CN105427305B - A kind of green tide information extracting method - Google Patents
A kind of green tide information extracting method Download PDFInfo
- Publication number
- CN105427305B CN105427305B CN201510800816.7A CN201510800816A CN105427305B CN 105427305 B CN105427305 B CN 105427305B CN 201510800816 A CN201510800816 A CN 201510800816A CN 105427305 B CN105427305 B CN 105427305B
- Authority
- CN
- China
- Prior art keywords
- green tide
- image
- green
- formula
- region
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30181—Earth observation
- G06T2207/30188—Vegetation; Agriculture
Landscapes
- Image Processing (AREA)
Abstract
The invention belongs to technical field of remote sensing image processing, it is related to a kind of green tide information extracting method, conventionally obtain sea area satellite remote sensing images, it screens the image in green tide region and carries out the pretreatment operation of geometric correction and image mosaic, green tide range is calculated and determined using vegetation index again, the irregular image file that arbitrary shape region of interest generates the region containing green tide is cut again, then green tide information is extracted using the graceful fast projection algorithm of the division Donald Bragg of the old-Wei Si model based on variation level set two-phase image segmentation, finally use quantitative formula quantum chemical method green tide region, it is accurately extracted on satellite remote sensing images based on the Variation Model of image segmentation and quantifies green tide information, traditional artificial threshold method can be substituted completely, realize operational use；Its scientific in principle is reasonable, and human factor is few, and calculating speed is fast, as a result precise and stable, strong operability, and application environment is friendly, and practicability is good, easy to spread.
Description
Technical field:
The invention belongs to technical field of remote sensing image processing, are related to a kind of green tide information extracting method, using image segmentation
Variation Model extracted on satellite remote sensing images green tide information and to extract result quantify.
Background technique:
Enteromorpha is a category of Chlorophyta Chlorophyceae Ulvaceae, is also tongue bar and green moss, is grown in the sea, frond is in bud green
Color, plant is very very thin, and chlorophyll a and b are contained as terrestrial plant, visually seems in green filament shape, has and permitted
Multiple-limb.Enteromorpha itself is nontoxic, it might even be possible to which edible, the resident of Chinese one band of Jiangsu and Zhejiang Provinces can cultivate on a small scale Enteromorpha in offshore and eat
With Enteromorpha product；Enteromorpha assembles outburst under the marine environmental conditions of preference temperature, and the phenomenon of ecological environment exception is caused to be claimed
For Enteromorpha green tide, abbreviation green tide.The harm of green tide mainly has the following: first is that Enteromorpha floating consumes a large amount of oxygen across the sea
Gas causes to be covered on undersea biology by Enteromorpha because anoxic influences growth and development even death；Second is that the object of Enteromorpha secretion
Matter, which deposits to seabed, influences the growth and development of benthon；Third is that covering fishery cultivating base after Enteromorpha offshore, cultivation is caused
The underproduction of marine animal and plant is even dead, brings huge property and emotional distress to raiser；Fourth is that Enteromorpha is to seashore
It travels, spend a holiday and marine regatta sailing causes to seriously affect, destroy the Tourist Destination Image of coastal city, the finance for reducing government is received
Enter；Currently, the main monitoring mode of Enteromorpha green tide has satellite remote sensing, aircraft, ship and land bank to make an inspection tour, satellite remote sensing has monitoring
The advantage that range is wide, spatial and temporal resolution is high and monitoring means is various becomes the major way of Enteromorpha green tide monitoring, distant using satellite
Feel obtained green tide monitoring result, quickly analyzed, identified and interpreted, interpretation result is reported to relevant departments in time, is
Policymaker provides the monitoring analysis of forefront as a result, to formulate counter-measure in time, by Enteromorpha green tide disaster to the shadow of society
Ring be reduced to it is minimum, have important social effect；Green tide information extraction during interpretation directly affects subsequent statistical analysis
Generation and prediction and warning work with interpreting report.Green tide information extraction in the prior art mainly uses visual interpretation and artificial
The method of given threshold exists although this method is easy to operate and depends on expertise, and human factor causes result smart
Exactness difference and unstable problem, and when satellite remote sensing images difference in brightness is larger, it needs to set multiple threshold values and mentions respectively
It takes, each picture dot is involved in wave band calculating, and computational efficiency is low, in green tide large-scale outbreak, is unable to satisfy requirement.Therefore, it grinds
It sends out green tide information extracting method a kind of, extracts green tide information and right on satellite remote sensing images using the Variation Model of image segmentation
It extracts result to be quantified, there is good society and economic value, have a extensive future.
