CN117574272A - Ocean data processing and classifying method - Google Patents
Ocean data processing and classifying method Download PDFInfo
- Publication number
- CN117574272A CN117574272A CN202311368870.XA CN202311368870A CN117574272A CN 117574272 A CN117574272 A CN 117574272A CN 202311368870 A CN202311368870 A CN 202311368870A CN 117574272 A CN117574272 A CN 117574272A
- Authority
- CN
- China
- Prior art keywords
- data
- underwater
- cluster
- sub
- ocean
- 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
- 238000012545 processing Methods 0.000 title claims abstract description 57
- 238000000034 method Methods 0.000 title claims abstract description 47
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 85
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 51
- 230000005540 biological transmission Effects 0.000 claims abstract description 40
- 238000007781 pre-processing Methods 0.000 claims abstract description 39
- 238000007637 random forest analysis Methods 0.000 claims abstract description 38
- 238000012800 visualization Methods 0.000 claims abstract description 17
- 238000004891 communication Methods 0.000 claims description 52
- 230000002159 abnormal effect Effects 0.000 claims description 34
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 23
- 229910052760 oxygen Inorganic materials 0.000 claims description 23
- 239000001301 oxygen Substances 0.000 claims description 23
- 238000003860 storage Methods 0.000 claims description 18
- 238000001914 filtration Methods 0.000 claims description 16
- 238000001514 detection method Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 15
- 238000004140 cleaning Methods 0.000 claims description 13
- 238000012876 topography Methods 0.000 claims description 13
- 238000006243 chemical reaction Methods 0.000 claims description 12
- 230000003628 erosive effect Effects 0.000 claims description 12
- 230000003750 conditioning effect Effects 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 11
- 230000004071 biological effect Effects 0.000 claims description 10
- 238000011156 evaluation Methods 0.000 claims description 10
- 229930002875 chlorophyll Natural products 0.000 claims description 9
- 235000019804 chlorophyll Nutrition 0.000 claims description 9
- ATNHDLDRLWWWCB-AENOIHSZSA-M chlorophyll a Chemical compound C1([C@@H](C(=O)OC)C(=O)C2=C3C)=C2N2C3=CC(C(CC)=C3C)=[N+]4C3=CC3=C(C=C)C(C)=C5N3[Mg-2]42[N+]2=C1[C@@H](CCC(=O)OC\C=C(/C)CCC[C@H](C)CCC[C@H](C)CCCC(C)C)[C@H](C)C2=C5 ATNHDLDRLWWWCB-AENOIHSZSA-M 0.000 claims description 9
- 150000003839 salts Chemical class 0.000 claims description 9
- 230000009467 reduction Effects 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 7
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 230000000737 periodic effect Effects 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 5
- 239000013598 vector Substances 0.000 claims description 5
- 230000008021 deposition Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 4
- 239000010802 sludge Substances 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 238000013500 data storage Methods 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 claims description 3
- 230000007935 neutral effect Effects 0.000 claims description 3
- 238000002360 preparation method Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000010183 spectrum analysis Methods 0.000 claims description 3
- 238000004148 unit process Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 8
- 238000012544 monitoring process Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000005260 corrosion Methods 0.000 description 3
- 230000007797 corrosion Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 238000011423 initialization method Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000012502 risk assessment Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C13/00—Surveying specially adapted to open water, e.g. sea, lake, river or canal
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C13/00—Surveying specially adapted to open water, e.g. sea, lake, river or canal
- G01C13/008—Surveying specially adapted to open water, e.g. sea, lake, river or canal measuring depth of open water
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/02—Instruments for indicating weather conditions by measuring two or more variables, e.g. humidity, pressure, temperature, cloud cover or wind speed
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/38—Services specially adapted for particular environments, situations or purposes for collecting sensor information
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Environmental & Geological Engineering (AREA)
- Hydrology & Water Resources (AREA)
- Software Systems (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Mathematical Physics (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention discloses a marine data processing and classifying method, which belongs to the technical field of marine science and data processing, and comprises the steps of firstly utilizing a self-adaptive underwater system and a water surface system to collect marine data, transmitting the data to a data preprocessing system for preprocessing, clustering the preprocessed data through a K-means algorithm model to obtain cluster information data of the marine data, transmitting the cluster information data to a random forest algorithm model for classification, classifying the cluster information data into different predefined categories by using the random forest algorithm model, and transmitting a classifying result to a visualization device. According to the invention, the marine data is comprehensively collected, the data quality and accuracy are improved through data preprocessing, the problems of complex marine environment, numerous marine data, high difficulty in data transmission, processing and classification and the like caused by the fact that the marine data are numerous are effectively solved, and the marine data preprocessing method has higher practical application and popularization values.
Description
Technical Field
The invention belongs to the technical field of ocean science and data processing, and particularly relates to an ocean data processing and classifying method.
Background
With the rapid development of technology, the acquisition and processing of ocean data become an important research field, and the ocean data have wide application, such as ocean environment monitoring, climate prediction, ocean biological research, submarine geographic exploration and the like. However, marine data is numerous and complex, including various types of data, such as physical, chemical, biological, and geographical, and the processing and classification difficulties are great, and moreover, the collection and transmission of data is also facing great challenges due to the specificity of the marine environment.
Therefore, how to collect ocean data, effectively process and classify the ocean data is of great significance to scientific research by utilizing the ocean data. The invention patent with the authorized bulletin number of CN114372538B discloses a method for classifying scale vortex time series convolution in a towed sensor array, which is characterized in that the towed sensor array is used for collecting sea profile observation data, amplifying the sea profile observation data and labeling sample types; selecting optimal shape local characteristics with distinguishability by establishing a time convolution network, and converting a multidimensional stereoscopic observation time sequence into a distance vector between the multidimensional stereoscopic observation time sequence and the shape; the online sequence overrun learning machine is constructed, the training samples of the obtained distance vectors and the class labels corresponding to the training samples are input into the online sequence overrun learning machine network for training, then the test samples are input into the trained online sequence overrun learning machine network model, finally the target classification result is obtained, and the three-dimensional monitoring of the marine environment is realized.
