CN117274819B - Remote sensing information extraction system for intelligent monitoring of multiple types of sea areas - Google Patents
Remote sensing information extraction system for intelligent monitoring of multiple types of sea areas Download PDFInfo
- Publication number
- CN117274819B CN117274819B CN202311534117.3A CN202311534117A CN117274819B CN 117274819 B CN117274819 B CN 117274819B CN 202311534117 A CN202311534117 A CN 202311534117A CN 117274819 B CN117274819 B CN 117274819B
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- information
- suspended matter
- particle size
- sets
- 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.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 58
- 238000000605 extraction Methods 0.000 title claims abstract description 34
- 239000002245 particle Substances 0.000 claims abstract description 80
- 239000000725 suspension Substances 0.000 claims abstract description 76
- 238000002310 reflectometry Methods 0.000 claims abstract description 58
- 238000000034 method Methods 0.000 claims abstract description 27
- 230000004927 fusion Effects 0.000 claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims description 38
- 238000012360 testing method Methods 0.000 claims description 38
- 238000012549 training Methods 0.000 claims description 29
- 238000003921 particle size analysis Methods 0.000 claims description 27
- 238000001228 spectrum Methods 0.000 claims description 23
- 238000010276 construction Methods 0.000 claims description 9
- 238000010801 machine learning Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 239000007787 solid Substances 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims 4
- 230000000694 effects Effects 0.000 abstract description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000005303 weighing Methods 0.000 description 2
- 238000013075 data extraction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/06—Investigating concentration of particle suspensions
-
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- 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
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
-
- 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
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G06V10/806—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Chemical & Material Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Analytical Chemistry (AREA)
- Pathology (AREA)
- Immunology (AREA)
- Biochemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Dispersion Chemistry (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a remote sensing information extraction system for intelligent monitoring of multiple types of sea areas, which relates to the technical field of data processing and comprises the following steps: dividing a designated sea area to be monitored for suspended matters into a plurality of sea area areas; acquiring a plurality of remote sensing information sets and a plurality of remote sensing reflectivity sets based on multi-resolution remote sensing; analyzing to obtain a plurality of suspension concentration information, a plurality of suspension particle size information, multi-scale fusion remote sensing information and suspension area information; integrating the suspension concentration information, the suspension particle size information and the suspension area information to serve as a suspension remote sensing monitoring result of the designated sea area. The method solves the technical problems of incomplete suspension information extraction and low accuracy caused by too much dependence on the accuracy of remote sensing data in sea area monitoring in the prior art, and achieves the technical effects of improving the accuracy of suspension information extraction and improving the intelligent monitoring level of multiple sea areas through multi-scale remote sensing information extraction.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a remote sensing information extraction system for intelligent monitoring of multiple types of sea areas.
Background
At present, a method for monitoring the concentration of suspended matters mainly comprises a traditional weighing method, a numerical simulation method and a remote sensing inversion method, but the traditional weighing method has larger limitations in the aspects of monitoring range, frequency and the like, the prediction result of the numerical simulation method is generally difficult to verify, the accuracy is lower, the remote sensing inversion method can be used for realizing continuous monitoring of the concentration of suspended matters in a large range, at low cost and periodically, but the prior art is mainly used for monitoring by depending on remote sensing data precision, so that the information extraction accuracy of the suspended matters is low, and the data extraction is incomplete.
Disclosure of Invention
The application provides a remote sensing information extraction system for intelligent monitoring of multi-type sea areas, which is used for solving the technical problems that in the prior art, sea area monitoring is too dependent on remote sensing data precision, so that suspended matter information extraction is incomplete and accuracy is low.
In a first aspect of the present application, there is provided a remote sensing information extraction method for intelligent monitoring of multiple types of sea areas, the method comprising: the sea area division module is used for dividing a designated sea area to be subjected to suspended matter monitoring into a plurality of sea area areas, wherein suspended matters in the designated sea area are generated based on sea reclamation construction; the remote sensing information acquisition module is used for acquiring remote sensing information of a plurality of sea areas based on multi-resolution remote sensing to obtain a plurality of remote sensing information sets, and testing the plurality of sea areas to obtain a plurality of remote sensing reflectivity sets; the suspended matter concentration information acquisition module is used for analyzing and calculating suspended matter concentrations in the sea areas based on the remote sensing reflectivity sets and the integrated multi-scale inversion model to obtain a plurality of suspended matter concentration information; the suspended matter particle size information acquisition module is used for carrying out analysis and calculation on the suspended matter particle sizes in the sea areas based on integrated multi-scale particle size analysis according to the remote sensing information sets to obtain a plurality of suspended matter particle size information; the multi-scale fused remote sensing information acquisition module is used for carrying out multi-scale remote sensing information fusion on the sea area areas based on the remote sensing information sets to acquire multi-scale fused remote sensing information; the suspended matter area information acquisition module is used for carrying out traversal feature extraction processing in the multi-scale fusion remote sensing information based on the suspended matter concentration information and the suspended matter particle size information to obtain suspended matter area information; and the suspended matter remote sensing monitoring result integrating module is used for integrating the suspended matter concentration information, the suspended matter particle size information and the suspended matter area information to serve as a suspended matter remote sensing monitoring result of a designated sea area.