Summary of the invention:
It is an object of the invention to overcome disadvantage of the existing technology, seek to design a kind of green tide information extracting method,
Green tide information is extracted on satellite remote sensing images using the Variation Model of image segmentation, to the green tide result amount of progress extracted
Change, obtains accurate green tide and extract area.
To achieve the goals above, green tide information extracting method technique of the present invention includes screening and preprocessed data
Source is calculated and determined green tide range, cuts irregular green tide range image, extracts green tide information and quantum chemical method green tide region five
A step:
(1), screening with preprocessed data source: it is fine, partly cloudy, without rain and without the weather of typhoon, according to conventional side
Method obtains sea area satellite remote sensing images, and screening noise is few, clarity is high, brightness uniformity and image resolution ratio are less than or equal to 30 meters simultaneously
And the image in green tide region can be covered, geometric correction is carried out to the image data filtered out and the pretreatment of image mosaic is grasped
Make, completes the screening and pretreatment of data source；
(2), green tide range is calculated and determined: for the biological nature and spectral characteristic of Enteromorpha, being planted using normalization difference
Green tide range is calculated and determined by index, formula is as follows:
IR indicates the near infrared band in satellite remote sensing images in formula, selects 4 wave band datas, and R is red wave band, selects 3 waves
Segment data, the vegetation index for completing satellite remote sensing images calculate；
(3), it cuts irregular green tide range image: determining green tide model according to vegetation index calculation method
It encloses, cut arbitrary shape region of interest and records the pixel total number and the gross area of region of interest, generate region of interest data
Collect file, the irregular image in the region containing green tide is generated after remote sensing processing platform imports the region of interest document data set generated
File completes the cutting of irregular green tide range image；
(4), green tide information is extracted:
Using the graceful fast projection algorithm of division Donald Bragg of the old-Wei Si model based on variation level set two-phase image segmentation
Green tide information is extracted, old-Wei Si model energy functional is as follows:
In formula (1-2),
R(u1, u2)=α1(u1-f)2-α2(u2-f)2 (1-4)
Wherein u1And u2The foreground and background of image is respectively indicated, f indicates the original image of Noise, introduces auxiliary division
Variable enablesBecause with equivalence, by formula (1-2) and formula (1-3) difference
It is rewritten as formula (1-5) and formula (1-6):
In formula (1-5),
For the graceful iterative parameter of Donald Bragg, the punishment parameter that θ is positive,WithInitial value is 0,
Formula (1-5) and formula (1-6) are solved to obtain:
In formula (1-9), g is edge indicator function, formula are as follows:
In formula (1-10), GσThe gaussian kernel function for being σ for standard deviation, sets initial parameter as γ=1, and σ=1, θ=
3000, it brings formula (1-8) into and formula (1-9) is solved to obtain green tide information extraction as a result, completing the extraction of green tide information；
(5), quantum chemical method green tide region: the green tide information quantization that step (4) is extracted using quantitative formula, quantitative formula
Are as follows:
Last_area=(clip_pixel-last_sum)/clip_pixel × clip_area (1-11)
In formula (1-11), Last_area is the gross area for the green tide information extracted, and clip_pixel is irregular image
Picture element total number, last_sum are the background pixels point total number in irregular image, and clip_area is irregular image
The gross area, unit are square kilometre to bring given data into, obtain the green tide information gross area, and unit is square kilometre to complete green tide
The quantum chemical method of information.