However, considering the complexity and diversity of the ocean data, although the above scheme has the benefit of fast classifying the ocean data collected by the sensor, in the technology, the ocean data source mainly depends on the towed sensor array, and the problem of single data collection mode exists, so that the processing and classification of various different types of data are difficult to realize.
Disclosure of Invention
Aiming at the problems of complex and various ocean data and difficult processing and classification in the prior art, the invention provides an ocean data processing and classifying method, which solves the problems of complex ocean environment and numerous ocean data, such as data acquisition, transmission, processing and classification.
The invention is realized by adopting the following technical scheme: a method of marine data processing and classification comprising the steps of:
step S1, marine data acquisition and transmission: the ocean data comprises underwater data and water data, the underwater data and the water data are respectively and correspondingly acquired through the self-adaptive underwater system and the water surface system, and the acquired ocean data are transmitted to the data preprocessing system and the visualization equipment through the water surface system;
step S2, preprocessing ocean data: carrying out data preprocessing on the ocean data based on a data preprocessing system, wherein the preprocessing step comprises data signal optimization, data filtering and cleaning and data conversion, and storing the preprocessed ocean data and transmitting the ocean data into a K-means algorithm model;
s3, clustering the K-means algorithm model: clustering the preprocessed ocean data by using a K-means algorithm model to obtain clustered information data of the clustered ocean data, and synchronously transmitting the clustered information to a random forest algorithm model and visualization equipment;
s4, classifying a random forest algorithm model: and classifying the cluster information data into different predefined categories by using the random forest algorithm model, and transmitting the classification result of the random forest algorithm model to the visualization equipment for visualization display.
Further, in the step S1, the adaptive underwater system includes an adaptive control device, an underwater acoustic communication device a end, an underwater seabed frame, an underwater power supply device, an underwater transmission device and an underwater acquisition device, wherein the underwater power supply device is located inside the underwater seabed frame, the underwater transmission device is connected with the underwater acoustic communication device a end through a waterproof cable, underwater data acquired by the underwater acquisition device is transmitted to the underwater acoustic communication device a end through the underwater transmission device, and an underwater data storage unit is arranged in the underwater transmission device;
the underwater acquisition device comprises a duplex acoustic releaser, a multi-parameter water quality instrument, a PH value sensor, a dissolved oxygen sensor, a Doppler flow velocity profile instrument and a sound sounding instrument, wherein the underwater data acquired by the underwater acquisition device comprises: underwater temperature data, chlorophyll data, depth data, salinity data, and turbidity data monitored by a multiparameter water quality meter; PH data monitored by a PH sensor; dissolved oxygen data monitored by a dissolved oxygen sensor; flow velocity data measured by an acoustic Doppler flow profiler; seismic data being explored by duplex acoustic releases; erosion and fouling data detected by sonar sounding equipment, seafloor topography data, underwater object data, and marine organism data.
Furthermore, an adjusting module is arranged in the self-adaptive control device, and the acquisition frequency and the priority of each unit of the underwater acquisition device are dynamically adjusted through the adjusting module according to the underwater data acquired in real time, so that when a sudden unstable condition occurs to a certain parameter of the underwater data, the sudden unstable condition can be timely identified and acquired.
Further, in the step S1, the water surface system includes a water sound communication machine B end, a water surface communication buoy, a water surface communication storage module, a warning system, a water surface power supply device, a mooring device and a water surface acquisition device, wherein a 4G transmission module, a Beidou transmission module and a weather observation sensor are fixedly arranged in the water surface communication buoy; the outside of the water surface communication buoy is fixedly provided with a tide level meter and a wave sensor, the A end of the underwater acoustic communication machine is in communication connection with the B end of the underwater acoustic communication machine, underwater data are transmitted to the B end of the underwater acoustic communication machine through the A end of the underwater acoustic communication machine, the underwater data are stored in a water surface communication storage module through the B end of the underwater acoustic communication machine, and the water surface communication storage module is connected with a wired interface of a water surface acquisition device to store the acquired water surface data and the underwater data; the water surface communication storage module transmits the underwater data and the water surface data to the data preprocessing system through the 4G transmission module and the Beidou transmission module; ,
the water surface acquisition device comprises a tide level instrument, a wave sensor and an meteorological observation sensor, and the acquired water surface data comprise: wind speed data, water temperature data, air humidity data and barometric pressure data monitored by the meteorological observation sensor; tide level data monitored by a tide level gauge; wave data monitored by the wave sensor.
Further, in step S2, the data preprocessing system includes a preprocessing data receiving module, a signal conditioning processing module, a filtering and cleaning module, a data conversion module, a preprocessing storage module, a data processing module and a data processing module, wherein the preprocessing storage module is used for receiving ocean data transmitted by the 4G transmission module and the beidou transmission module, and is connected with the K-means algorithm model through an application program interface;
the signal conditioning processing module is used for signal optimization and comprises a signal intensity conditioning unit and a filter unit, wherein the signal intensity conditioning unit dynamically adjusts the signal intensity by adopting an AGC technology so as to keep the constant intensity of an output signal; the filter unit performs frequency screening on the signals in a high-pass filtering mode so as to reserve a signal part related to actual ocean data;
the filtering and cleaning module is used for filtering and cleaning data and comprises a noise reduction processing unit, a missing value processing unit and an abnormal value detection unit, wherein the noise reduction processing unit adopts a moving average method to smooth the data, and the periodic components in the data are identified through frequency spectrum analysis and are used for detecting and processing periodic noise; the missing value processing unit processes missing values in the data by adopting a mode method, and sets an alarm function, and when the missing rate of a certain parameter in the ocean data is too high, an alarm is automatically sent out; the abnormal value detection unit adopts a box diagram method to identify abnormal values and process the abnormal values, the abnormal value detection unit is also provided with a function of recording abnormal value information, the abnormal value information comprises the numerical value, the position, the time stamp and the processing mode of the abnormal values, the abnormal value detection unit is provided with an alarm function, and when the abnormal rate of a certain parameter in ocean data is too high, the abnormal value detection unit automatically gives an alarm;
the data conversion module is used for data conversion and comprises a data normalization unit and a characteristic engineering unit, wherein the data normalization unit adopts a minimum-maximum scaling method to eliminate the influence brought by different dimensions and value ranges; the feature engineering unit adopts a PCA method to extract and construct a feature vector useful for the K-means algorithm model.
