CN117075092A - Underwater sonar side-scan image small target detection method based on forest algorithm - Google Patents
Underwater sonar side-scan image small target detection method based on forest algorithm Download PDFInfo
- Publication number
- CN117075092A CN117075092A CN202311138389.1A CN202311138389A CN117075092A CN 117075092 A CN117075092 A CN 117075092A CN 202311138389 A CN202311138389 A CN 202311138389A CN 117075092 A CN117075092 A CN 117075092A
- Authority
- CN
- China
- Prior art keywords
- sonar
- forest algorithm
- scan image
- small target
- target detection
- 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
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 39
- 238000001514 detection method Methods 0.000 title claims abstract description 22
- 238000003066 decision tree Methods 0.000 claims abstract description 20
- 238000007637 random forest analysis Methods 0.000 claims abstract description 20
- 238000000034 method Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 11
- 238000012360 testing method Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 4
- 244000140747 Iris setosa Species 0.000 claims description 3
- 235000000827 Iris setosa Nutrition 0.000 claims description 3
- 208000025174 PANDAS Diseases 0.000 claims description 3
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 claims description 3
- 235000016496 Panda oleosa Nutrition 0.000 claims description 3
- 238000013145 classification model Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 3
- 240000000220 Panda oleosa Species 0.000 claims 1
- 230000011218 segmentation Effects 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 6
- 238000012876 topography Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000012935 Averaging Methods 0.000 description 2
- 240000004718 Panda Species 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/52—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00
- G01S7/539—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S15/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
- G01S15/88—Sonar systems specially adapted for specific applications
- G01S15/89—Sonar systems specially adapted for specific applications for mapping or imaging
- G01S15/8902—Side-looking sonar
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/54—Extraction of image or video features relating to texture
-
- 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
- G06V10/765—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects using rules for classification or partitioning the feature space
-
- 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
- G06V10/771—Feature selection, e.g. selecting representative features from a multi-dimensional feature space
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/05—Underwater scenes
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Acoustics & Sound (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
The invention relates to the technical field of underwater multi-beam sonar, and discloses an underwater sonar side-scan image small target detection method based on a forest algorithm, which is characterized by comprising the following steps of: step one, using information with a sonar graphic target to establish a sonar side-scan graphic set with target information; and secondly, randomly selecting N samples from the sonar side-scan pattern set to form a new target detection data subset. The technical scheme of the invention adopts a forest algorithm, and the forest algorithm has the following advantages: (1) The growth of the tree in the random forest algorithm gives additional randomness to the model, creating diversity, each node is segmented into optimal features that minimize errors, the random forest selects random features to construct the optimal segmentation, only a random subset of the nodes is considered for segmentation, and the tree can even be made more random by using random thresholds on each feature, rather than searching for optimal thresholds as in a normal decision tree.
Description
Technical Field
The invention relates to the technical field of underwater multi-beam sonar, in particular to an underwater sonar side-scan image small target detection method based on a forest algorithm.
Background
The multi-beam sonar technology is an advanced sonar technology, and can simultaneously transmit a plurality of acoustic beams, so that the detection efficiency and the detection precision of sonar are improved. The technology is widely applied to the fields of sea exploration, submarine topography mapping, underwater target detection and the like.
The use of multi-beam sonar technology is very widespread, particularly in marine exploration and sub-sea topographic mapping. In marine exploration, multi-beam sonar technology can help scientists to more accurately detect seafloor topography and marine life. In terms of submarine topography mapping, multi-beam sonar technology can help scientists to map submarine topography more accurately.
However, manually analyzing the massive amount of underwater sonar image data generated every day is a tedious and time-consuming task, and the large number of small targets greatly increases the workload. Therefore, a small target detection system for a classified and identified sonar side-scan image has important practical value for reducing time-consuming and expensive manual input.
Disclosure of Invention
The invention mainly aims to provide a method for detecting a small target of an underwater sonar side-scan image based on a forest algorithm, which aims to detect the small target in a mode of combining target identical target information and filtering target information which does not belong to the return characteristic, effectively improves the small target detection performance of the underwater sonar side-scan image, and has practicability and reliability.
