executes the computer instructions to execute the deep learning based ship feature optimization method according to any one of claims 1-5.
8. A computer-readable storage medium storing computer instructions for causing a computer to perform the deep learning based vessel feature optimization method according to any one of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911282577.5A CN111612028A (en) | 2019-12-13 | 2019-12-13 | Ship feature optimization method and device based on deep learning and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911282577.5A CN111612028A (en) | 2019-12-13 | 2019-12-13 | Ship feature optimization method and device based on deep learning and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111612028A true CN111612028A (en) | 2020-09-01 |
Family
ID=72201128
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911282577.5A Pending CN111612028A (en) | 2019-12-13 | 2019-12-13 | Ship feature optimization method and device based on deep learning and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111612028A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107845390A (en) * | 2017-09-21 | 2018-03-27 | 太原理工大学 | A kind of Emotional speech recognition system based on PCNN sound spectrograph Fusion Features |
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 |
US20190108639A1 (en) * | 2017-10-09 | 2019-04-11 | The Board Of Trustees Of The Leland Stanford Junior University | Systems and Methods for Semantic Segmentation of 3D Point Clouds |
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 |
CN111639513A (en) * | 2019-12-10 | 2020-09-08 | 珠海大横琴科技发展有限公司 | Ship shielding identification method and device and electronic equipment |
-
2019
- 2019-12-13 CN CN201911282577.5A patent/CN111612028A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107845390A (en) * | 2017-09-21 | 2018-03-27 | 太原理工大学 | A kind of Emotional speech recognition system based on PCNN sound spectrograph Fusion Features |
US20190108639A1 (en) * | 2017-10-09 | 2019-04-11 | The Board Of Trustees Of The Leland Stanford Junior University | Systems and Methods for Semantic Segmentation of 3D Point Clouds |
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 |
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 |
CN111639513A (en) * | 2019-12-10 | 2020-09-08 | 珠海大横琴科技发展有限公司 | Ship shielding identification method and device and electronic equipment |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
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 |
---|---|---|
CN108710847B (en) | Scene recognition method and device and electronic equipment | |
EP3757890A1 (en) | Method and device for image processing, method and device for training object detection model | |
US20200175062A1 (en) | Image retrieval method and apparatus, and electronic device | |
CN112102204B (en) | Image enhancement method and device and electronic equipment | |
CN109492577B (en) | Gesture recognition method and device and electronic equipment | |
EP2806374A1 (en) | Method and system for automatic selection of one or more image processing algorithm | |
CN111814810A (en) | Image recognition method and device, electronic equipment and storage medium | |
US10846531B2 (en) | Image-based anomaly detection for aerial inspection using self-supervised learning | |
CN111639513A (en) | Ship shielding identification method and device and electronic equipment | |
CN108318773B (en) | Transmission conductor strand breakage detection method and system | |
CN113688947B (en) | Method and system for identifying faults of infrared image of power distribution equipment | |
CN110942456B (en) | Tamper image detection method, device, equipment and storage medium | |
CN110991385A (en) | Method and device for identifying ship driving track and electronic equipment | |
CN110969213A (en) | Ship detection method and device based on fast RCNN and electronic equipment | |
CN112580581A (en) | Target detection method and device and electronic equipment | |
CN112750104A (en) | Method and device for automatically matching optimal camera by monitoring ship through multiple cameras | |
CN113158773B (en) | Training method and training device for living body detection model | |
CN111612028A (en) | Ship feature optimization method and device based on deep learning and electronic equipment | |
CN112883762A (en) | Living body detection method, device, system and storage medium | |
CN112329729B (en) | Small target ship detection method and device and electronic equipment | |
CN112395993A (en) | Method and device for detecting ship sheltered based on monitoring video data and electronic equipment | |
CN112270326B (en) | Detection optimization method and device for ship sheltering and electronic equipment | |
CN114140427A (en) | Object detection method and device | |
CN111695398A (en) | Small target ship identification method and device and electronic equipment | |
CN112714284A (en) | Power equipment detection method and device and mobile terminal |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200901 |