Vegetation index of the present invention is to determine the coverage condition of green vegetation in satellite remote sensing field
Mainly by the way of, vegetation index is higher, and expression green vegetation is more intensive, and it is better to grow；Enteromorpha frond is in fresh
Green, containing chlorophyll a and b, the spectral characteristic of Enteromorpha and seawater is had differences, Enteromorpha blue wave band and red band have compared with
Strong absorbability, reflectivity is low, has stronger reflection peak in green band, reflectivity is slightly higher, has near infrared band very strong
Reflection peak, reflectivity are very high；The reflectivity changes of seawater do not have that Enteromorpha is obvious, and entire reflectance curve is on a declining curve, green
Color wave band, seawater equally have stronger reflection peak with Enteromorpha, and the reflection peak of seawater is than more gentle, near infrared band, seawater
Reflectivity is very low, and the curve of spectrum of seawater and Enteromorpha has apparent difference, can be calculated simultaneously based on vegetation index
Determine green tide range；It is that vegetation index is calculated the result is that black white image, black are oceans, white is land
And Enteromorpha, a wide range of of seawater and Enteromorpha can be seen clearly, need to cut irregular green tide range image, extract green tide information and quantization
It calculates green tide region and obtains the green tide gross area；Satellite remote sensing images include the non-green tide area of land and cloud, utilize conventional threshold values side
Method and Variation Model extract green tide information can not automatic identification green tide area and non-green tide area, the normalization difference of land and Enteromorpha plants
Result is close after being calculated by index, and land picture dot participates in image segmentation, causes green tide information extraction result inaccurate；Regular image
There is no operational use value in reality, needs to carry out the cutting of irregular image.
The principle that green tide information extracting method of the present invention is quantified is: when using regular image, level set
Prospect in function segmentation result image is 1, and background is 0；When using irregular image, level set function segmentation result image
Interior prospect is 0, and background is 1, and the region outside image is entirely 0；Prospect represents green tide, background pp seawater.
Compared with prior art, the present invention accurately being extracted simultaneously on satellite remote sensing images based on the Variation Model of image segmentation
Quantify green tide information, traditional artificial threshold method can be substituted completely, realizes operational use；Its scientific in principle is reasonable, people
Few for factor, calculating speed is fast, as a result precise and stable, strong operability, and application environment is friendly, and practicability is good, easy to spread.
Specific embodiment:
Below by embodiment, the invention will be further described.
Embodiment 1:
The green tide information extracting method that the present embodiment is related to is completed by means of computer and satellite remote sensing images, technique packet
Include screening and preprocessed data source, green tide range is calculated and determined, cuts irregular green tide range image, extract green tide information and
Five, quantum chemical method green tide region step:
(1), screening with preprocessed data source: it is fine, partly cloudy, without rain and without the weather of typhoon, according to conventional side
Method obtains sea area satellite remote sensing images, and screening noise is few, clarity is high, brightness uniformity and image resolution ratio are less than or equal to 30 meters simultaneously
And the image in green tide region can be covered, and geometric correction and the pretreatment of image mosaic behaviour are carried out to the image data filtered out
Make, completes the screening and pretreatment of data source；
(2), green tide range is calculated and determined:
For the biological nature and spectral characteristic of Enteromorpha, calculated using vegetation index (NDVI index) and true
Determine green tide range, formula is as follows:
IR indicates the near infrared band in satellite remote sensing images in formula, selects 4 wave band datas, and R is red wave band, selects 3 waves
Segment data, the NDVI index for completing satellite remote sensing images calculate；
(3), it cuts irregular green tide range image: determining green tide range according to NDVI index calculation method, cut any
Shape region of interest and the pixel total number and the gross area for recording region of interest generate region of interest document data set, distant
Feel the irregular image file that processing platform imports generation region containing green tide after the region of interest document data set generated, completes not
The cutting of regular green tide range image；
(4), green tide information is extracted:
Division Donald Bragg using old-Wei Si (Chan-Vese) model based on variation level set two-phase image segmentation is graceful
(Bregman) fast projection algorithm extracts green tide information, and Chan-Vese model energy functional is as follows:
In formula (1-2),
R(u1, u2)=α1(u1-f)2-α2(u2-f)2 (1-4)
Wherein u1And u2The foreground and background of image is respectively indicated, f indicates the original image of Noise, introduces auxiliary division
Variable enablesBecause with equivalence, by formula (1-2) and formula (1-3) difference
It is rewritten as formula (1-5) and (1-6):
In formula (1-5),
For the punishment parameter that Bregman iterative parameter, θ are positive,WithInitial value is 0.