In order to achieve the purpose, the underwater sonar side-scan image small target detection method based on the forest algorithm is characterized by comprising the following steps of:
step one, using information with a sonar graphic target to establish a sonar side-scan graphic set with target information;
randomly selecting N samples from the sonar side-scan pattern set to form a new target detection data subset;
randomly selecting M features from all features to be selected of the sonar graph by each data subset, and selecting features conforming to an input image from the M features as input features of a decision tree;
splitting according to the selected characteristics to obtain a child node;
step five, repeating the step 2-3, respectively obtaining corresponding decision trees through each data subset, and forming a random forest by a plurality of decision trees together;
step six, counting the number of each characteristic by an algorithm, and selecting the class with the largest number as a result to return;
and step seven, combining the classification results, and detecting the small target of the underwater sonar side-scan image by combining the same target information of the targets and filtering the target information which does not belong to the return characteristic.
In an embodiment, in the step one, fuzzy standards are performed on all targets in the collected sonar graphic information by adopting rectangular frames, large targets are marked one by one through the rectangular frames, and small targets are marked in a region through the rectangular frames.
In one embodiment, each internal node in the decision tree represents a feature, each branch represents a value for this effect, and each leaf node represents a classification.
In one embodiment, each subset in the second step contains data with the same characteristics, and a classification model based on random forests is implemented by using scikit-learn library of Python.
In one embodiment, the sonar side-scan pattern set in the step one uses an Iris dataset model, and the three categories in the dataset are Iris Setosa, iris Versicolour, and Iris virginiana respectively, and the dataset has 150 samples, wherein each category has 50 samples.
In one embodiment, the second step also requires converting the dataset into a DataFrame format using the Pandas library, and setting the number of trees and the maximum depth of the trees as default parameters.
In one embodiment, the sonar graphic features are texture, shape, and edges.
In an embodiment, in the fifth step, the data set is divided into a training set and a testing set, the training set is used for training the model, and the testing set is used for testing the prediction accuracy of the model.
In an embodiment, the training set trains N weak learners independently, and the final strong learners are obtained for the N weak learners by a set strategy, the prediction classes are { c1, c2, … … ck }, and the prediction results of the N weak learners are (h) for any one prediction sample x 1 (x),h 2 (x),……h T (x))。
In one embodiment, each individual learner has a weight w, then the final prediction is:
wherein,H(x)represents the final prediction result corresponding to the prediction sample x, w i Is individual studyDevice h i Is usually:
;
wherein,Trepresenting the number of samples.
The technical scheme of the invention has the following advantages by adopting a forest algorithm:
(1) The growth of trees in random forest algorithms can bring additional randomness to the model, creating diversity. Each node is segmented into optimal features that minimize errors, and a random forest selects random features to construct the optimal segmentation. Therefore, the detection method only considers the random subset used for dividing the node, the tree can be made more random by using a random threshold value on each feature, the optimal threshold value is not searched as a normal decision tree, and features can be better refined when the underwater small target side-scanning image is processed, so that the discrimination is improved.
(2) The relative importance of each feature to the prediction is easily measured. It measures the importance of a feature by looking at how much of the non-purity of all trees in a forest is reduced by using the feature. The method automatically calculates the score of each feature after training, and normalizes the results so that the sum of the importance of all the features is equal to 1, so that the weight of each feature can be considered when small target identification is carried out, and the method achieves fine processing and is more practical for small target image processing.
(3) The processing capability of the high-dimensional data set is good, thousands of input variables can be processed, the most important variables are determined, the multi-feature characteristics of the small target image are further processed, and the recognition degree is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is an algorithm flow chart of the underwater sonar side-scan image small target detection method based on the forest algorithm.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that, if directional indications (such as up, down, left, right, front, and rear … …) are included in the embodiments of the present invention, the directional indications are merely used to explain the relative positional relationship, movement conditions, etc. between the components in a specific posture (as shown in the drawings), and if the specific posture is changed, the directional indications are correspondingly changed.
In the present invention, unless specifically stated and limited otherwise, the terms "connected," "affixed," and the like are to be construed broadly, and for example, "affixed" may be a fixed connection, a removable connection, or an integral body; can be mechanically or electrically connected; either directly or indirectly, through intermediaries, or both, may be in communication with each other or in interaction with each other, unless expressly defined otherwise. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.
In addition, if there is a description of "first", "second", etc. in the embodiments of the present invention, the description of "first", "second", etc. is for descriptive purposes only and is not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In addition, the meaning of "and/or" as it appears throughout includes three parallel schemes, for example "A and/or B", including the A scheme, or the B scheme, or the scheme where A and B are satisfied simultaneously. In addition, the technical solutions of the embodiments may be combined with each other, but it is necessary to base that the technical solutions can be realized by those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not within the scope of protection claimed in the present invention.