Further, in the step S3, the clustering process of the preprocessed ocean data by using the K-means algorithm model is as follows:
(1) Initializing: initializing by adopting a K-means++ algorithm:
firstly, selecting a sample point in a data set as a first clustering center randomly, calculating the nearest square distance between the sample point in the data set and all current clustering centers, then selecting a new sample point as a new clustering center, wherein the probability of selection is in direct proportion to the nearest square distance between the sample point in the data set and all current clustering centers, and repeating the steps until K clustering centers are selected;
(2) Assign samples to the nearest cluster center: for each sample point in the data set, calculating the distance between the sample point and each cluster center by adopting the mahalanobis distance, and then distributing each sample to the nearest cluster center;
(3) Recalculating the cluster center: for each cluster, calculating the average value of all samples of the cluster, wherein the average value is a new cluster center;
(4) Convergence checking: stopping the algorithm if the clustering center is not changed or reaches the preset iteration times, and returning to the step (2) if the clustering center is not reached;
(5) Dynamically determining the number of clusters: evaluation of contour coefficients is adopted: for different K values, calculating the contour coefficient of the clustering result, and selecting the K value with the largest contour coefficient as the optimal cluster number;
(6) Parallel computing: dividing the dataset into a plurality of sub-datasets, running steps (1) through (5) in parallel on a plurality of processors or compute nodes, merging results from different processors or compute nodes;
(7) Outputting a result: and outputting cluster information, wherein the cluster information comprises sub-cluster information of different labels.
Further, the cluster information comprises a water quality basic data sub-cluster, a biological activity related data sub-cluster, a physical attribute data sub-cluster, a submarine topography data sub-cluster, an erosion and deposition data sub-cluster, a seismic data sub-cluster, an underwater biological data sub-cluster, a climate and weather data sub-cluster, a tide and fluctuation data sub-cluster and a dissolved oxygen data sub-cluster;
the water quality basic data sub-cluster comprises underwater temperature data, salinity data and turbidity data; biological activity related data sub-clusters include chlorophyll data and PH data; the physical attribute data sub-cluster includes depth data and flow rate data; the sub-cluster of sub-sea floor topographic data comprises topographic data; the erosion-sludge data sub-cluster includes erosion-sludge data; the seismic data sub-cluster includes seismic data; the sub-cluster of underwater biological data comprises underwater biological data; the climate and weather data sub-clusters comprise wind speed data, water temperature data, air humidity data and atmospheric pressure data; the tide and wave data sub-cluster comprises tide level data and wave data; the dissolved oxygen data sub-cluster includes dissolved oxygen data.
Further, in the step S4, the process of classifying the cluster information data into different predefined categories by the random forest algorithm model is as follows:
1) Data preparation: taking cluster information data clustered by the K-Means model as input data, wherein each data point comprises a label of a cluster and characteristic data related to each sub-cluster;
2) And (3) dividing data: dividing the data into a training set and a testing set, and evaluating the performance of the model by adopting cross verification;
3) And (3) model construction: constructing a random forest classifier, and setting the quantity, depth and characteristic sub-sampling proportion super-parameters of decision trees;
4) Model training: training a random forest model by using training data, wherein the random forest is trained on a plurality of trees at the same time;
5) Model evaluation: evaluating model performance using the test dataset, evaluating classification performance using accuracy, precision, recall, F1 score indicators;
6) Feature importance: analyzing the importance of the features provided by the random forest model to understand the effect of the features on classification;
7) And (3) tuning: performing super-parameter tuning according to the evaluation result to improve the performance of the random forest model;
8) And (3) applying a model: inputting new cluster information data, and predicting a predefined category of cluster sub-information by a random forest model according to the characteristics of the new input cluster information data;
further, the predefined categories include:
water quality base data sub-clusters:
turbidity data: clear |medium| turbidity;
underwater temperature data: low temperature |medium temperature| high temperature;
salinity data: low salt |medium salt| high salt;
biological activity-related data sub-clusters:
PH value data: acidity |neutral| basicity;
chlorophyll data: low/medium/high;
physical attribute data sub-clusters:
depth data: shallow sea |middle-deep|deep sea;
flow rate data: slow/medium speed/fast;
submarine topography data sub-clusters:
terrain type: flat cliff seaditch;
erosion and fouling data sub-clusters:
erosion fouling degree: mild/moderate/severe;
seismic data sub-clusters:
seismic intensity: slight |moderate| strong;
underwater biological data sub-clusters:
biological species: less |medium| more;
climate and weather data sub-clusters:
wind speed data: breeze |and |strong wind;
water temperature data: low temperature |medium temperature| high temperature;
air humidity data: low humidity |medium| high humidity;
atmospheric pressure data: low-voltage |normal|high-voltage;
tidal and wave data sub-clusters:
tide level data: low tide |average|high tide;
wave data: wavelet |medium wave|large wave;
dissolved oxygen data sub-clusters:
dissolved oxygen content: low| medium high.
Further, the visualization device adopts an outdoor splicing point array type full-color screen display device, and is provided with a multi-source input support function so as to simultaneously receive a plurality of input sources.
Compared with the prior art, the invention has the advantages and positive effects that:
1. and (3) data collection is comprehensive: the comprehensive acquisition of underwater data and water data is realized through the combination of the underwater system and the water surface system, rich resources are provided for researching and utilizing ocean data, and when a certain parameter in the ocean is changed drastically, the change reason is known in time by setting the preferential acquisition and increasing the acquisition frequency;
2. the data processing is accurate and reliable: the data preprocessing system carries out multiple processing on the received ocean data, reduces noise and restores data authenticity, thereby improving data quality and accuracy and providing a reliable basis for subsequent processing and analysis;
3. stable and accurate data transmission: the 4G transmission module and the Beidou transmission module are adopted, so that the difficulty of data transmission in the marine environment is overcome, and the stability and accuracy of transmission are improved;
4. accurate classification and label assignment: the K-means algorithm model and the random forest algorithm model are respectively applied to the preprocessed data, so that accurate clustering and further classification of ocean data are realized.
Drawings
Fig. 1 is a flow chart of a data processing and classifying method according to an embodiment of the invention.
Detailed Description
In order that the above objects, features and advantages of the invention will be more readily understood, a further description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced otherwise than as described herein, and therefore the present invention is not limited to the specific embodiments disclosed below.
The embodiment provides a marine data processing and classifying method, as shown in fig. 1, comprising the following steps:
step S1, data acquisition and transmission: the marine data comprises underwater data and water data, the underwater data is collected by the self-adaptive underwater system, the water data is collected by the water surface system, and the collected marine data is transmitted to the data preprocessing system and the visualization equipment through the water surface system;
step S2, data preprocessing: carrying out data preprocessing on the ocean data based on a data preprocessing system, wherein the preprocessing step comprises data signal optimization, data filtering and cleaning and data conversion, and storing the preprocessed ocean data and transmitting the ocean data into a K-means algorithm model;
s3, clustering the K-means algorithm model: clustering the preprocessed ocean data by using a K-means algorithm model to obtain clustered information data of the clustered ocean data, and synchronously transmitting the clustered information to a random forest algorithm model and visualization equipment;
s4, classifying a random forest algorithm model: classifying the cluster information data into different predefined categories by utilizing a random forest algorithm model, and transmitting the classification result of the random forest algorithm model to the visualization equipment;
step S5, visual display: and displaying ocean data, K-means clustering results and random mode forest model classification results by using a visualization device.
The present invention will be described in detail below for a clearer understanding of the present invention:
in step S1, multi-azimuth acquisition of ocean data is achieved through the adaptive underwater system and the surface system:
an adaptive underwater system: the underwater power supply device comprises a self-adaptive control device, an underwater acoustic communication machine A end, an underwater seabed frame, an underwater power supply device, an underwater conveying device and an underwater acquisition device; the underwater power supply device is positioned in the underwater seabed frame and used for providing power for the self-adaptive underwater system; the underwater transmission device is connected with the A end of the underwater acoustic communication machine through a waterproof cable, underwater data acquired by the underwater acquisition device are transmitted to the A end of the underwater acoustic communication machine through the underwater transmission device, and an underwater data storage unit is arranged in the underwater transmission device so as to temporarily store the underwater data when the communication condition is poor; in addition, an adjusting module is arranged in the self-adaptive control device, and the acquisition frequency and the priority of each unit of the underwater acquisition device are dynamically adjusted according to the underwater data acquired in real time through the adjusting module, so that when a sudden unstable condition occurs to a certain parameter of the underwater data, the underwater data can be timely identified and acquired.
The underwater acquisition device comprises a duplex acoustic releaser, a multi-parameter water quality instrument, a PH value sensor, a dissolved oxygen sensor, a Doppler flow velocity profile instrument, a sonar sounding instrument and the like, and the underwater data acquired by the underwater acquisition device comprises: underwater temperature data, chlorophyll data, depth data, salinity data, and turbidity data monitored by the multi-parameter water quality meter; PH data monitored by the PH sensor; dissolved oxygen data monitored by the dissolved oxygen sensor; flow velocity data measured by the acoustic Doppler flow profiler; seismic data being explored by the duplex acoustic releaser; erosion and deposition data detected by the sonar depth finder, seafloor topography data, underwater object data, and marine organism data.
Surface system: the system comprises an underwater sound communication machine B end, a water surface communication buoy, a water surface communication storage module, a warning system, a water surface power supply device, a mooring device and a water surface acquisition device, wherein a 4G transmission module, a Beidou transmission module and a meteorological observation sensor are fixedly arranged in the water surface communication buoy; the outside of the water surface communication buoy is fixedly provided with a tide level meter and a wave sensor, the A end of the underwater acoustic communication machine is in communication connection with the B end of the underwater acoustic communication machine, underwater data are transmitted to the B end of the underwater acoustic communication machine through the A end of the underwater acoustic communication machine, and the underwater data are stored in the water surface communication storage module through the B end of the underwater acoustic communication machine; the water surface communication storage module is connected with the water surface acquisition device through a wired interface and used for storing acquired water surface data and underwater data, and the water surface communication storage module is used for transmitting the underwater data and the water surface data to the data preprocessing system through the 4G transmission module and the Beidou transmission module.
The water surface acquisition device comprises a tide level instrument, a wave sensor and an meteorological observation sensor, and the acquired water surface data comprise: wind speed data, water temperature data, air humidity data and barometric pressure data monitored by the meteorological observation sensor; tide level data monitored by a tide level gauge; wave data monitored by the wave sensor.
In step S2, the data preprocessing system includes: the system comprises a preprocessing data receiving module, a signal adjusting and processing module, a filtering and cleaning module, a data conversion module and a preprocessing storage module, wherein the preprocessing data receiving module is used for receiving ocean data transmitted by a 4G transmission module and a Beidou transmission module, and the preprocessing storage module is used for storing the data and is connected with the K-means algorithm model through an application program interface;
the signal conditioning processing module is used for signal optimization and comprises a signal intensity conditioning unit and a filter unit, wherein the signal intensity conditioning unit dynamically adjusts the signal intensity by adopting an AGC technology so as to keep the constant intensity of an output signal; the filter unit performs frequency screening on the signal in a high-pass filtering manner to preserve the signal portion associated with the actual marine data.
The filtering and cleaning module is used for filtering and cleaning data and comprises a noise reduction processing unit, a missing value processing unit and an abnormal value detection unit, wherein the noise reduction processing unit adopts a moving average method to smooth the data, and identifies periodic components in the data through frequency spectrum analysis and is used for detecting and processing periodic noise; the missing value processing unit processes missing values in the data by adopting a mode method, and sets an alarm function, and when the missing rate of a certain parameter in the ocean data is too high, the missing value processing unit automatically sends out an alarm; the abnormal value detection unit adopts a box line diagram method to identify abnormal values and process the abnormal values, the abnormal value detection unit is further provided with an abnormal value information recording function, the abnormal value information comprises numerical values, positions, time stamps and processing modes of the abnormal values, the abnormal value detection unit is provided with an alarm function, and when the abnormal rate of a certain parameter in ocean data is too high, the abnormal value detection unit automatically gives an alarm.
The data conversion module is used for data conversion and comprises a data normalization unit and a characteristic engineering unit, wherein the data normalization unit adopts a minimum-maximum scaling method to eliminate the influence brought by different dimensions and value ranges; the feature engineering unit adopts a PCA method to extract and construct a feature vector useful for the K-means algorithm model.
Therefore, in the data preprocessing stage, a perfect processing process is provided, and particularly, in the data filtering and cleaning stage, a perfect processing method is provided for abnormal data and missing data, so that the problems that marine data are numerous, noise is large in influence and difficult to process are solved.
In step S3, the clustering process of the K-means algorithm model on the preprocessed ocean data is as follows:
(1) Initializing: initializing by adopting a K-means++ algorithm:
firstly, selecting a sample point in a data set as a first clustering center randomly, calculating the nearest square distance between the sample point in the data set and all current clustering centers, then selecting a new sample point as a new clustering center, wherein the probability of selection is in direct proportion to the nearest square distance between the sample point in the data set and all current clustering centers, and repeating the steps until K clustering centers are selected;
(2) Assign samples to the nearest cluster center: for each sample point in the data set, calculating the distance between the sample point and each cluster center by adopting the mahalanobis distance, and then distributing each sample to the nearest cluster center;
(3) Recalculating the cluster center: for each cluster, calculating the average value of all samples of the cluster, wherein the average value is a new cluster center;
(4) Convergence checking: stopping the algorithm if the clustering center is not changed or reaches the preset iteration times, and returning to the step (2) if the clustering center is not reached;
(5) Dynamically determining the number of clusters: evaluation of contour coefficients is adopted: for different K values, calculating the contour coefficient of the clustering result, and selecting the K value with the largest contour coefficient as the optimal cluster number;
(6) Parallel computing: dividing the dataset into a plurality of sub-datasets, running steps (1) through (5) in parallel on a plurality of processors or compute nodes, merging results from different processors or compute nodes;
(7) Outputting a result:
outputting cluster information, wherein the cluster information comprises sub-cluster information of different labels, and the cluster information comprises a water quality basic data sub-cluster, a biological activity related data sub-cluster, a physical attribute data sub-cluster, a submarine topography data sub-cluster, an erosion and deposition data sub-cluster, a seismic data sub-cluster, an underwater biological data sub-cluster, a climate and weather data sub-cluster, a tide and fluctuation data sub-cluster and a dissolved oxygen data sub-cluster:
the water quality basic data sub-cluster is used for judging basic data of the marine environment and comprises underwater temperature data, salinity data and turbidity data;
biological activity related data sub-clusters for assessing marine biological activity abundance and health status, including chlorophyll data and PH data;
a physical attribute data sub-cluster comprising depth data and flow rate data;
sub-clusters of seafloor topography data, including topography data, useful for mapping seafloor topography and understanding seafloor topography conditions;
monitoring and studying corrosion fouling data sub-clusters of seafloor corrosion and fouling phenomena, including corrosion fouling data;
a seismic data sub-cluster that facilitates seismic activity monitoring and seismic risk assessment, including seismic data;
sub-clusters of underwater biodata, including underwater biodata, that facilitate estimation of biological resources and study of biodiversity;
climate and weather data sub-clusters providing basis for meteorological observation and forecast, including wind speed data, water temperature data, air humidity data and atmospheric pressure data;
tidal and wave data sub-clusters reflecting dynamic changes of the ocean and having important roles in navigation and disaster prevention and reduction, including tidal level data and wave data;
the dissolved oxygen data sub-clusters directly affecting the living environment and quality of the water body organisms comprise dissolved oxygen data.
The K-means algorithm model and the random forest algorithm model are respectively applied to the preprocessed data, so that the precise clustering and further classification of the ocean data are realized, and the problem that the ocean data are difficult to classify is solved.
In the K-means algorithm model clustering step, different from the traditional algorithm, the number of clusters is dynamically determined by adopting a contour coefficient evaluation method to solve the problem of excessive clustering or insufficient clustering, the similarity between data is measured by utilizing a Markov distance formula, the problem of instability and slow convergence speed of results caused by improper selection of the center of an initial cluster is solved by adopting a K-means++ algorithm initialization method, and the calculation efficiency of the K-means algorithm model is effectively improved by adopting a parallel calculation mode.
In step S4, the cluster information data is classified into different predefined categories using a random forest algorithm model, the predefined categories including:
water quality base data sub-clusters:
turbidity data: clear |medium| turbidity;
underwater temperature data: low temperature |medium temperature| high temperature;
salinity data: low salt |medium salt| high salt;
biological activity-related data sub-clusters:
PH value data: acidity |neutral| basicity;
chlorophyll data: low/medium/high;
physical attribute data sub-clusters:
depth data: shallow sea |middle-deep|deep sea;
flow rate data: slow/medium speed/fast;
submarine topography data sub-clusters:
terrain type: flat cliff seaditch;
erosion and fouling data sub-clusters:
erosion fouling degree: mild/moderate/severe;
seismic data sub-clusters:
seismic intensity: slight |moderate| strong;
underwater biological data sub-clusters:
biological species: less |medium| more;
climate and weather data sub-clusters:
wind speed data: breeze |and |strong wind;
water temperature data: low temperature |medium temperature| high temperature;
air humidity data: low humidity |medium| high humidity;
atmospheric pressure data: low-voltage |normal|high-voltage;
tidal and wave data sub-clusters:
tide level data: low tide |average|high tide;
wave data: wavelet |medium wave|large wave;
dissolved oxygen data sub-clusters:
dissolved oxygen content: low| medium high.
The method comprises the following steps:
1) Data preparation: taking cluster information data clustered by the K-Means model as input data, wherein each data point comprises a label of a cluster and characteristic data related to each sub-cluster;
2) And (3) dividing data: dividing the data into a training set and a testing set, and evaluating the performance of the model by adopting cross verification;
3) And (3) model construction: constructing a random forest classifier, and setting the quantity, depth and characteristic sub-sampling proportion super-parameters of decision trees;
4) Model training: training a random forest model by using training data, wherein the random forest is trained on a plurality of trees at the same time;
5) Model evaluation: evaluating model performance using the test dataset, evaluating classification performance using accuracy, precision, recall, F1 score indicators;
6) Feature importance: analyzing the importance of the features provided by the random forest model to understand the effect of the features on classification;
7) And (3) tuning: performing super-parameter tuning according to the evaluation result to improve the performance of the random forest model;
8) And (3) applying a model: inputting new cluster information data, and predicting a predefined category of cluster sub-information by a random forest model according to the characteristics of the new input cluster information data;
the ocean data is further classified by utilizing a random forest algorithm model, and the current class status of the ocean data is intuitively reflected through a visualization device, wherein the visualization device can adopt an outdoor splicing point array type full-color screen display device, and a multi-source input support function can be set so as to simultaneously receive a plurality of input sources.
The present invention is not limited to the above-mentioned embodiments, and any equivalent embodiments which can be changed or modified by the technical content disclosed above can be applied to other fields, but any simple modification, equivalent changes and modification made to the above-mentioned embodiments according to the technical substance of the present invention without departing from the technical content of the present invention still belong to the protection scope of the technical solution of the present invention.
Claims (10)
1. A method of marine data processing and classification comprising the steps of:
step S1, marine data acquisition and transmission: the ocean data comprises underwater data and water data, the underwater data and the water data are respectively and correspondingly acquired through the self-adaptive underwater system and the water surface system, and the acquired ocean data are transmitted to the data preprocessing system and the visualization equipment through the water surface system;
step S2, preprocessing ocean data: carrying out data preprocessing on the ocean data based on a data preprocessing system, wherein the preprocessing step comprises data signal optimization, data filtering and cleaning and data conversion, and storing the preprocessed ocean data and transmitting the ocean data into a K-means algorithm model;
s3, clustering the K-means algorithm model: clustering the preprocessed ocean data by using a K-means algorithm model to obtain clustered information data of the clustered ocean data, and synchronously transmitting the clustered information to a random forest algorithm model and visualization equipment;
s4, classifying a random forest algorithm model: and classifying the cluster information data into different predefined categories by using the random forest algorithm model, and transmitting the classification result of the random forest algorithm model to the visualization equipment for visualization display.
2. The marine data processing and classifying method according to claim 1, wherein in the step S1, the adaptive underwater system comprises an adaptive control device, an underwater acoustic communication machine a end, an underwater seabed frame, an underwater power supply device, an underwater transmission device and an underwater acquisition device, wherein the underwater power supply device is positioned inside the underwater seabed frame, the underwater transmission device is connected with the underwater acoustic communication machine a end through a waterproof cable, the underwater data acquired by the underwater acquisition device is transmitted to the underwater acoustic communication machine a end through the underwater transmission device, and an underwater data storage unit is arranged in the underwater transmission device;
the underwater acquisition device comprises a duplex acoustic releaser, a multi-parameter water quality instrument, a PH value sensor, a dissolved oxygen sensor, a Doppler flow velocity profile instrument and a sound sounding instrument, wherein the underwater data acquired by the underwater acquisition device comprises: underwater temperature data, chlorophyll data, depth data, salinity data, turbidity data, PH data, dissolved oxygen data, flow rate data, seismic data, erosion and fouling data, seafloor topography data, underwater object data, and marine organism data.
3. The ocean data processing and classifying method according to claim 2, wherein an adjusting module is arranged in the self-adaptive control device, and the collecting frequency and the priority of each unit of the underwater collecting device are dynamically adjusted according to the underwater data collected in real time through the adjusting module.
4. The ocean data processing and classifying method according to claim 1, wherein in the step S1, the water surface system comprises a water sound communication machine B end, a water surface communication buoy, a water surface communication storage module, a water surface power supply device, a mooring device and a water surface acquisition device, wherein a 4G transmission module, a Beidou transmission module and a weather observation sensor are fixedly arranged in the water surface communication buoy; the outside of the water surface communication buoy is fixedly provided with a tide level meter and a wave sensor, the A end of the underwater acoustic communication machine is in communication connection with the B end of the underwater acoustic communication machine, underwater data are transmitted to the B end of the underwater acoustic communication machine through the A end of the underwater acoustic communication machine, the underwater data are stored in a water surface communication storage module through the B end of the underwater acoustic communication machine, and the water surface communication storage module is connected with a wired interface of a water surface acquisition device to store the acquired water surface data and the underwater data; the water surface communication storage module transmits the underwater data and the water surface data to the data preprocessing system through the 4G transmission module and the Beidou transmission module;
the water surface acquisition device comprises a tide level meter, a wave sensor and an meteorological observation sensor, and the acquired water data comprise wind speed data, water temperature data, air humidity data, atmospheric pressure data, tide level data and wave data.
5. The ocean data processing and classifying method according to claim 1, wherein in the step S2, the data preprocessing system comprises a preprocessing data receiving module, a signal conditioning processing module, a filtering and cleaning module, a data conversion module and a preprocessing storage module, wherein the preprocessing data receiving module is used for receiving ocean data transmitted by the 4G transmission module and the beidou transmission module, and the preprocessing storage module is used for storing the data and is connected with the K-means algorithm model through an application program interface;
the signal conditioning processing module is used for signal optimization and comprises a signal intensity conditioning unit and a filter unit, wherein the signal intensity conditioning unit dynamically adjusts the signal intensity by adopting an AGC technology; the filter unit performs frequency screening on the signals in a high-pass filtering mode so as to reserve a signal part related to actual ocean data;
the filtering and cleaning module is used for filtering and cleaning data and comprises a noise reduction processing unit, a missing value processing unit and an abnormal value detection unit, wherein the noise reduction processing unit adopts a moving average method to smooth the data, and the periodic components in the data are identified through frequency spectrum analysis and are used for detecting and processing periodic noise; the missing value processing unit processes missing values in the data by adopting a mode method, and sets an alarm function, and when the missing rate of a certain parameter in the ocean data is too high, an alarm is automatically sent out; the abnormal value detection unit adopts a box diagram method to identify abnormal values and process the abnormal values, the abnormal value detection unit is also provided with a function of recording abnormal value information, the abnormal value information comprises the numerical value, the position, the time stamp and the processing mode of the abnormal values, the abnormal value detection unit is provided with an alarm function, and when the abnormal rate of a certain parameter in ocean data is too high, the abnormal value detection unit automatically gives an alarm;
the data conversion module is used for data conversion and comprises a data normalization unit and a characteristic engineering unit, wherein the data normalization unit adopts a minimum-maximum scaling method to eliminate the influence brought by different dimensions and value ranges; the feature engineering unit adopts a PCA method to extract and construct a feature vector useful for the K-means algorithm model.
6. The method for processing and classifying ocean data according to claim 1, wherein in the step S3, the pre-processing ocean data clustering process using the K-means algorithm model is as follows:
(1) Initializing: initializing by adopting a K-means++ algorithm:
firstly, selecting a sample point in a data set as a first clustering center randomly, calculating the nearest square distance between the sample point in the data set and all current clustering centers, then selecting a new sample point as a new clustering center, wherein the probability of selection is in direct proportion to the nearest square distance between the sample point in the data set and all current clustering centers, and repeating the steps until K clustering centers are selected;
(2) Assign samples to the nearest cluster center: for each sample point in the data set, calculating the distance between the sample point and each cluster center by adopting the mahalanobis distance, and then distributing each sample to the nearest cluster center;
(3) Recalculating the cluster center: for each cluster, calculating the average value of all samples of the cluster, wherein the average value is a new cluster center;
(4) Convergence checking: stopping the algorithm when the clustering center is not changed or reaches the preset iteration times, and returning to the step (2) if the clustering center is not reached;
(5) Dynamically determining the number of clusters: evaluation of contour coefficients is adopted: for different K values, calculating the contour coefficient of the clustering result, and selecting the K value with the largest contour coefficient as the optimal cluster number;
(6) Parallel computing: dividing the dataset into a plurality of sub-datasets, running steps (1) through (5) in parallel on a plurality of processors or compute nodes, merging results from different processors or compute nodes;
(7) Outputting a result: and outputting cluster information, wherein the cluster information comprises sub-cluster information of different labels.
7. The marine data processing and classification method of claim 6, wherein the cluster information comprises a water quality base data sub-cluster, a biological activity related data sub-cluster, a physical attribute data sub-cluster, a seafloor topography data sub-cluster, an erosion and deposition data sub-cluster, a seismic data sub-cluster, an underwater biological data sub-cluster, a climate and weather data sub-cluster, a tide and wave data sub-cluster, a dissolved oxygen data sub-cluster;
the water quality basic data sub-cluster comprises underwater temperature data, salinity data and turbidity data; biological activity related data sub-clusters include chlorophyll data and PH data; the physical attribute data sub-cluster includes depth data and flow rate data; the sub-cluster of sub-sea floor topographic data comprises topographic data; the erosion-sludge data sub-cluster includes erosion-sludge data; the seismic data sub-cluster includes seismic data; the sub-cluster of underwater biological data comprises underwater biological data; the climate and weather data sub-clusters comprise wind speed data, water temperature data, air humidity data and atmospheric pressure data; the tide and wave data sub-cluster comprises tide level data and wave data; the dissolved oxygen data sub-cluster includes dissolved oxygen data.
8. The method for processing and classifying marine data according to claim 1, wherein in the step S4, the process of classifying the cluster information data into different predefined categories by the random forest algorithm model is as follows:
1) Data preparation: taking cluster information data clustered by the K-Means model as input data, wherein each data point comprises a label of a cluster and characteristic data related to each sub-cluster;
2) Dividing data: dividing the data into a training set and a testing set, and evaluating the performance of the model by adopting cross verification;
3) Model construction: constructing a random forest classifier, and setting the quantity, depth and characteristic sub-sampling proportion super-parameters of decision trees;
4) Model training: training a random forest model by using training data, wherein the random forest is trained on a plurality of trees at the same time;
5) Model evaluation: evaluating model performance using the test dataset, evaluating classification performance using accuracy, precision, recall, F1 score indicators;
6) Feature importance: analyzing the importance of the features provided by the random forest model to understand the effect of the features on classification;
7) And (3) adjusting: performing super-parameter tuning according to the evaluation result to improve the performance of the random forest model;
8) And (3) applying a model: new cluster information data is input, and the random forest model predicts the predefined category of the cluster sub information according to the characteristics of the new input cluster information data.
9. A method of marine data processing and classification as claimed in claim 8, wherein said predefined categories include: water quality base data sub-clusters:
turbidity data: clear |medium| turbidity;
underwater temperature data: low temperature |medium temperature| high temperature;
salinity data: low salt |medium salt| high salt;
biological activity-related data sub-clusters:
PH value data: acidity |neutral| basicity;
chlorophyll data: low/medium/high;
physical attribute data sub-clusters:
depth data: shallow sea |middle-deep|deep sea;
flow rate data: slow/medium speed/fast;
submarine topography data sub-clusters:
terrain type: flat cliff seaditch;
erosion and fouling data sub-clusters:
erosion fouling degree: mild/moderate/severe;
seismic data sub-clusters:
seismic intensity: slight |moderate| strong;
underwater biological data sub-clusters:
biological species: less |medium| more;
climate and weather data sub-clusters:
wind speed data: breeze |and |strong wind;
water temperature data: low temperature |medium temperature| high temperature;
air humidity data: low humidity |medium| high humidity;
atmospheric pressure data: low-voltage |normal|high-voltage;
tidal and wave data sub-clusters:
tide level data: low tide |average|high tide;
wave data: wavelet |medium wave|large wave;
dissolved oxygen data sub-clusters:
dissolved oxygen content: low| medium high.
10. A marine data processing and sorting method according to claim 1, characterized in that the visualization device is an outdoor splice point array full color screen display device, and is provided with a multi-source input support function for receiving a plurality of input sources simultaneously.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311368870.XA CN117574272A (en) | 2023-10-20 | 2023-10-20 | Ocean data processing and classifying method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311368870.XA CN117574272A (en) | 2023-10-20 | 2023-10-20 | Ocean data processing and classifying method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117574272A true CN117574272A (en) | 2024-02-20 |
Family
ID=89888894
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311368870.XA Pending CN117574272A (en) | 2023-10-20 | 2023-10-20 | Ocean data processing and classifying method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117574272A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140082385A (en) * | 2012-12-24 | 2014-07-02 | 계명대학교 산학협력단 | medical image retrieval method based on image clustering |
| CN108846259A (en) * | 2018-04-26 | 2018-11-20 | 河南师范大学 | A kind of gene sorting method and system based on cluster and random forests algorithm |
| CN111242206A (en) * | 2020-01-08 | 2020-06-05 | 吉林大学 | High-resolution ocean water temperature calculation method based on hierarchical clustering and random forests |
| WO2022016884A1 (en) * | 2020-07-22 | 2022-01-27 | 江苏科技大学 | Method for extracting sea surface wind speed on basis of k-means clustering algorithm |
| JP7116852B1 (en) * | 2021-07-08 | 2022-08-10 | 自然資源部第一海洋研究所 | Parallel coupled ocean acoustic prediction system and method of operation |
| CN114912900A (en) * | 2022-06-06 | 2022-08-16 | 南通沪通海洋科技有限公司 | Ocean engineering design data acquisition method and system |
| CN115294404A (en) * | 2022-08-30 | 2022-11-04 | 中国科学院海洋研究所 | Benthonic animal hyperspectral data classification method based on random forest algorithm |
| CN116822381A (en) * | 2023-08-30 | 2023-09-29 | 中国海洋大学 | Marine temperature salt structure inversion method based on artificial intelligence |
-
2023
- 2023-10-20 CN CN202311368870.XA patent/CN117574272A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140082385A (en) * | 2012-12-24 | 2014-07-02 | 계명대학교 산학협력단 | medical image retrieval method based on image clustering |
| CN108846259A (en) * | 2018-04-26 | 2018-11-20 | 河南师范大学 | A kind of gene sorting method and system based on cluster and random forests algorithm |
| CN111242206A (en) * | 2020-01-08 | 2020-06-05 | 吉林大学 | High-resolution ocean water temperature calculation method based on hierarchical clustering and random forests |
| WO2022016884A1 (en) * | 2020-07-22 | 2022-01-27 | 江苏科技大学 | Method for extracting sea surface wind speed on basis of k-means clustering algorithm |
| JP7116852B1 (en) * | 2021-07-08 | 2022-08-10 | 自然資源部第一海洋研究所 | Parallel coupled ocean acoustic prediction system and method of operation |
| CN114912900A (en) * | 2022-06-06 | 2022-08-16 | 南通沪通海洋科技有限公司 | Ocean engineering design data acquisition method and system |
| CN115294404A (en) * | 2022-08-30 | 2022-11-04 | 中国科学院海洋研究所 | Benthonic animal hyperspectral data classification method based on random forest algorithm |
| CN116822381A (en) * | 2023-08-30 | 2023-09-29 | 中国海洋大学 | Marine temperature salt structure inversion method based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| André et al. | Listening to the deep: live monitoring of ocean noise and cetacean acoustic signals | |
| Wilson et al. | Multiscale terrain analysis of multibeam bathymetry data for habitat mapping on the continental slope | |
| Baumgartner et al. | A generalized baleen whale call detection and classification system | |
| Korneliussen | Acoustic target classification | |
| Sueur et al. | Global estimation of animal diversity using automatic acoustic sensors | |
| Guihen et al. | An assessment of the use of ocean gliders to undertake acoustic measurements of zooplankton: the distribution and density of Antarctic krill (Euphausia superba) in the Weddell Sea. | |
| CN112331231B (en) | Broiler feed intake detection system based on audio technology | |
| Ward et al. | Beaked whale (Mesoplodon densirostris) passive acoustic detection in increasing ambient noise | |
| CN117198330B (en) | Sound source identification method and system and electronic equipment | |
| CN111624586B (en) | Underwater acoustic target ranging method based on feature extraction and neural network | |
| CN114818839B (en) | Deep learning-based optical fiber sensing underwater acoustic signal identification method and device | |
| Cazau et al. | Measuring the marine soundscape of the Indian Ocean with southern elephant seals used as acoustic gliders of opportunity | |
| Belghith et al. | Acoustic diversity classifier for automated marine big data analysis | |
| CN113516626A (en) | Side-scan sonar seabed sand wave detection method based on multi-scale convolution and pooling strategy | |
| CN114018317B (en) | Data acquisition device and method for marine environment | |
| CN101614819A (en) | Submerged plants in shallow water lake automatic identification technology and device | |
| CN114722926A (en) | Convolution clustering method for scale vortex time sequence diagram in towed sensor array | |
| CN117574272A (en) | Ocean data processing and classifying method | |
| CN114662603A (en) | Ship-borne multivariate marine environment information acquisition visualization algorithm | |
| CN115248075A (en) | Underwater acoustic detection efficiency evaluation system for multi-source acoustic big data fusion | |
| CN110806444B (en) | Seabed sediment recognition and classification method based on shallow stratum profiler and SVM | |
| McCarthy et al. | Reduced-order machine-learning model for transmission loss prediction in underwater acoustics | |
| Zuba et al. | UPAL: Underwater passive aquatic listener | |
| CN112906458B (en) | Group intelligent optimized underwater laser multi-target end-to-end automatic identification system | |
| Ziegenhorn | Some like it cool: A study of odontocete ecology in the Hawaiian islands using passive acoustic monitoring |
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 |