In a second aspect of the present application, there is provided a remote sensing information extraction system for intelligent monitoring of multiple types of sea areas, the system comprising: dividing a designated sea area to be subjected to suspended matter monitoring into a plurality of sea area areas, wherein suspended matters in the designated sea area are generated based on reclamation sea construction; based on multi-resolution remote sensing, acquiring remote sensing information of a plurality of sea areas to obtain a plurality of remote sensing information sets, and testing the plurality of sea areas to obtain a plurality of remote sensing reflectivity sets; based on the remote sensing reflectivity sets and the integrated multi-scale inversion model, analyzing and calculating the suspended matter concentration in the sea areas to obtain a plurality of suspended matter concentration information; according to the remote sensing information sets, based on integrated multi-scale particle size analysis, analyzing and calculating the particle sizes of suspended matters in the sea areas to obtain the particle size information of the suspended matters; based on the multiple remote sensing information sets, performing multi-scale remote sensing information fusion of the multiple sea areas to obtain multi-scale fused remote sensing information; performing traversal feature extraction processing in the multi-scale fusion remote sensing information based on the suspension concentration information and the suspension particle size information to obtain suspension area information; integrating the suspension concentration information, the suspension particle size information and the suspension area information to serve as a suspension remote sensing monitoring result of the designated sea area.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
the application provides a remote sensing information extraction method for intelligent monitoring of multi-type sea areas, which relates to the technical field of data processing, and aims at solving the technical problems that the sea area monitoring is too dependent on the accuracy of remote sensing data, so that the suspended matter information extraction is incomplete and low in accuracy in the prior art, the accuracy of the suspended matter information extraction is improved, and the technical effect of intelligent monitoring level of the multi-type sea areas is improved by acquiring a plurality of remote sensing information sets and a plurality of remote sensing reflectivity sets, analyzing and acquiring a plurality of suspended matter concentration information, a plurality of suspended matter particle size information, multi-scale fusion remote sensing information and suspended matter area information, integrating the plurality of suspended matter concentration information, the plurality of suspended matter particle size information and the suspended matter area information.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a remote sensing information extraction method for intelligent monitoring of multiple types of sea areas according to an embodiment of the present application;
fig. 2 is a schematic flow chart of obtaining particle size information of a plurality of suspended matters in a remote sensing information extraction method for intelligent monitoring of a plurality of sea areas according to an embodiment of the present application;
fig. 3 is a schematic flow chart of calculating and obtaining the suspended area information in the remote sensing information extraction method for intelligent monitoring of multi-type sea areas according to the embodiment of the present application;
fig. 4 is a schematic structural diagram of a remote sensing information extraction system for intelligent monitoring of multiple sea areas according to an embodiment of the present application.
Reference numerals illustrate: the system comprises a sea area dividing module 11, a remote sensing information acquisition module 12, a suspended matter concentration information acquisition module 13, a suspended matter particle size information acquisition module 14, a multi-scale fusion remote sensing information acquisition module 15, a suspended matter area information acquisition module 16 and a suspended matter remote sensing monitoring result integration module 17.
Detailed Description
The application provides a remote sensing information extraction method for intelligent monitoring of multi-type sea areas, which is used for solving the technical problems of incomplete suspension information extraction and low accuracy caused by too much dependence on remote sensing data precision in sea area monitoring in the prior art.
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
Embodiment one:
as shown in fig. 1, the present application provides a remote sensing information extraction method for intelligent monitoring of multiple types of sea areas, the method comprising:
t10: dividing a designated sea area to be subjected to suspended matter monitoring into a plurality of sea area areas, wherein suspended matters in the designated sea area are generated based on reclamation sea construction;
optionally, the application monitors the sea area of suspended matters generated by the reclamation sea construction, divides the appointed sea area to be monitored into a plurality of sea area areas, and can divide the appointed sea area into a plurality of sea area areas with the same area according to a certain area, or divide the appointed sea area into a plurality of sea area areas according to a preset quantity, so that the suspended matters can be respectively monitored for the plurality of sea area areas, and the comprehensiveness and accuracy of monitoring data are improved.
T20: based on multi-resolution remote sensing, acquiring remote sensing information of a plurality of sea areas to obtain a plurality of remote sensing information sets, and testing the plurality of sea areas to obtain a plurality of remote sensing reflectivity sets;
further, step T20 of the embodiment of the present application further includes:
t21: based on a plurality of remote sensing resolutions, acquiring remote sensing monitoring information of a plurality of sea areas to obtain a plurality of remote sensing information sets;
t22: and based on the remote sensing test spectrums of the plurality of wavelengths, testing and collecting the plurality of sea areas to obtain a plurality of remote sensing reflectivity sets.
In one possible embodiment of the present application, remote sensing information of a plurality of sea areas is collected based on multi-resolution remote sensing, where the multi-resolution remote sensing refers to a plurality of remote sensing technologies with different resolutions, where the resolution of a remote sensing image refers to the size of identifiable or distinguishable minimum details or features in an image, the remote sensing information refers to a remote sensing image, and a plurality of remote sensing information sets are formed by a plurality of remote sensing images, and each remote sensing information set includes remote sensing image information with different resolutions of the same sea area.
Further, a plurality of remote sensing test spectrums with different wavelengths are adopted to respectively test and calculate the reflectivity of the plurality of sea areas, an arbitrary sea area is selected by way of example, a portable spectrum radiometer is adopted to conduct water body spectrum observation according to the plurality of remote sensing test spectrums with different wavelengths, actual measurement spectrum data is obtained, the actual measurement spectrum data comprises spectrum radiance information of standard gray plates, water bodies and sky lights, further, remote sensing reflectivity calculation is conducted according to the spectrum radiance information of the standard gray plates, the water bodies and the sky lights and by combining a remote sensing reflectivity calculation formula, a plurality of remote sensing reflectances of the sea area are obtained, and a remote sensing reflectivity set of the sea area is formed. And by analogy, testing and collecting the plurality of sea areas to obtain a plurality of remote sensing reflectivity sets of the plurality of sea areas.