Referring to fig. 1, in an embodiment of the invention, a method for detecting a small target of an underwater sonar side-scan image based on a forest algorithm is characterized by comprising the following steps:
step one, using information with a sonar graphic target to establish a sonar side-scan graphic set with target information;
randomly selecting N samples from the sonar side-scan pattern set to form a new target detection data subset;
randomly selecting M features from all features to be selected of the sonar graph by each data subset, and selecting features conforming to an input image from the M features as input features of a decision tree;
splitting according to the selected characteristics to obtain a child node;
step five, repeating the step 2-3, respectively obtaining corresponding decision trees through each data subset, and forming a random forest by a plurality of decision trees together;
step six, counting the number of each characteristic by an algorithm, and selecting the class with the largest number as a result to return;
and step seven, combining the classification results, and detecting the small target of the underwater sonar side-scan image by combining the same target information of the targets and filtering the target information which does not belong to the return characteristic.
The technical scheme of the invention has high robustness to noise, abnormal values and other adverse factors through the random forest algorithm, the random forest algorithm trains the data by using a plurality of decision trees at the same time, and a more stable and reliable prediction result can be obtained through an averaging or voting mechanism. In addition, the random forest algorithm can automatically process missing values in the data set, and robustness of the algorithm is further enhanced. The random forest algorithm can effectively avoid the problem of overfitting by using a random subset and random features, each decision tree is trained on different random subsets, so that the difference between each decision tree is larger, the variance of a model is reduced, the random forest algorithm can process high-dimensional data, only a part of random features are selected for training, the random forest algorithm does not need to calculate all the features, the algorithm efficiency can be improved, and the random forest algorithm can evaluate the importance of the features by calculating the importance of each feature in all the decision trees. This importance index may help us select the most relevant features, thereby improving the efficiency and accuracy of the algorithm.
In general, each target in the collected sonar graphic information is subjected to fuzzy standard by adopting a rectangular frame, large targets are marked one by one through the rectangular frame, and small targets are marked in a region through the rectangular frame.
Accordingly, each internal node in the decision tree represents a feature, each branch represents a value for this effect, and each leaf node represents a classification.
In this embodiment, each subset contains data with the same characteristics, and a classification model based on random forests is implemented by using the scikit-learn library of Python.
Of course, the sonar side-scan pattern set adopts an Iris data set model, three categories in the data set are Iris Setosa, iris Versicolour and Iris Virginiana respectively, the data set has 150 samples, wherein each category has 50 samples
In this embodiment, it is also necessary to use the Pandas library to convert the dataset into the DataFrame format, and set the number of trees and the maximum depth of the trees as default parameters
Further, sonar graphics features are texture, shape, and edges.
However, the design is not limited thereto, and in other embodiments, the training set is independently trained on the basis of the aboveN weak learners, and obtaining final strong learners for the N weak learners through an aggregation strategy, wherein the prediction categories are { c1, c2, … … ck }, and the prediction results of the N weak learners are (h) for any one prediction sample x 1 (x),h 2 (x),……h T (x))。
In this embodiment, each individual learner has a weight w, and the final prediction is:
wherein,H(x)represents the final prediction result corresponding to the prediction sample x, w i Is an individual learner h i Is usually:
;
wherein,Trepresenting the number of samples.
It should be noted that, the random forest is an integrated learning algorithm, and is widely used in fields such as data mining and machine learning due to its excellent performance. The random forest trains the data set by using a plurality of decision trees at the same time, and obtains a final prediction result by a voting mechanism or an averaging mode, wherein the random mainly refers to two aspects: first, randomly selecting samples, namely, sampling with replacement from an original data set to obtain sub data sets, wherein the sample size of the sub data sets is kept consistent with that of the original data set, elements among different sub data sets can be repeated, and elements among the same sub data set can also be repeated. Second, randomly selecting features, similar to the random sample selection process, the sub-dataset selects a certain number of feature subsets from all the original features to be selected, and then selects from the re-selected feature subsets. And forming a decision tree through the data subset and the feature subset selected each time, and finally obtaining a random forest algorithm.
The foregoing description is only of the preferred embodiments of the present invention and is not intended to limit the scope of the invention, and all equivalent structural changes made by the description of the present invention and the accompanying drawings or direct/indirect application in other related technical fields are included in the scope of the invention.
Claims (10)
1. The underwater sonar side-scan image small target detection method based on the forest algorithm is characterized by comprising the following steps of:
step one, using information with a sonar graphic target to establish a sonar side-scan graphic set with target information;
randomly selecting N samples from the sonar side-scan pattern set to form a new target detection data subset;
randomly selecting M features from all features to be selected of the sonar graph by each data subset, and selecting features conforming to an input image from the M features as input features of a decision tree;
splitting according to the selected characteristics to obtain a child node;
step five, repeating the step 2-3, respectively obtaining corresponding decision trees through each data subset, and forming a random forest by a plurality of decision trees together;
step six, counting the number of each characteristic by an algorithm, and selecting the class with the largest number as a result to return;
and step seven, combining the classification results, and detecting the small target of the underwater sonar side-scan image by combining the same target information of the targets and filtering the target information which does not belong to the return characteristic.
2. The method for detecting the small targets in the underwater sonar side-scan image based on the forest algorithm according to claim 1, wherein in the first step, fuzzy standards are carried out on all targets in the collected sonar graphic information by adopting rectangular frames, large targets are marked one by one through the rectangular frames, and small targets are marked in a region through the rectangular frames.
3. A method for detecting a small target in an underwater sonar sideways scanned image based on a forest algorithm as defined in claim 1, wherein each internal node in said decision tree represents a feature, each branch represents a value of the special effect, and each leaf node represents a classification.
4. The underwater sonar side-scan image small target detection method based on the forest algorithm as set forth in claim 1, wherein each subset in the second step contains data with the same characteristics, and a classification model based on random forests is realized by adopting a scikit-learn library of Python.
5. The method for detecting the small target of the underwater sonar side-scan image based on the forest algorithm as set forth in claim 1, wherein the sonar side-scan image set in the step one adopts an Iris data set model, and three categories in the data set are Iris Setosa, iris verisimolour and Iris virginiana respectively, and the data set has 150 samples, wherein each category has 50 samples.
6. The underwater sonar side-scan image small target detection method based on the forest algorithm as claimed in claim 1, wherein in the second step, the Pandas library is further required to be used for converting the data set into a DataFrame format, and the number of trees and the maximum depth of the trees are set as default parameters.
7. The underwater sonar side-scan image small target detection method based on the forest algorithm as defined in claim 1, wherein the sonar figure features are texture, shape and edge.
8. The underwater sonar side-scan image small target detection method based on the forest algorithm as claimed in claim 1, wherein in the fifth step, the data set is divided into a training set and a test set, the training set is used for training a model, and the test set is used for testing the prediction accuracy of the model.
9. The underwater sonar side-scan image small target detection method based on forest algorithm as claimed in claim 8, wherein said training set trains N weak learners independently, and the final strong learners are obtained for the N weak learners by means of a set strategy, the prediction class is { c1, c2, … … ck }, and the prediction results of the N weak learners for any one prediction sample x are (h 1 (x),h 2 (x),……h T (x))。
10. A method for detecting a small target in an underwater sonar side-scan image based on a forest algorithm as defined in claim 9, wherein each individual learner has a weight w, then the final prediction is:
wherein,H(x)represents the final prediction result corresponding to the prediction sample x, w i Is an individual learner h i Is usually:
;
wherein,Trepresenting the number of samples.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311138389.1A CN117075092A (en) | 2023-09-05 | 2023-09-05 | Underwater sonar side-scan image small target detection method based on forest algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311138389.1A CN117075092A (en) | 2023-09-05 | 2023-09-05 | Underwater sonar side-scan image small target detection method based on forest algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117075092A true CN117075092A (en) | 2023-11-17 |
Family
ID=88708005
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311138389.1A Pending CN117075092A (en) | 2023-09-05 | 2023-09-05 | Underwater sonar side-scan image small target detection method based on forest algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117075092A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117538881A (en) * | 2024-01-10 | 2024-02-09 | 海底鹰深海科技股份有限公司 | Sonar water imaging beam forming method, system, equipment and medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107886050A (en) * | 2017-10-16 | 2018-04-06 | 电子科技大学 | Utilize time-frequency characteristics and the Underwater targets recognition of random forest |
CN109448038A (en) * | 2018-11-06 | 2019-03-08 | 哈尔滨工程大学 | Sediment sonar image feature extracting method based on DRLBP and random forest |
WO2021022970A1 (en) * | 2019-08-05 | 2021-02-11 | 青岛理工大学 | Multi-layer random forest-based part recognition method and system |
CN114155428A (en) * | 2021-11-26 | 2022-03-08 | 中国科学院沈阳自动化研究所 | Underwater sonar side-scan image small target detection method based on Yolo-v3 algorithm |
CN116340746A (en) * | 2023-03-28 | 2023-06-27 | 哈尔滨理工大学 | Feature selection method based on random forest improvement |
-
2023
- 2023-09-05 CN CN202311138389.1A patent/CN117075092A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107886050A (en) * | 2017-10-16 | 2018-04-06 | 电子科技大学 | Utilize time-frequency characteristics and the Underwater targets recognition of random forest |
CN109448038A (en) * | 2018-11-06 | 2019-03-08 | 哈尔滨工程大学 | Sediment sonar image feature extracting method based on DRLBP and random forest |
WO2021022970A1 (en) * | 2019-08-05 | 2021-02-11 | 青岛理工大学 | Multi-layer random forest-based part recognition method and system |
CN114155428A (en) * | 2021-11-26 | 2022-03-08 | 中国科学院沈阳自动化研究所 | Underwater sonar side-scan image small target detection method based on Yolo-v3 algorithm |
CN116340746A (en) * | 2023-03-28 | 2023-06-27 | 哈尔滨理工大学 | Feature selection method based on random forest improvement |
Non-Patent Citations (2)
Title |
---|
张蕾: "基于随机森林方法的卫星高光谱影像道路提取与分析", 航天器工程, vol. 32, no. 4 * |
王楠: "从决策树到集成学习", 智能系统与技术丛书 自然语言理解与行业知识图谱 概念方法与工程落地, pages 79 - 82 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117538881A (en) * | 2024-01-10 | 2024-02-09 | 海底鹰深海科技股份有限公司 | Sonar water imaging beam forming method, system, equipment and medium |
CN117538881B (en) * | 2024-01-10 | 2024-05-07 | 海底鹰深海科技股份有限公司 | Sonar water imaging beam forming method, system, equipment and medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110443281B (en) | Text classification self-adaptive oversampling method based on HDBSCAN (high-density binary-coded decimal) clustering | |
CN111444769B (en) | Laser radar human leg detection method based on multi-scale self-adaptive random forest | |
Wang et al. | Fast image segmentation using two-dimensional Otsu based on estimation of distribution algorithm | |
CN117075092A (en) | Underwater sonar side-scan image small target detection method based on forest algorithm | |
Zhong et al. | Detecting, classifying, and counting blue whale calls with Siamese neural networks | |
Kim et al. | Combined visually and geometrically informative link hypothesis for pose-graph visual SLAM using bag-of-words | |
CN113343123B (en) | Training method and detection method for generating confrontation multiple relation graph network | |
Von Rohrscheidt et al. | Topological singularity detection at multiple scales | |
CN110929801B (en) | Improved Euclid distance KNN classification method and system | |
CN115880189B (en) | Multi-beam point cloud filtering method for submarine topography | |
JP2015232805A (en) | Image processing method, image processor, and image processing program | |
CN109767545B (en) | Method and system for classifying defects of valuable bills | |
CN116343825A (en) | Sample mapping-based few-sample underwater sound target identification method | |
CN111931749B (en) | Face clustering method and device based on neural network | |
KR102427861B1 (en) | Apparatus and method for generating underwater image data | |
CN115082784A (en) | Multi-temporal remote sensing image unsupervised change detection pseudo sample automatic generation method | |
CN114662603A (en) | Ship-borne multivariate marine environment information acquisition visualization algorithm | |
CN110889436B (en) | Underwater multi-class target classification method based on credibility estimation | |
Fakiris et al. | Sonar Class: A MATLAB toolbox for the classification of side scan sonar imagery, using local textural and reverberational characteristics | |
CN111524098A (en) | Neural network output layer cutting and template frame size determining method based on self-organizing clustering | |
Wei et al. | Automatic cleaning of outliers in multibeam bathymetric data with clustering algorithm | |
Chavez et al. | Automated species counting using a hierarchical classification approach with haar cascades and multi-descriptor random forests | |
CN117495891B (en) | Point cloud edge detection method and device and electronic equipment | |
Roa et al. | Taxonomical classification of reef fish with broadband backscattering models and machine learning approaches | |
JPH0981684A (en) | Pattern recognition device and character segmentation system |
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 |