Formula (1-5) and formula (1-6) are solved to obtain:
In formula (1-9), g is edge indicator function, formula are as follows:
In formula (1-10), GσThe gaussian kernel function for being σ for standard deviation, sets initial parameter as γ=1, and σ=1, θ=
3000, it brings formula (1-8) into and formula (1-9) is solved to obtain green tide information extraction as a result, completing the extraction of green tide information；
(5), quantum chemical method green tide region: the green tide information quantization that step (4) is extracted using quantitative formula, quantitative formula
Are as follows:
Last_area=(clip_pixel-last_sum)/clip_pixel × clip_area (1-11)
In formula (1-11), Last_area is the gross area for the green tide information extracted, and clip_pixel is irregular image
Picture element total number, last_sum are the background pixels point total number in irregular image, and clip_area is irregular image
The gross area, unit are square kilometre.It brings given data into, obtains the green tide information gross area, unit is square kilometre to complete green tide
The quantum chemical method of information.
The NDVI index that the present embodiment is related to is to determine that the coverage condition of green vegetation mainly uses in satellite remote sensing field
Mode, NDVI index is higher, and expression green vegetation is more intensive, grow it is better；Since the biological nature of Enteromorpha is that Enteromorpha frond is in
Emerald green, as the terrestrial plants such as moss and dimension pipe, Enteromorpha contains chlorophyll a and b, the spectral characteristic presence of Enteromorpha and seawater
Notable difference, the spectral characteristic of Enteromorpha are that Enteromorpha in blue wave band (436nm) and red band (670nm) has stronger absorption
Property, it is low in the two wave band reflectivity, there is stronger reflection peak in green band (546nm), it is slightly higher in this wave band reflectivity,
There is very strong reflection peak near infrared band (750nm-850nm), it is very high in this wave band reflectivity；The spectral characteristic of seawater is
Its reflectivity changes does not have that Enteromorpha is obvious, and entire reflectance curve is on a declining curve, and in green band, seawater equally has with Enteromorpha
Stronger reflection peak, the reflection peak of seawater is very low in the reflectivity of near infrared band, seawater than more gentle, seawater and Enteromorpha
The curve of spectrum has apparent difference.Thus it is possible to which green tide range is calculated and determined based on NDVI index；NDVI index is calculated
The result is that black white image, black is ocean, and white is land and Enteromorpha, can see a wide range of of seawater and Enteromorpha clearly, need to cut out
It cuts irregular green tide range image, extraction green tide information and quantum chemical method green tide region and obtains the green tide gross area；Satellite remote sensing figure
As including land and Yun Dengfei green tide area, using conventional threshold values method and Variation Model extract green tide information can not automatic identification it is green
Tidal zone and non-green tide area, because result is close after land is calculated with the NDVI index of Enteromorpha, land picture dot participates in image segmentation, makes
At green tide information extraction result inaccuracy；Regular image does not have operational use value in reality, it is therefore desirable to not advised
The then cutting of image.