T30: based on the remote sensing reflectivity sets and the integrated multi-scale inversion model, analyzing and calculating the suspended matter concentration in the sea areas to obtain a plurality of suspended matter concentration information;
further, step T30 in the embodiment of the present application further includes:
t31: acquiring a sample suspended matter concentration information set and a plurality of sample remote sensing reflectivity sets based on monitoring data records of suspended matters in the sea area, wherein each sample remote sensing reflectivity set comprises a plurality of remote sensing reflectivities of different sample suspended matter concentrations on a remote sensing test spectrum;
t32: respectively adopting the plurality of sample remote sensing reflectivity sets, and carrying out integrated multi-scale training by combining the sample suspended matter concentration information sets to obtain the integrated multi-scale inversion model;
t33: and respectively inputting the remote sensing reflectivity sets into a plurality of inversion paths in the integrated multi-scale inversion model to obtain a multi-scale inversion result set, and inputting the multi-scale inversion result set into an integrated calculation branch to obtain a plurality of suspension concentration information.
Specifically, based on monitoring data records of sea area suspended matters in a sea area to be monitored, a plurality of sample suspended matter concentration information and a plurality of sample remote sensing reflectivities are collected to form a sample suspended matter concentration information set and a plurality of sample remote sensing reflectivities, wherein each sample remote sensing reflectivity set comprises a plurality of remote sensing reflectivities of different sample suspended matter concentrations to one remote sensing test spectrum.
Further, the plurality of sample remote sensing reflectivity sets and the sample suspended matter concentration information sets are respectively adopted as training data to perform integrated multi-scale training, namely, the inversion model is a model for describing the relationship between remote sensing signals and water body properties under different resolution remote sensing, a plurality of inversion paths aiming at different remote sensing test spectrums can be constructed based on a convolutional neural network, the training data are adopted to perform supervised training until the output of the inversion paths achieves convergence and meets the preset accuracy requirement, and the integrated multi-scale inversion model is formed by the inversion paths.
Further, the multiple remote sensing reflectivity sets are respectively input into multiple inversion paths in the integrated multi-scale inversion model to obtain multi-scale inversion result sets under different remote sensing test spectrums, and are input into an integrated calculation branch to perform weighted calculation to obtain multiple suspended matter concentration information, so that suspended matter concentrations of multiple sea areas can be reflected.
Further, step T32 in the embodiment of the present application further includes:
t32-1: according to machine learning, constructing a plurality of inversion paths, wherein the input of the inversion paths is remote sensing reflectivity obtained by remote sensing test spectrum tests with different wavelengths, and the output is suspended matter concentration;
t32-2: respectively adopting the plurality of sample remote sensing reflectivity sets and the sample suspended matter concentration information sets to carry out supervision training update on network parameters of the plurality of inversion paths;
t32-3: based on a plurality of inversion paths after training, testing and acquiring a plurality of accuracy information;
t32-4: constructing a suspension concentration calculation rule for weighting calculation according to the plurality of accuracy information, and constructing an integrated calculation branch;
t32-5: and connecting the inversion paths with the integrated calculation branch to obtain the integrated multi-scale inversion model.
It should be understood that according to machine learning, a plurality of inversion paths are constructed, wherein the machine learning refers to that a computer automatically learns rules and modes from data and predicts and makes decisions according to the rules and modes, the inversion paths are a plurality of suspended matter concentration prediction models, the input of the inversion paths is the remote sensing reflectivity obtained by remote sensing test spectrum test of different wavelengths, the output is suspended matter concentration, and the training process can be to respectively adopt the plurality of sample remote sensing reflectivity sets and sample suspended matter concentration information sets as training data, and to perform supervised training update on network parameters of the inversion paths until the output reached by the inversion paths reaches convergence.
Further, based on a plurality of inversion paths which are completed through training, a set of test data is adopted to perform accuracy testing, a plurality of accuracy information corresponding to the inversion paths is obtained, a plurality of weight coefficients of the inversion paths are set according to the accuracy, a suspension concentration calculation rule is built according to the weight coefficients, further, an integrated calculation branch is built according to the suspension concentration calculation rule, the suspension concentrations corresponding to remote sensing test spectrums with different wavelengths are used for weighting, so that the suspension concentrations of corresponding sea areas are obtained, and further, the inversion paths are connected with the integrated calculation branch, so that the integrated multi-scale inversion model is obtained.
T40: according to the remote sensing information sets, based on integrated multi-scale particle size analysis, analyzing and calculating the particle sizes of suspended matters in the sea areas to obtain the particle size information of the suspended matters;
further, as shown in fig. 2, step T40 in the embodiment of the present application further includes:
t41: acquiring a sample suspended matter particle size information set and a plurality of sample remote sensing information sets based on remote sensing data records of suspended matters in the sea area, wherein each sample remote sensing information set comprises a plurality of sample remote sensing information with different suspended matter particle sizes under one remote sensing resolution;
t42: respectively adopting the plurality of sample remote sensing information sets, and carrying out integrated multi-scale training by combining the sample suspension particle size information sets to obtain an integrated multi-scale particle size analysis model;
t43: inputting the remote sensing information with different resolutions in the remote sensing information sets into a plurality of particle size analysis paths in the integrated multi-scale particle size analysis model respectively to obtain a plurality of suspended matter particle size information sets;
t44: and respectively carrying out integrated weighted calculation on the plurality of suspension particle size information sets to obtain the plurality of suspension particle size information.
The method includes the steps of collecting and acquiring a sample suspension particle size information set and a plurality of sample remote sensing information sets based on remote sensing data records of sea area suspension in a sea area to be monitored, wherein each sample remote sensing information set contains sample remote sensing information with different suspension particle sizes under one remote sensing resolution, the method is referred to in the step T32, the plurality of sample remote sensing information sets and the sample suspension particle size information set are respectively adopted as training data to perform integrated multi-scale training, a plurality of particle size analysis paths aiming at different remote sensing resolutions can be constructed based on a convolutional neural network, and the training data is adopted to perform supervised training until the output of the plurality of particle size analysis paths is converged and meets the preset accuracy requirement, and the integrated multi-scale particle size analysis model is formed by the plurality of particle size analysis paths.
Further, remote sensing information with different resolutions in the remote sensing information sets is respectively input into a plurality of particle size analysis paths in the integrated multi-scale particle size analysis model to obtain a plurality of suspended matter particle size information sets, corresponding weight coefficients are distributed according to output accuracy of the plurality of particle size analysis paths, integrated weighted calculation is carried out on the plurality of suspended matter particle size information sets to obtain the plurality of suspended matter particle size information, and the suspended matter particle sizes of a plurality of sea areas can be represented.
T50: based on the multiple remote sensing information sets, performing multi-scale remote sensing information fusion of the multiple sea areas to obtain multi-scale fused remote sensing information;
optionally, based on the multiple remote sensing information sets, acquiring multiple remote sensing image sets with different resolutions, and respectively fusing the multi-scale remote sensing information of the multiple sea area areas according to the collection positions and the shooting picture ranges, so as to obtain an integral remote sensing image of the whole sea area to be monitored, namely fused remote sensing information, wherein the multi-scale refers to multiple remote sensing resolutions, and the multi-scale fused remote sensing information, namely the integral remote sensing image of the sea area to be monitored with different resolutions, can be obtained according to the remote sensing image sets with different resolutions.
T60: performing traversal feature extraction processing in the multi-scale fusion remote sensing information based on the suspension concentration information and the suspension particle size information to obtain suspension area information;
further, as shown in fig. 3, step T60 in the embodiment of the present application further includes:
t61: performing traversal division of local areas in a plurality of pieces of fused remote sensing information in the multi-scale fused remote sensing information according to the size of the preset local areas and the preset traversal step length to obtain a plurality of local area sets;
t62: based on the suspension concentration information, matching is carried out in the traversing dividing process of the local area sets, and a plurality of first matching areas are obtained;
t63: based on the particle size information of the suspended matters, matching is carried out in the traversing dividing process of the local area set, and a plurality of second matching areas are obtained;
t64: and screening out non-repeated local areas in the first matching areas and the second matching areas, fusing to obtain suspended matter areas, and calculating to obtain the suspended matter area information.
It should be understood that, in the multiple fused remote sensing information in the multi-scale fused remote sensing information, that is, in the multiple integrated remote sensing images with different resolutions, the local area is divided according to the size of a preset local area and a preset traversal step, where the preset local area is a preset feature extraction area, for example, a pixel grid of 5*5, and the preset traversal step is the distance of each movement of the preset local area, and the integrated remote sensing image is segmented according to the size of the preset local area, so as to obtain multiple local area sets.
Further, based on the plurality of suspension concentration information, suspension concentration matching is performed in the traversing dividing process of the plurality of local area sets, that is, suspension concentration identification and matching are performed in the plurality of local areas according to the plurality of suspension concentration information, a plurality of areas consistent with the plurality of suspension concentration information are screened out as a plurality of first matching areas, and similarly, suspension particle size matching is performed in the traversing dividing process of the local area sets based on the plurality of suspension particle size information, so that a plurality of second matching areas are obtained. Further, the non-repeated local areas in the first matching areas and the second matching areas are screened and extracted, the non-repeated local areas are fused to obtain a suspended matter area image, and further the suspended matter area information is obtained through area calculation, so that the suspended matter area of the sea area to be monitored can be represented.
Further, step T62 of the embodiment of the present application further includes:
t62-1: dividing local areas in the fused remote sensing information according to the size of the preset local area and preset traversal compensation to obtain a plurality of first local areas;
t62-2: performing identification and matching of suspended matter concentrations in the plurality of first local areas based on the plurality of suspended matter concentration information;
t62-3: and continuing to divide the local areas and identify and match the suspended solids concentration so as to obtain the plurality of first matching areas.
The method comprises the steps of dividing local areas in the fused remote sensing information according to the size of the preset local areas and preset traversal compensation, namely dividing the fused remote sensing image according to the size of the local areas and preset traversal compensation, obtaining a plurality of first local areas, further, based on the suspended matter concentration information, carrying out identification and matching of suspended matter concentration in the first local areas, namely identifying suspended matter concentration of each local area, carrying out identification based on the integrated multi-scale inversion model, then carrying out matching on the identified suspended matter concentration and the suspended matter concentration, and if the identified suspended matter concentration is similar to or the same as any one of the suspended matter concentrations, taking the area as the matched suspended matter area, and further, continuing to carry out division and suspended matter concentration identification matching of the first local areas, namely obtaining a plurality of first matching areas.
T70: integrating the suspension concentration information, the suspension particle size information and the suspension area information to serve as a suspension remote sensing monitoring result of the designated sea area.
Optionally, the suspension concentration information, the suspension particle size information and the suspension area information are used together as suspension remote sensing monitoring results of the designated sea area, and the suspension concentration and the suspension particle size of the designated sea area and the whole suspension area size of the designated sea area can be reflected.
In summary, the embodiments of the present application have at least the following technical effects:
the method comprises the steps of dividing a designated sea area to be monitored for suspended matters into a plurality of sea areas, acquiring a plurality of remote sensing information sets and a plurality of remote sensing reflectivity sets based on multi-resolution remote sensing, analyzing and acquiring a plurality of suspended matter concentration information, a plurality of suspended matter particle size information, multi-scale fusion remote sensing information and suspended matter area information, and integrating the plurality of suspended matter concentration information, the plurality of suspended matter particle size information and the suspended matter area information to serve as suspended matter remote sensing monitoring results of the designated sea area.
The technical effects of improving the accuracy of suspended matter information extraction and improving the intelligent monitoring level of multiple types of sea areas through multi-scale remote sensing information extraction are achieved.
Embodiment two:
based on the same inventive concept as the remote sensing information extraction method for multi-type sea area intelligent monitoring in the foregoing embodiments, as shown in fig. 4, the present application provides a remote sensing information extraction system for multi-type sea area intelligent monitoring, and the system and method embodiments in the embodiments of the present application are based on the same inventive concept. Wherein the system comprises:
a sea area division module 11, wherein the sea area division module 11 is configured to divide a specified sea area to be monitored for suspended matters into a plurality of sea areas, and the suspended matters in the specified sea area are generated based on the construction of the reclamation sea;
the remote sensing information acquisition module 12 is used for acquiring remote sensing information of a plurality of sea areas based on multi-resolution remote sensing to obtain a plurality of remote sensing information sets, and testing the plurality of sea areas to obtain a plurality of remote sensing reflectivity sets;
the suspended matter concentration information acquisition module 13 is used for analyzing and calculating suspended matter concentrations in the sea areas based on the remote sensing reflectivity sets and the integrated multi-scale inversion model to obtain a plurality of suspended matter concentration information;
the suspended matter particle size information obtaining module 14 is configured to perform analysis and calculation of suspended matter particle sizes in the plurality of sea areas based on integrated multi-scale particle size analysis according to the plurality of remote sensing information sets, so as to obtain a plurality of suspended matter particle size information;
the multi-scale fused remote sensing information obtaining module 15 is configured to perform multi-scale remote sensing information fusion of the plurality of sea areas based on the plurality of remote sensing information sets, so as to obtain multi-scale fused remote sensing information;
the suspended matter area information acquisition module 16, wherein the suspended matter area information acquisition module 16 is configured to perform traversal feature extraction processing in the multiscale fused remote sensing information based on the plurality of suspended matter concentration information and the plurality of suspended matter particle size information, so as to obtain suspended matter area information;
the suspended matter remote sensing monitoring result integrating module 17 is used for integrating the suspended matter concentration information, the suspended matter particle size information and the suspended matter area information to serve as a suspended matter remote sensing monitoring result of the designated sea area.
Further, the remote sensing information acquisition module 12 is further configured to perform the following steps:
based on a plurality of remote sensing resolutions, acquiring remote sensing monitoring information of a plurality of sea areas to obtain a plurality of remote sensing information sets;
and based on the remote sensing test spectrums of the plurality of wavelengths, testing and collecting the plurality of sea areas to obtain a plurality of remote sensing reflectivity sets.
Further, the suspension concentration information obtaining module 13 is further configured to perform the following steps:
acquiring a sample suspended matter concentration information set and a plurality of sample remote sensing reflectivity sets based on monitoring data records of suspended matters in the sea area, wherein each sample remote sensing reflectivity set comprises a plurality of remote sensing reflectivities of different sample suspended matter concentrations on a remote sensing test spectrum;
respectively adopting the plurality of sample remote sensing reflectivity sets, and carrying out integrated multi-scale training by combining the sample suspended matter concentration information sets to obtain the integrated multi-scale inversion model;
and respectively inputting the remote sensing reflectivity sets into a plurality of inversion paths in the integrated multi-scale inversion model to obtain a multi-scale inversion result set, and inputting the multi-scale inversion result set into an integrated calculation branch to obtain a plurality of suspension concentration information.
Further, the suspension concentration information obtaining module 13 is further configured to perform the following steps:
according to machine learning, constructing a plurality of inversion paths, wherein the input of the inversion paths is remote sensing reflectivity obtained by remote sensing test spectrum tests with different wavelengths, and the output is suspended matter concentration;
respectively adopting the plurality of sample remote sensing reflectivity sets and the sample suspended matter concentration information sets to carry out supervision training update on network parameters of the plurality of inversion paths;
based on a plurality of inversion paths after training, testing and acquiring a plurality of accuracy information;
constructing a suspension concentration calculation rule for weighting calculation according to the plurality of accuracy information, and constructing an integrated calculation branch;
and connecting the inversion paths with the integrated calculation branch to obtain the integrated multi-scale inversion model.
Further, the suspension particle size information obtaining module 14 is further configured to perform the following steps:
acquiring a sample suspended matter particle size information set and a plurality of sample remote sensing information sets based on remote sensing data records of suspended matters in the sea area, wherein each sample remote sensing information set comprises a plurality of sample remote sensing information with different suspended matter particle sizes under one remote sensing resolution;
respectively adopting the plurality of sample remote sensing information sets, and carrying out integrated multi-scale training by combining the sample suspension particle size information sets to obtain an integrated multi-scale particle size analysis model;
inputting the remote sensing information with different resolutions in the remote sensing information sets into a plurality of particle size analysis paths in the integrated multi-scale particle size analysis model respectively to obtain a plurality of suspended matter particle size information sets;
and respectively carrying out integrated weighted calculation on the plurality of suspension particle size information sets to obtain the plurality of suspension particle size information.
Further, the suspended matter area information obtaining module 16 is further configured to perform the following steps:
performing traversal division of local areas in a plurality of pieces of fused remote sensing information in the multi-scale fused remote sensing information according to the size of the preset local areas and the preset traversal step length to obtain a plurality of local area sets;
based on the suspension concentration information, matching is carried out in the traversing dividing process of the local area sets, and a plurality of first matching areas are obtained;
based on the particle size information of the suspended matters, matching is carried out in the traversing dividing process of the local area set, and a plurality of second matching areas are obtained;
and screening out non-repeated local areas in the first matching areas and the second matching areas, fusing to obtain suspended matter areas, and calculating to obtain the suspended matter area information.
Further, the suspended matter area information obtaining module 16 is further configured to perform the following steps:
dividing local areas in the fused remote sensing information according to the size of the preset local area and preset traversal compensation to obtain a plurality of first local areas;
performing identification and matching of suspended matter concentrations in the plurality of first local areas based on the plurality of suspended matter concentration information;
and continuing to divide the local areas and identify and match the suspended solids concentration so as to obtain the plurality of first matching areas.
The foregoing description of the preferred embodiments of the present application is not intended to limit the invention to the particular embodiments of the present application, but to limit the scope of the invention to the particular embodiments of the present application.
Claims (6)
1. A remote sensing information extraction system for intelligent monitoring of multiple types of sea areas, the system comprising:
the sea area division module is used for dividing a designated sea area to be subjected to suspended matter monitoring into a plurality of sea area areas, wherein suspended matters in the designated sea area are generated based on sea reclamation construction;
the remote sensing information acquisition module is used for acquiring remote sensing information of a plurality of sea areas based on multi-resolution remote sensing to obtain a plurality of remote sensing information sets, and testing the plurality of sea areas to obtain a plurality of remote sensing reflectivity sets;
the suspended matter concentration information acquisition module is used for analyzing and calculating suspended matter concentrations in the sea areas based on the remote sensing reflectivity sets and the integrated multi-scale inversion model to obtain a plurality of suspended matter concentration information;
the suspended matter particle size information acquisition module is used for carrying out analysis and calculation on the suspended matter particle sizes in the sea areas based on integrated multi-scale particle size analysis according to the remote sensing information sets to obtain a plurality of suspended matter particle size information;
the multi-scale fused remote sensing information acquisition module is used for carrying out multi-scale remote sensing information fusion on the sea area areas based on the remote sensing information sets to acquire multi-scale fused remote sensing information;
the suspended matter area information acquisition module is used for carrying out traversal feature extraction processing in the multi-scale fusion remote sensing information based on the suspended matter concentration information and the suspended matter particle size information to obtain suspended matter area information;
the suspended matter remote sensing monitoring result integrating module is used for integrating the suspended matter concentration information, the suspended matter particle size information and the suspended matter area information to serve as a suspended matter remote sensing monitoring result of a designated sea area;
the system further comprises:
the system comprises a sample suspended matter concentration information acquisition module, a remote sensing detection module and a remote sensing detection module, wherein the sample suspended matter concentration information acquisition module is used for acquiring a sample suspended matter concentration information set and a plurality of sample remote sensing reflectivity sets based on monitoring data records of sea area suspended matters, and each sample remote sensing reflectivity set comprises a plurality of remote sensing reflectivities of different sample suspended matter concentrations for one remote sensing test spectrum;
the integrated multi-scale inversion model training module is used for carrying out integrated multi-scale training by respectively adopting the plurality of sample remote sensing reflectivity sets and combining the sample suspended matter concentration information set to obtain the integrated multi-scale inversion model;
the suspended matter concentration information calculation module is used for respectively inputting the remote sensing reflectivity sets into a plurality of inversion paths in the integrated multi-scale inversion model to obtain a multi-scale inversion result set, and inputting the multi-scale inversion result set into an integrated calculation branch to obtain a plurality of suspended matter concentration information;
the system comprises a sample suspended matter particle size information acquisition module, a remote sensing resolution detection module and a remote sensing resolution detection module, wherein the sample suspended matter particle size information acquisition module is used for acquiring a sample suspended matter particle size information set and a plurality of sample remote sensing information sets based on remote sensing data records of sea area suspended matters, and each sample remote sensing information set comprises a plurality of sample remote sensing information with different suspended matter particle sizes under the remote sensing resolution;
the integrated multi-scale particle size analysis model training module is used for respectively adopting the plurality of sample remote sensing information sets and combining the sample suspended matter particle size information sets to perform integrated multi-scale training to obtain an integrated multi-scale particle size analysis model;
the suspended matter particle size information set acquisition module is used for respectively inputting the remote sensing information with different resolutions in the plurality of remote sensing information sets into a plurality of particle size analysis paths in the integrated multi-scale particle size analysis model to obtain a plurality of suspended matter particle size information sets;
and the suspension particle size information calculation module is used for respectively carrying out integrated weighted calculation on the suspension particle size information sets to obtain the suspension particle size information.
2. The system according to claim 1, characterized in that the system comprises:
the remote sensing information collection acquisition module is used for acquiring remote sensing monitoring information of the sea areas based on a plurality of remote sensing resolutions to acquire a plurality of remote sensing information collections;
the remote sensing reflectivity set acquisition module is used for testing and acquiring the sea areas based on remote sensing test spectrums of a plurality of wavelengths to obtain a plurality of remote sensing reflectivity sets.
3. The system according to claim 1, characterized in that the system comprises:
the inversion path construction module is used for constructing a plurality of inversion paths according to machine learning, wherein the input of the inversion paths is remote sensing reflectivity obtained by remote sensing test spectrum tests with different wavelengths, and the output is suspended matter concentration;
the monitoring training updating module is used for performing monitoring training updating on network parameters of the inversion paths by adopting the plurality of sample remote sensing reflectivity sets and the sample suspended matter concentration information sets respectively;
the accuracy testing module is used for testing and acquiring a plurality of accuracy information based on a plurality of inversion paths after training;
the integrated calculation branch construction module is used for constructing a suspension concentration calculation rule for weighting calculation according to the plurality of accuracy information and constructing an integrated calculation branch;
the integrated multi-scale inversion model connecting module is used for connecting the inversion paths with the integrated calculation branches to obtain the integrated multi-scale inversion model.
4. The system according to claim 1, characterized in that the system comprises:
the local area set acquisition module is used for carrying out traversal division on the local area in the multiple pieces of fused remote sensing information in the multi-scale fused remote sensing information according to the size of the preset local area and the preset traversal step length to obtain multiple local area sets;
the first matching region acquisition module is used for carrying out matching in the traversing dividing process of the local region sets based on the suspended matter concentration information to obtain a plurality of first matching regions;
the second matching region acquisition module is used for carrying out matching in the traversing dividing process of the local region set based on the particle size information of the suspended matters to obtain a plurality of second matching regions;
and the suspended matter area calculation module is used for screening out non-repeated local areas in the plurality of first matching areas and the plurality of second matching areas, fusing the non-repeated local areas to obtain suspended matter areas, and calculating to obtain the suspended matter area information.
5. The system of claim 4, wherein the system comprises:
the first local area acquisition module is used for carrying out local area division in the plurality of fused remote sensing information according to the size of the preset local area and preset traversal compensation to obtain a plurality of first local areas;
a suspended matter concentration identification matching module for identifying and matching suspended matter concentrations in the plurality of first partial areas based on the plurality of suspended matter concentration information;
the first matching region acquisition module is used for continuing to divide local regions and identify and match suspended solids concentration so as to obtain the plurality of first matching regions.
6. A method for extracting remote sensing information for intelligent monitoring of multiple types of sea areas, the method comprising:
dividing a designated sea area to be subjected to suspended matter monitoring into a plurality of sea area areas, wherein suspended matters in the designated sea area are generated based on reclamation sea construction;
based on multi-resolution remote sensing, acquiring remote sensing information of a plurality of sea areas to obtain a plurality of remote sensing information sets, and testing the plurality of sea areas to obtain a plurality of remote sensing reflectivity sets;
based on the remote sensing reflectivity sets and the integrated multi-scale inversion model, analyzing and calculating the suspended matter concentration in the sea areas to obtain a plurality of suspended matter concentration information;
according to the remote sensing information sets, based on integrated multi-scale particle size analysis, analyzing and calculating the particle sizes of suspended matters in the sea areas to obtain the particle size information of the suspended matters;
based on the multiple remote sensing information sets, performing multi-scale remote sensing information fusion of the multiple sea areas to obtain multi-scale fused remote sensing information;
performing traversal feature extraction processing in the multi-scale fusion remote sensing information based on the suspension concentration information and the suspension particle size information to obtain suspension area information;
integrating the suspension concentration information, the suspension particle size information and the suspension area information to serve as suspension remote sensing monitoring results of the designated sea area;
based on an integrated multi-scale inversion model, analyzing and calculating the suspended matter concentration in the plurality of sea areas to obtain a plurality of suspended matter concentration information, wherein the method comprises the following steps:
acquiring a sample suspended matter concentration information set and a plurality of sample remote sensing reflectivity sets based on monitoring data records of suspended matters in the sea area, wherein each sample remote sensing reflectivity set comprises a plurality of remote sensing reflectivities of different sample suspended matter concentrations on a remote sensing test spectrum;
respectively adopting the plurality of sample remote sensing reflectivity sets, and carrying out integrated multi-scale training by combining the sample suspended matter concentration information sets to obtain the integrated multi-scale inversion model;
respectively inputting the remote sensing reflectivity sets into a plurality of inversion paths in the integrated multi-scale inversion model to obtain a multi-scale inversion result set, and inputting the multi-scale inversion result set into an integrated calculation branch to obtain a plurality of suspension concentration information;
based on the integrated multi-scale particle size analysis, performing an analytical calculation of the particle size of the suspended matter in the plurality of sea areas to obtain a plurality of pieces of information of the particle size of the suspended matter, including:
acquiring a sample suspended matter particle size information set and a plurality of sample remote sensing information sets based on remote sensing data records of suspended matters in the sea area, wherein each sample remote sensing information set comprises a plurality of sample remote sensing information with different suspended matter particle sizes under one remote sensing resolution;
respectively adopting the plurality of sample remote sensing information sets, and carrying out integrated multi-scale training by combining the sample suspension particle size information sets to obtain an integrated multi-scale particle size analysis model;
inputting the remote sensing information with different resolutions in the remote sensing information sets into a plurality of particle size analysis paths in the integrated multi-scale particle size analysis model respectively to obtain a plurality of suspended matter particle size information sets;
and respectively carrying out integrated weighted calculation on the plurality of suspension particle size information sets to obtain the plurality of suspension particle size information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311534117.3A CN117274819B (en) | 2023-11-17 | 2023-11-17 | Remote sensing information extraction system for intelligent monitoring of multiple types of sea areas |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311534117.3A CN117274819B (en) | 2023-11-17 | 2023-11-17 | Remote sensing information extraction system for intelligent monitoring of multiple types of sea areas |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117274819A CN117274819A (en) | 2023-12-22 |
| CN117274819B true CN117274819B (en) | 2024-02-23 |
Family
ID=89204814
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311534117.3A Active CN117274819B (en) | 2023-11-17 | 2023-11-17 | Remote sensing information extraction system for intelligent monitoring of multiple types of sea areas |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117274819B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005052697A (en) * | 2003-08-06 | 2005-03-03 | Hitachi Ltd | Water quality monitoring system |
| CN105606498A (en) * | 2016-02-03 | 2016-05-25 | 南京信息工程大学 | Method for monitoring particle size distribution of suspended sediment in water |
| CN106960204A (en) * | 2017-05-05 | 2017-07-18 | 南京信息工程大学 | A kind of remote sensing technique of estimation coastal waters water phytoplankton Grain grade structure |
| CN107219171A (en) * | 2017-05-18 | 2017-09-29 | 杭州师范大学 | The remote-sensing monitoring method of polycyclic aromatic hydrocarbon in a kind of sea surface Suspended Particles in Water |
| CN109917098A (en) * | 2019-04-09 | 2019-06-21 | 中国科学院深圳先进技术研究院 | Shenzhen waters quality evaluation method and system |
| KR102143649B1 (en) * | 2019-04-26 | 2020-08-11 | 주식회사 하이드로봇테크앤리서치 | Observation Buoy of the concentration of Suspended Sediment and Real-time Efficiency Estimation System of Silt Protector |
| CN111538940A (en) * | 2020-05-12 | 2020-08-14 | 自然资源部第二海洋研究所 | Suspended matter concentration inversion model determination method and suspended matter concentration determination method |
| CN112051226A (en) * | 2020-09-03 | 2020-12-08 | 山东省科学院海洋仪器仪表研究所 | Method for estimating total suspended matter concentration of offshore area based on unmanned aerial vehicle-mounted hyperspectral image |
| CN115797760A (en) * | 2023-01-29 | 2023-03-14 | 水利部交通运输部国家能源局南京水利科学研究院 | Active and passive fusion water quality three-dimensional remote sensing inversion method and system and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107144506B (en) * | 2017-06-21 | 2023-08-22 | 华南理工大学 | Suspended matter dynamic monitoring method and device based on annular interweaved array |
-
2023
- 2023-11-17 CN CN202311534117.3A patent/CN117274819B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005052697A (en) * | 2003-08-06 | 2005-03-03 | Hitachi Ltd | Water quality monitoring system |
| CN105606498A (en) * | 2016-02-03 | 2016-05-25 | 南京信息工程大学 | Method for monitoring particle size distribution of suspended sediment in water |
| CN106960204A (en) * | 2017-05-05 | 2017-07-18 | 南京信息工程大学 | A kind of remote sensing technique of estimation coastal waters water phytoplankton Grain grade structure |
| CN107219171A (en) * | 2017-05-18 | 2017-09-29 | 杭州师范大学 | The remote-sensing monitoring method of polycyclic aromatic hydrocarbon in a kind of sea surface Suspended Particles in Water |
| CN109917098A (en) * | 2019-04-09 | 2019-06-21 | 中国科学院深圳先进技术研究院 | Shenzhen waters quality evaluation method and system |
| KR102143649B1 (en) * | 2019-04-26 | 2020-08-11 | 주식회사 하이드로봇테크앤리서치 | Observation Buoy of the concentration of Suspended Sediment and Real-time Efficiency Estimation System of Silt Protector |
| CN111538940A (en) * | 2020-05-12 | 2020-08-14 | 自然资源部第二海洋研究所 | Suspended matter concentration inversion model determination method and suspended matter concentration determination method |
| CN112051226A (en) * | 2020-09-03 | 2020-12-08 | 山东省科学院海洋仪器仪表研究所 | Method for estimating total suspended matter concentration of offshore area based on unmanned aerial vehicle-mounted hyperspectral image |
| CN115797760A (en) * | 2023-01-29 | 2023-03-14 | 水利部交通运输部国家能源局南京水利科学研究院 | Active and passive fusion water quality three-dimensional remote sensing inversion method and system and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| Remote sensing monitoring of total suspended solids concentration in Jiaozhou Bay based on multi-source data;Xiang Zhang;《Ecological Indicators》;第154卷;全文 * |
| 填海施工区悬浮物浓度遥感监测方法研究;宋南奇;《应用基础与工程科学学报》;第28卷(第5期);1108-1121 * |
| 填海施工区悬浮物浓度遥感监测方法研究;宋南奇;王诺;吴暖;林婉妮;;应用基础与工程科学学报(第05期);107-120 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117274819A (en) | 2023-12-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103528617B (en) | A kind of cockpit instrument identifies and detection method and device automatically | |
| CN108846835B (en) | Image change detection method based on depth separable convolutional network | |
| CN111797890A (en) | Method and system for detecting defects of power transmission line equipment | |
| CN111598881A (en) | Image anomaly detection method based on variational self-encoder | |
| Puno et al. | Determination of soil nutrients and pH level using image processing and artificial neural network | |
| CN109800631A (en) | Fluorescence-encoded micro-beads image detecting method based on masked areas convolutional neural networks | |
| CN113989662B (en) | Remote sensing image fine-grained target identification method based on self-supervision mechanism | |
| CN108615046A (en) | A kind of stored-grain pests detection recognition methods and device | |
| Li et al. | Automatic bridge crack identification from concrete surface using ResNeXt with postprocessing | |
| CN110045015A (en) | A kind of concrete structure Inner Defect Testing method based on deep learning | |
| CN104680185B (en) | Hyperspectral image classification method based on boundary point reclassification | |
| CN117152152B (en) | Production management system and method for detection kit | |
| CN111709775A (en) | House property price evaluation method and device, electronic equipment and storage medium | |
| CN116843650A (en) | SMT welding defect detection method and system integrating AOI detection and deep learning | |
| CN109450406A (en) | A kind of filter construction based on Recognition with Recurrent Neural Network | |
| CN108960413A (en) | A kind of depth convolutional neural networks method applied to screw surface defects detection | |
| EP3818468A1 (en) | Method and apparatus for training a neural network classifier to classify an image depicting one or more objects of a biological sample | |
| CN117274819B (en) | Remote sensing information extraction system for intelligent monitoring of multiple types of sea areas | |
| CN108664936B (en) | Diagnosis method and system based on machine fault | |
| CN117076928A (en) | Bridge health state monitoring method, device and system and electronic equipment | |
| CN111368823B (en) | Pointer type instrument reading identification method and device | |
| CN115601293A (en) | Object detection method and device, electronic equipment and readable storage medium | |
| CN115375925A (en) | Underwater sonar image matching algorithm based on phase information and deep learning | |
| Sheta et al. | Autonomous robot system for pavement crack inspection based cnn model | |
| CN111898314B (en) | Lake water parameter inspection method and device, electronic equipment and storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |