CN116486177A - Underwater target identification and classification method based on deep learning - Google Patents
Underwater target identification and classification method based on deep learning Download PDFInfo
- Publication number
- CN116486177A CN116486177A CN202310542407.6A CN202310542407A CN116486177A CN 116486177 A CN116486177 A CN 116486177A CN 202310542407 A CN202310542407 A CN 202310542407A CN 116486177 A CN116486177 A CN 116486177A
- Authority
- CN
- China
- Prior art keywords
- underwater target
- classification
- deep learning
- recognition
- module
- 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
- 238000013135 deep learning Methods 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000013145 classification model Methods 0.000 claims abstract description 28
- 238000013136 deep learning model Methods 0.000 claims abstract description 10
- 238000002372 labelling Methods 0.000 claims abstract description 4
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 241000251468 Actinopterygii Species 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 1
- 238000009364 mariculture Methods 0.000 description 1
Classifications
-
- 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
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of 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
- G06V20/05—Underwater scenes
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The invention provides an underwater target recognition and classification method based on deep learning. The underwater target recognition and classification method based on deep learning comprises the following steps: acquiring an underwater target image, and labeling the underwater target image; constructing a plurality of deep-learning underwater target classification models, training the underwater target classification models, and classifying the underwater target images by using the trained underwater target classification models to obtain a plurality of underwater target classification results; and carrying out fine classification recognition work of the underwater target based on the deep learning model which is trained by the underwater target characteristics, and accurately recognizing the type of the underwater target. According to the underwater target recognition and classification method based on deep learning, the picture recognition and classification is automatically completed by processing the picture, a user does not need to have any knowledge on classification standards, the operation is simple and convenient, and the accuracy rate of classifying and recognizing the underwater target domain data is higher.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to an underwater target recognition and classification method based on deep learning.
Background
In the field of image processing technology, object detection refers to detecting and identifying whether an appointed object exists in an image, and determining the position of the appointed object, and has wide application in many aspects such as traffic safety, civil security, public property safety and the like.
With the development of machine vision technology, machine vision-based object recognition technology (such as vehicle type recognition, garbage type recognition and the like) has also gained more and more attention, and underwater object identifiers are respectively used for detecting underwater objects from sonar sensing images or camera videos and classifying the underwater objects to determine what type the underwater objects belong to.
In the prior art, identification and classification of some underwater target data are generally carried out by means of personal knowledge experience, and certain requirements are provided for relevant knowledge reserves of participants. For example, the classification of submarine garbage is not high in participation enthusiasm and ocean garbage classification accuracy of people in various places at present due to the influence of various garbage types, different specific classification standards of garbage in various places, limited personal knowledge time and energy and other factors. For another example, deep sea fishing or mariculture generally involves classification of various fish species, and most people cannot accurately distinguish fish species due to limited knowledge storage or culture experience, so that wide-range popularization is difficult.
Therefore, it is necessary to provide a deep learning-based underwater target recognition and classification method to solve the above technical problems.
Disclosure of Invention
The invention provides an underwater target recognition and classification method based on deep learning, which solves the problems in the background technology.
In order to solve the technical problems, the underwater target recognition and classification method based on deep learning provided by the invention comprises the following steps:
acquiring an underwater target image, and labeling the underwater target image;
constructing a plurality of deep-learning underwater target classification models, training the underwater target classification models, and classifying the underwater target images by using the trained underwater target classification models to obtain a plurality of underwater target classification results;
and carrying out fine classification recognition work of the underwater target based on the deep learning model which is trained by the underwater target characteristics, and accurately recognizing the type of the underwater target.
Preferably, the method further comprises the step of carrying out data enhancement on the classified images, and carrying out scale transformation and normalization on the images to be identified and classified.
Preferably, the classified image is input into a deep-learning underwater target classified model for feature extraction, and an output result of the deep learning is obtained.
Preferably, the outputting the result includes: and performing image enhancement processing on the classified images, and cutting the classified images subjected to the image enhancement processing into sub-images with preset sizes.
Preferably, the sub-image is input into a plurality of deep learning underwater target classification models to perform feature extraction, and an output result of each deep learning is obtained.
Preferably, the performing the classification recognition of the underwater target based on the deep learning model completed by the training of the underwater target features includes: voting the plurality of classification results to obtain the number of votes corresponding to each classification result in the plurality of classification results; and determining the classification result with the largest number of votes and the half number of votes as the underwater target classification.
Preferably, the deep learning underwater target classification model comprises a power supply module, a network transmission module, a scanning module, a comparison module, an application program and a storage module.
Preferably, the power supply module supplies power to the application program, the network transmission module provides a network for the application program, a processor is arranged in the application program and is connected with the comparison module, and the comparison module is connected with the storage module.
Preferably, the deep learning underwater target classification model further comprises a user login module, a fault repair module and an identification analysis module, and the storage module is used for storing data.
Compared with the related art, the underwater target recognition and classification method based on deep learning has the following beneficial effects:
the invention provides an underwater target recognition and classification method based on deep learning, which comprises the steps of obtaining an underwater target image and marking the underwater target image; constructing a plurality of deep-learning underwater target classification models, training the underwater target classification models, and classifying the underwater target images by using the trained underwater target classification models to obtain a plurality of underwater target classification results; the underwater target type is accurately identified by carrying out fine classification identification work on the underwater target based on the deep learning model which is completed by the underwater target feature training, the picture identification and classification is automatically completed by processing the picture, a user does not need to have any knowledge on classification standards, the operation is simple and convenient, and the accuracy of classifying and identifying the underwater target domain data is higher.
Drawings
Fig. 1 is a schematic diagram of a preferred embodiment of an underwater target recognition classification method based on deep learning according to the present invention.
Detailed Description
The invention will be further described with reference to the drawings and embodiments.
Referring to fig. 1 in combination, fig. 1 is a schematic diagram of an embodiment of a deep learning-based underwater target recognition classification method according to the present invention. The underwater target recognition and classification method based on deep learning comprises the following steps:
acquiring an underwater target image, and labeling the underwater target image;
constructing a plurality of deep-learning underwater target classification models, training the underwater target classification models, and classifying the underwater target images by using the trained underwater target classification models to obtain a plurality of underwater target classification results;
and carrying out fine classification recognition work of the underwater target based on the deep learning model which is trained by the underwater target characteristics, and accurately recognizing the type of the underwater target.
The method also comprises the steps of carrying out data enhancement on the classified images, and carrying out scale transformation and normalization on the images to be identified and classified.
And inputting the classified images into a deep learning underwater target classified model for feature extraction to obtain a deep learning output result.
The output result includes: and performing image enhancement processing on the classified images, and cutting the classified images subjected to the image enhancement processing into sub-images with preset sizes.
And inputting the sub-images into a plurality of deep learning underwater target classification models to perform feature extraction, and obtaining an output result of each deep learning.
The deep learning model based on the training of the underwater target features carries out the fine classification recognition work of the underwater target, and the deep learning model comprises the following steps: voting the plurality of classification results to obtain the number of votes corresponding to each classification result in the plurality of classification results; and determining the classification result with the largest number of votes and the half number of votes as the underwater target classification.
The deep learning underwater target classification model comprises a power supply module, a network transmission module, a scanning module, a comparison module, an application program and a storage module.
The power supply module supplies power to the application program, the network transmission module provides a network for the application program, a processor is arranged in the application program and is connected with the comparison module, and the comparison module is connected with the storage module.
The deep learning underwater target classification model further comprises a user login module, a fault repair module and an identification analysis module, and the storage module is used for storing data.
Compared with the related art, the underwater target recognition and classification method based on deep learning has the following beneficial effects:
marking the underwater target image by acquiring the underwater target image; constructing a plurality of deep-learning underwater target classification models, training the underwater target classification models, and classifying the underwater target images by using the trained underwater target classification models to obtain a plurality of underwater target classification results; the underwater target type is accurately identified by carrying out fine classification identification work on the underwater target based on the deep learning model which is completed by the underwater target feature training, the picture identification and classification is automatically completed by processing the picture, a user does not need to have any knowledge on classification standards, the operation is simple and convenient, and the accuracy of classifying and identifying the underwater target domain data is higher.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent structures or equivalent processes or direct or indirect application in other related technical fields are included in the scope of the present invention.
Claims (9)
1. An underwater target recognition and classification method based on deep learning is characterized by comprising the following steps:
acquiring an underwater target image, and labeling the underwater target image;
constructing a plurality of deep-learning underwater target classification models, training the underwater target classification models, and classifying the underwater target images by using the trained underwater target classification models to obtain a plurality of underwater target classification results;
and carrying out fine classification recognition work of the underwater target based on the deep learning model which is trained by the underwater target characteristics, and accurately recognizing the type of the underwater target.
2. The deep learning-based underwater target recognition classification method according to claim 1, further comprising performing data enhancement on the classified images, and performing scale transformation and normalization processing on the images to be recognized and classified.
3. The deep learning-based underwater target recognition classification method according to claim 1, wherein the classification image is input into a deep learning underwater target classification model for feature extraction to obtain an output result of the deep learning.
4. A deep learning based underwater target recognition classification method in accordance with claim 3, wherein said outputting results comprises: and performing image enhancement processing on the classified images, and cutting the classified images subjected to the image enhancement processing into sub-images with preset sizes.
5. The deep learning-based underwater target recognition classification method according to claim 1, wherein the sub-images are input into a plurality of deep learning underwater target classification models to perform feature extraction, and an output result of each deep learning is obtained.
6. The deep learning-based underwater target recognition classification method according to claim 1, wherein the performing of the fine classification recognition of the underwater target based on the deep learning model completed by the training of the underwater target features comprises: voting the plurality of classification results to obtain the number of votes corresponding to each classification result in the plurality of classification results; and determining the classification result with the largest number of votes and the half number of votes as the underwater target classification.
7. The deep learning-based underwater target recognition classification method of claim 1, wherein the deep learning underwater target classification model comprises a power supply module, a network transmission module, a scanning module, a comparison module, an application program and a storage module.
8. The deep learning-based underwater target recognition and classification method according to claim 7, wherein the power supply module supplies power to the application program, the network transmission module provides a network for the application program, a processor is arranged in the application program, the processor is connected with a comparison module, and the comparison module is connected with a storage module.
9. The deep learning-based underwater target recognition and classification method according to claim 7, wherein the deep learning underwater target classification model further comprises a user login module, a fault repair module and a recognition analysis module, and the storage module is used for storing data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310542407.6A CN116486177A (en) | 2023-05-15 | 2023-05-15 | Underwater target identification and classification method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310542407.6A CN116486177A (en) | 2023-05-15 | 2023-05-15 | Underwater target identification and classification method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116486177A true CN116486177A (en) | 2023-07-25 |
Family
ID=87216295
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310542407.6A Pending CN116486177A (en) | 2023-05-15 | 2023-05-15 | Underwater target identification and classification method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116486177A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117809168A (en) * | 2024-01-08 | 2024-04-02 | 中国电子科技集团公司第十五研究所 | Method and device for detecting inherent attribute characteristics based on underwater target |
-
2023
- 2023-05-15 CN CN202310542407.6A patent/CN116486177A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117809168A (en) * | 2024-01-08 | 2024-04-02 | 中国电子科技集团公司第十五研究所 | Method and device for detecting inherent attribute characteristics based on underwater target |
| CN117809168B (en) * | 2024-01-08 | 2024-05-17 | 中国电子科技集团公司第十五研究所 | Method and device for detecting inherent attribute characteristics based on underwater target |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108171184B (en) | Method for re-identifying pedestrians based on Simese network | |
| CN110659582A (en) | Image conversion model training method, heterogeneous face recognition method, device and equipment | |
| CN109829467A (en) | Image labeling method, electronic device and non-transient computer-readable storage medium | |
| CN105574550A (en) | Vehicle identification method and device | |
| CN109472280B (en) | Method for updating species recognition model library, storage medium and electronic equipment | |
| CN114155244B (en) | Defect detection method, device, equipment and storage medium | |
| CN110390308B (en) | Video behavior identification method based on space-time confrontation generation network | |
| CN109389105B (en) | Multitask-based iris detection and visual angle classification method | |
| CN116486177A (en) | Underwater target identification and classification method based on deep learning | |
| CN114626476A (en) | Bird fine-grained image recognition method and device based on Transformer and component feature fusion | |
| CN111680577A (en) | Face detection method and device | |
| CN111783541A (en) | Text recognition method and device | |
| CN114494773A (en) | Part sorting and identifying system and method based on deep learning | |
| CN113033297B (en) | Method, device, equipment and storage medium for programming real object | |
| CN116704526B (en) | Staff scanning robot and method thereof | |
| CN110334818B (en) | Method and system for automatically identifying pipeline | |
| CN115797970B (en) | Dense pedestrian target detection method and system based on YOLOv5 model | |
| CN114693554B (en) | Big data image processing method and system | |
| CN110728316A (en) | Classroom behavior detection method, system, device and storage medium | |
| CN112580739B (en) | Method and device for determining training sample set | |
| CN113158878B (en) | Heterogeneous migration fault diagnosis method, system and model based on subspace | |
| CN109325557B (en) | Data intelligence acquisition method based on computer visual image identification | |
| CN113947780A (en) | Sika deer face recognition method based on improved convolutional neural network | |
| CN111259843B (en) | Multimedia navigator testing method based on visual stability feature classification registration | |
| CN117274197A (en) | PCB defect detection method based on YOLO v5 algorithm improvement |
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 |