The quantization principles that the present embodiment is related to are: when using regular image, in level set function segmentation result image
Prospect is 1, and background is 0；When using irregular image, the prospect in level set function segmentation result image is 0, and background is 1,
Region outside image is entirely 0；Prospect represents green tide, background pp seawater.
Embodiment 2:
The green tide information extracting method step that the present embodiment is related to is with embodiment 1, and obtained green tide information is using biography
The artificial cognition of system and the threshold method of setting show that the data parameters of green tide information are shown in Table 1；Based on Variation Model and quantization side
Method show that the data parameters of green tide information are shown in Table 2, and the time in Tables 1 and 2 is from starting to carry out green tide information extraction to obtaining
The working time of the green tide information gross area, the gross area extracted based on Variation Model and quantization method is according to the difference of initial parameter
It is slightly different, wherein α value influences operational efficiency, is crucial initial parameter, and analytical table 1 is obtained when NDVI threshold value is -0.17
When the green tide information gross area (51.0678) that extracts be it is most accurate, in table 2 except 1,5 and 6 the green tide information gross area with
51.0678 have outside deviation, and other green tide information gross areas and 51.0678 are substantially close in extraction green tide information gross area number
According to upper, the result based on Variation Model and quantization method is more stable, and precision is higher；Analytical table 1 show that runing time is shortest
2 ' 10 ", longest is 2 ' 40 ", it is 14.4301 that runing time is shortest in table 2 ", and longest is 41.0595 ", at runtime between
On, the calculating speed based on Variation Model and quantization method is faster.
Table 1: the data parameters of conventional threshold values method extraction green tide information
| Serial number | NDVI threshold value | The gross area (km2) | Time |
| 1 | -0.07 | 30.3012 | 2′30″ |
| 2 | -0.08 | 30.8826 | 2′20″ |
| 3 | -0.09 | 32.7393 | 2′25″ |
| 4 | -0.1 | 34.8390 | 2′20″ |
| 5 | -0.11 | 35.6310 | 2′15″ |
| 6 | -0.12 | 38.5047 | 2′30″ |
| 7 | -0.13 | 40.7160 | 2′30″ |
| 8 | -0.14 | 42.1542 | 2′30″ |
| 9 | -0.15 | 44.0622 | 2′10″ |
| 10 | -0.16 | 47.5317 | 2′40″ |
Table 2: the data parameters of green tide information are extracted based on Variation Model and quantization method
| Serial number | Method | α | The number of iterations | The gross area (km2) | Time |
| 1 | SBPM | 1 | 6 | 54.5173 | 14.4301″ |
| 2 | SBPM | 1 | 7 | 50.4127 | 14.8513″ |
| 3 | SBPM | 1 | 8 | 49.1825 | 15.4750″ |
| 4 | SBPM | 1 | 9 | 48.7244 | 17.0041″ |
| 5 | SBPM | 0.1 | 33 | 59.7911 | 37.0970″ |
| 6 | SBPM | 0.1 | 34 | 55.3782 | 38.5166″ |
| 7 | SBPM | 0.1 | 35 | 52.3157 | 38.3294″ |
| 8 | SBPM | 0.1 | 36 | 50.3048 | 41.0595″ |
Claims (3)
1. a kind of green tide information extracting method, it is characterised in that technique include screening with preprocessed data source, be calculated and determined it is green
Damp range cuts irregular green tide range image, extracts green tide information and five, quantum chemical method green tide region step:
(1), screening with preprocessed data source: it is fine, partly cloudy, without rain and without the weather of typhoon, conventionally obtain
Sea area satellite remote sensing images are taken, screening noise is few, clarity is high, brightness uniformity and image resolution ratio are less than or equal to 30 meters and energy
The image for enough covering green tide region carries out the pretreatment operation of geometric correction and image mosaic to the image data filtered out, complete
At the screening and pretreatment of data source；
(2), green tide range is calculated and determined: for the biological nature and spectral characteristic of Enteromorpha, being referred to using normalization difference vegetation
Green tide range is calculated and determined in number, and formula is as follows:
IR indicates the near infrared band in satellite remote sensing images in formula, selects 4 wave band datas, and R is red wave band, selects 3 wave band numbers
According to the vegetation index for completing satellite remote sensing images calculates；
(3), it cuts irregular green tide range image: determining green tide range according to vegetation index calculation method, cut out
It cuts arbitrary shape region of interest and records the pixel total number and the gross area of region of interest, generate region of interest data set text
Part generates the irregular image text in the region containing green tide after remote sensing processing platform imports the region of interest document data set generated
Part completes the cutting of irregular green tide range image；
(4), green tide information is extracted:
It is extracted using the graceful fast projection algorithm of the division Donald Bragg of the old-Wei Si model based on variation level set two-phase image segmentation
Green tide information, old-Wei Si model energy functional are as follows:
Wherein u1And u2The foreground and background of image is respectively indicated, f indicates the original image of Noise, introduces auxiliary division variableIt enablesBecauseWithEquivalence changes formula (1-2) and formula (1-3) respectively
It is written as formula (1-5) and formula (1-6):
In formula (1-5),
For the graceful iterative parameter of Donald Bragg, the punishment parameter that θ is positive,WithInitial value is 0,
Formula (1-5) and formula (1-6) are solved to obtain:
In formula (1-9), g is edge indicator function, formula are as follows:
In formula (1-10), GσThe gaussian kernel function for being σ for standard deviation sets initial parameter as γ=1, and σ=1, θ=3000 are brought into
Formula (1-8) and formula (1-9) are solved to obtain green tide information extraction as a result, completing the extraction of green tide information；
(5), quantum chemical method green tide region: the green tide information quantization that step (4) is extracted using quantitative formula, quantitative formula are as follows:
Last_area=(clip_pixel-last_sum)/clip_pixel × clip_area (1-11)
In formula (1-11), Last_area is the gross area for the green tide information extracted, and clip_pixel is the pixel of irregular image
Point total number, last_sum are the background pixels point total number in irregular image, and clip_area is total face of irregular image
Product, unit are square kilometre to bring given data into, obtain the green tide information gross area, and unit is square kilometre to complete green tide information
Quantum chemical method.
2. green tide information extracting method according to claim 1, it is characterised in that the vegetation index is
Satellite remote sensing field determine the coverage condition of green vegetation mainly by the way of, the higher expression of vegetation index
Green vegetation is more intensive, and it is better to grow；Enteromorpha frond is in emerald green, containing chlorophyll a and b, the spectral characteristic of Enteromorpha and seawater
It has differences, Enteromorpha has stronger absorbability in blue wave band and red band, and reflectivity is low, has in green band stronger anti-
Peak is penetrated, reflectivity is slightly higher, has very strong reflection peak near infrared band, reflectivity is very high；The reflectivity changes of seawater do not have waterside
Tongue fur is obvious, and entire reflectance curve is on a declining curve, and in green band, seawater equally has stronger reflection peak, seawater with Enteromorpha
Reflection peak than more gentle, very low in the reflectivity of near infrared band, seawater, the curve of spectrum of seawater and Enteromorpha has apparent area
Not, green tide range can be calculated and determined based on vegetation index；What vegetation index was calculated
The result is that black white image, black is ocean, and white is land and Enteromorpha, can see a wide range of of seawater and Enteromorpha clearly, need to cut
Irregular green tide range image, extraction green tide information and quantum chemical method green tide region obtain the green tide gross area.
3. green tide information extracting method according to claim 1, it is characterised in that the principle quantified is: being advised when using
Then when image, the prospect in level set function segmentation result image is 1, and background is 0；When using irregular image, level set
Prospect in function segmentation result image is 0, and background is 1, and the region outside image is entirely 0；Prospect represents green tide, background pp
Seawater.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510800816.7A CN105427305B (en) | 2015-11-19 | 2015-11-19 | A kind of green tide information extracting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510800816.7A CN105427305B (en) | 2015-11-19 | 2015-11-19 | A kind of green tide information extracting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105427305A CN105427305A (en) | 2016-03-23 |
| CN105427305B true CN105427305B (en) | 2018-12-21 |
Family
ID=55505486
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510800816.7A Active CN105427305B (en) | 2015-11-19 | 2015-11-19 | A kind of green tide information extracting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105427305B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12118781B2 (en) * | 2020-10-23 | 2024-10-15 | The Second Institute of Oceanography (SIO), MNR | Method and device for determining extraction model of green tide coverage ratio based on mixed pixels |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108253943B (en) * | 2017-12-24 | 2020-08-21 | 航天恒星科技有限公司 | Integrated monitoring method for enteromorpha in red tide based on satellite remote sensing image |
| CN109781626B (en) * | 2019-03-11 | 2021-07-06 | 王祥 | Near-shore high-suspended sand water body green tide remote sensing identification method based on spectral analysis |
| CN110543863B (en) * | 2019-07-04 | 2023-02-03 | 自然资源部第一海洋研究所 | Green tide remote sensing automatic detection method and system based on neighborhood edge-preserving level set |
| CN111027766B (en) * | 2019-12-09 | 2023-09-26 | 国家海洋局北海预报中心 | Green tide biomass forecasting method, device, equipment and medium |
| CN112308901B (en) * | 2020-10-28 | 2022-11-08 | 山东省科学院海洋仪器仪表研究所 | Method for estimating green tide coverage area of sea surface under MODIS image cloud |
| CN112712553B (en) * | 2020-12-30 | 2022-09-02 | 自然资源部第一海洋研究所 | Enteromorpha shore resistance amount estimation method |
| CN113484923A (en) * | 2021-07-13 | 2021-10-08 | 山东省海洋预报减灾中心 | Remote sensing monitoring and evaluating method for green tide disasters |
| CN116467565B (en) * | 2023-06-20 | 2023-09-22 | 国家海洋局北海预报中心((国家海洋局青岛海洋预报台)(国家海洋局青岛海洋环境监测中心站)) | Enteromorpha green tide plaque optimal search area forecasting method |
| CN117408534B (en) * | 2023-12-14 | 2024-04-30 | 国家海洋局北海预报中心((国家海洋局青岛海洋预报台)(国家海洋局青岛海洋环境监测中心站)) | Enteromorpha green tide salvaging effect short-term evaluation method based on satellite remote sensing |
| CN118155080B (en) * | 2024-05-10 | 2024-08-02 | 国家海洋局北海预报中心((国家海洋局青岛海洋预报台)(国家海洋局青岛海洋环境监测中心站)) | Enteromorpha coverage area prediction method based on exponential regression model |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102208100A (en) * | 2011-05-31 | 2011-10-05 | 重庆大学 | Total-variation (TV) regularized image blind restoration method based on Split Bregman iteration |
| CN102231190A (en) * | 2011-07-08 | 2011-11-02 | 中铁第四勘察设计院集团有限公司 | Automatic extraction method for alluvial-proluvial fan information |
| CN102254174A (en) * | 2011-07-08 | 2011-11-23 | 中铁第四勘察设计院集团有限公司 | Method for automatically extracting information of bare area in slumped mass |
| CN103604761A (en) * | 2013-10-29 | 2014-02-26 | 国家海洋局第一海洋研究所 | Red tide detection method based on AISA aerial hyperspectral image |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20150117127A (en) * | 2014-04-09 | 2015-10-19 | 한국원자력연구원 | Method alarming quality of water using image information of water |
-
2015
- 2015-11-19 CN CN201510800816.7A patent/CN105427305B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102208100A (en) * | 2011-05-31 | 2011-10-05 | 重庆大学 | Total-variation (TV) regularized image blind restoration method based on Split Bregman iteration |
| CN102231190A (en) * | 2011-07-08 | 2011-11-02 | 中铁第四勘察设计院集团有限公司 | Automatic extraction method for alluvial-proluvial fan information |
| CN102254174A (en) * | 2011-07-08 | 2011-11-23 | 中铁第四勘察设计院集团有限公司 | Method for automatically extracting information of bare area in slumped mass |
| CN103604761A (en) * | 2013-10-29 | 2014-02-26 | 国家海洋局第一海洋研究所 | Red tide detection method based on AISA aerial hyperspectral image |
Non-Patent Citations (3)
| Title |
|---|
| "基于小波框架的盲图像修复研究";刘纯利等;《计算机科学》;20130430;第40卷(第4期);第295-297页 * |
| "基于混合像元分解的MODIS绿潮覆盖面积精细化提取方法研究";辛蕾等;《激光生物学报》;20141231;第23卷(第6期);第585-589页 * |
| "多范数混合约束的正则化图像盲复原";李伟红等;《光学精密工程》;20130531;第21卷(第5期);第1357-1364页 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12118781B2 (en) * | 2020-10-23 | 2024-10-15 | The Second Institute of Oceanography (SIO), MNR | Method and device for determining extraction model of green tide coverage ratio based on mixed pixels |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105427305A (en) | 2016-03-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105427305B (en) | A kind of green tide information extracting method | |
| Ren et al. | Rapid expansion of coastal aquaculture ponds in China from Landsat observations during 1984–2016 | |
| Zhang et al. | A novel index for robust and large-scale mapping of plastic greenhouse from Sentinel-2 images | |
| CN104457626B (en) | A kind of plant leaf area assessment of indices method based on laser radar point cloud | |
| CN103063202B (en) | Cyanobacteria biomass spatial-temporal change monitoring and visualization method based on remote sensing image | |
| CN110378909A (en) | Single wooden dividing method towards laser point cloud based on Faster R-CNN | |
| CN107504923B (en) | Kelp culture area monitoring method integrating remote sensing image and extension rope information | |
| Sverdrup-Thygeson et al. | Can airborne laser scanning assist in mapping and monitoring natural forests? | |
| CN110414509A (en) | Stop Ship Detection in harbour based on the segmentation of extra large land and feature pyramid network | |
| CN105893977B (en) | A kind of rice drafting method based on adaptive features select | |
| Jia et al. | Monitoring loss and recovery of salt marshes in the Liao River Delta, China | |
| CN116883853B (en) | Crop space-time information remote sensing classification method based on transfer learning | |
| Kamal et al. | Comparison of Google Earth Engine (GEE)-based machine learning classifiers for mangrove mapping | |
| CN117975282B (en) | Remote sensing extraction method and system for laver cultivation area based on multi-element optical feature fusion | |
| Kawakubo et al. | Mapping coffee crops in southeastern Brazil using spectral mixture analysis and data mining classification | |
| CN114119618B (en) | Inland salt lake artemia strip remote sensing extraction method based on deep learning | |
| CN114119617B (en) | Inland salt lake artemia zone extraction method of multispectral satellite remote sensing image | |
| CN117611987B (en) | Automatic identification method, device and medium for sea for cultivation | |
| Xue et al. | Detection the expansion of marine aquaculture in Sansha Bay by remote sensing | |
| Cui et al. | Remote sensing identification of marine floating raft aquaculture area based on sentinel-2A and DEM data | |
| Pada et al. | Mangrove forest cover extraction of the coastal areas of Negros Occidental, Western Visayas, Philippines using LiDAR data | |
| Zhang et al. | A Mapping Approach for Eucalyptus Plantations Canopy and Single-Tree Using High-Resolution Satellite Images in Liuzhou, China | |
| CN109871774A (en) | A kind of mixed pixel decomposition method based on the close pixel of local | |
| Liao et al. | Study on mangrove of maximum likelihood: Reclassification method in Xiezhou bay | |
| Pineda et al. | Concentration and condition of American lobster postlarvae in small‐scale convergences |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |