CN114998658A - Intertidal zone beach extraction method and system based on tidal flat index - Google Patents
Intertidal zone beach extraction method and system based on tidal flat index Download PDFInfo
- Publication number
- CN114998658A CN114998658A CN202210690325.1A CN202210690325A CN114998658A CN 114998658 A CN114998658 A CN 114998658A CN 202210690325 A CN202210690325 A CN 202210690325A CN 114998658 A CN114998658 A CN 114998658A
- Authority
- CN
- China
- Prior art keywords
- tide
- synthetic
- synthetic image
- image
- tidal flat
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a tidal flat index-based intertidal zone beach extraction method and system, wherein the method comprises the following steps: acquiring a remote sensing image of a target area in a target period; generating a low tide synthetic image and a high tide synthetic image by adopting a quantile synthetic method based on the remote sensing image; calculating a tidal flat index corresponding to each pixel position by utilizing the low tide and high tide synthetic images; and carrying out object-oriented multi-scale segmentation by using the computation result of the tidal flat index to obtain a patch object, and then classifying the patch object by using a classification model to obtain the distribution result of intertidal zone tidal flat of the target area. According to the technical scheme, the low tide synthetic image and the high tide synthetic image are generated firstly, and the spectral reflectivity of the low tide and the high tide is brought into the calculation of the tidal flat index, so that the technical obstacle brought to the remote sensing extraction of the intertidal zone mudflat by the periodic flooding of tidal water is overcome, and the mudflat extraction precision is effectively improved.
Description
Technical Field
The invention belongs to the technical field of classification extraction of remote sensing images, and particularly relates to a tidal flat index-based intertidal zone tidal flat extraction method and system.
Background
The tidal flat is located in an intertidal zone area where the sea transits to the land environment, is an important component for connecting the sea, the land and a fresh water ecosystem, and plays an important ecological and economic role in maintaining the stability of a coastal zone, preventing wind and fixing sand, keeping biological diversity, developing human activities and the like. However, due to the influence of climate change and human activities, such as sea level rise, land reclamation for urbanization, coastal over-development, aquaculture expansion and the like, intertidal zone tidal flat area is less and less, which leads to serious damage to coastal wetland ecosystem and rapid decline of biodiversity. Therefore, under the severe situation that the coastal zone beaches are increasingly damaged, the intertidal zone beaches space distribution is rapidly and accurately extracted, the area and the distribution of the beaches are systematically mastered, and a data basis can be provided for scientific management, protection and ecological restoration of the coastal wetland, so that the method is very important for sustainable development of the coastal zone.
In recent years, the remote sensing technology has the advantages of accuracy, real time, multiple scales, short repeated workshop period and the like, and is widely applied to extraction and monitoring research of intertidal zones and mudflats. The conventional intertidal zone mudflat extraction method can be classified into three types: tidal model or terrain based methods, machine learning classification methods, knowledge based decision tree classification methods. The method based on the tidal model or the terrain has the technical defects of uncertainty, difficulty in obtaining early multi-source data and complex implementation process due to complex dynamic changes of the terrain and the tidal water level space of the coastal zone; the machine learning classification method needs to rely on a large amount of sample training, but the cost selection is high in labor intensity, high in cost and lack of repeatability; the decision tree classification method based on knowledge needs to determine threshold conditions according to experience, however, unified thresholds are not suitable for extraction of large-area mudflats, and the technical defects of more manual intervention and lack of automation exist.
Secondly, because tidal periodically submerges to bring great difficulty to tidal flat extraction, in the existing tidal flat extraction method based on remote sensing images, single-time-phase images are mostly adopted and depend on auxiliary data such as tidal water level, terrain information, training samples and the like aiming at data sources; in the identification method, the links such as the threshold value and the post-processing of manual intervention are mostly relied on. However, an index specially used for extracting the tidal flat is not provided according to the characteristic that the tidal flat is periodically submerged, and a set of tidal flat extraction method is formed by taking the constructed tidal flat index as a data source, so that the extraction result of the tidal flat in an intertidal zone is not accurate enough, and the extraction method has the characteristics of universality, intelligence and reproducibility.
Disclosure of Invention
The invention aims to overcome the problem of inaccurate intertidal zone mudflat extraction result in the prior art, improve the intertidal zone mudflat extraction precision, and especially fill the gap of the mudflat extraction technology considering the periodic submergence of tide in the prior art. The intertidal zone beach extraction method generates a low tide synthetic image and a high tide synthetic image, and simultaneously brings the spectral reflectivity of low tide and high tide into the calculation of the tidal zone index, thereby overcoming the difficulty brought by the periodic submergence of tidal water to the remote sensing extraction of the intertidal zone beach, solving the problems of excessive dependence on manual intervention and difficult acquisition of auxiliary data, and effectively improving the beach extraction precision.
On one hand, the intertidal zone beach extraction method based on the tidal flat index provided by the invention comprises the following steps:
step 1: acquiring a remote sensing image of a target area in a target period;
step 2: generating a low tide synthetic image and a high tide synthetic image by adopting a quantile synthetic method based on the remote sensing image;
and step 3: calculating a tidal flat index corresponding to each pixel position by utilizing the low tide and high tide synthetic images;
and 4, step 4: carrying out object-oriented multi-scale segmentation by utilizing the computation result of the tidal flat index to obtain a plaque object, and then classifying the plaque object by utilizing a classification model to obtain the distribution result of intertidal zones of a target area. Wherein, the distribution result of the intertidal zone tidal flat is the intertidal zone tidal flat of the target area extracted by the technical scheme of the invention.
And 4, setting the classification model related to the step 4 as a remote sensing classification method such as an expert decision tree, a neural network, a support vector machine, a nearest algorithm and the like, and training samples of different land classes to obtain the classification model trained by the remote sensing classification method. The input of the classification model is the characteristics of the plaque, and the output result is the type of the plaque; or the input of the classification model is the characteristics of the patch, and the output result is the classification result of whether the patch is the intertidal zone tidal flat or not. In addition, the training sample of the classification model is preferably a sample generated by the same type of remote sensing image.
Further optionally, the tidal flat index is calculated as follows:
wherein TFRI is the tidal flat index, Bandi low Is the spectral reflectance of the i-th band, Bandj, at a certain pixel position in the low tide synthetic image high And Bandk high The spectral reflectivities of the j wave band and the k wave band at the same pixel position in the climax synthetic image are respectively;
the ith wave band is a characteristic wave band for distinguishing the mudflat from other ground object types based on the low tide synthetic image; the j & ltth & gt and k & ltth & gt wave bands are characteristic wave bands for distinguishing the mudflat from other ground object types based on the climax synthetic images.
Further optionally, the ith wavelength band is an 8 th wavelength band, the jth wavelength band is a 6 th wavelength band, and the kth wavelength band is a 7 th wavelength band.
Further optionally, the i-th waveband, the j-th waveband and the k-th waveband are distinct characteristic wavebands extracted by a DNA coding method, and the specific process is as follows:
respectively collecting samples of known ground object types on the generated low tide synthetic image and the high tide synthetic image;
respectively calculating the spectral reflectivity mean value and/or the first order differential spectral reflectivity mean value of all ground object types on the low tide synthetic image and the high tide synthetic image based on the respective spectral reflectivity curve of the low tide synthetic image and the high tide synthetic image and/or based on the first order differential spectral reflectivity curve of the spectral reflectivity curve, and taking the spectral reflectivity mean value and/or the first order differential spectral reflectivity mean value as the middle threshold value T corresponding to the low tide synthetic image and the high tide synthetic image mid ；
Respectively obtaining the spectral reflectivity curve of the low tide and the high tide synthetic images and/or the spectral reflectivity curve of all pixels on the first order differential spectral reflectivity curve and/or the minimum value T of the first order differential spectral reflectivity min Maximum value T max And based on said minimum value T of the same curve min The maximum value T max And the intermediate threshold value T mid Form an interval [ T min ,T mid ]And [ T mid ,T max ](ii) a That is, the spectral reflectance curve and/or the first order differential spectral reflectance curve of the low tide synthetic image respectively correspond to a set of intervals [ T min ,T mid ]And [ T mid ,T max ](ii) a The spectral reflectance curve and/or first order differential spectral reflectance curve of the climax synthetic image respectively correspond to a set of intervals [ T min ,T mid ]And [ T mid ,T max ]；
Then, the spectral reflectivity curve and/or the first order differential spectral reflectivity curve of the low tide and high tide synthetic images respectively falling on the corresponding two intervals are averaged to obtain the corresponding 1/4 average threshold value T 1/4 Average threshold T with 3/4 3/4 Further, the interval [ T ] corresponding to the low tide and the high tide is formed min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ]；
Wherein the interval [ T min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ]Respectively corresponding to four character codes;
then, according to the spectral reflectivity and/or first order differential spectral reflectivity corresponding to each wave band of the pixel of each ground feature category on the low tide and high tide synthetic image and the interval [ T min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ]According to the wave band sequence, coding to form a DNA code chain of each ground object type corresponding to the low tide and high tide synthetic images;
wherein, the DNA code chain of each ground object type is formed by the codes of all wave bands;
and finally, comparing DNA code chains generated on the low tide and high tide synthetic images by each ground category, and respectively selecting the code segment with the largest difference relative to the mudflat, wherein the obtained code segments corresponding to the low tide and the high tide are the distinguishing characteristic wave segments corresponding to the low tide and the high tide.
Further optionally, if encoding is performed based on the spectral reflectance curve and the first-order differential spectral reflectance curve, a reflectance code chain is obtained based on the spectral reflectance curve, a waveform code chain is obtained based on the first-order differential spectral reflectance curve, and then the reflectance code chain and the waveform code chain of the same feature type are combined to obtain a DNA code chain.
Further optionally, the step 2 of generating the low tide synthetic image and the high tide synthetic image by using a quantile synthesis method based on the remote sensing image comprises the following steps:
step 21: counting the spectral reflectivity corresponding to each pixel in each wave band in the remote sensing image in the target period, and further generating a histogram corresponding to each pixel in each wave band;
step 22: respectively determining the spectral reflectances of which the cumulative ratio reaches a preset low ratio and a preset high ratio aiming at the histogram corresponding to each pixel in each wave band, and then calculating the spectral reflectance mean value of which the cumulative ratio is less than or equal to the preset low ratio and the spectral reflectance mean value of which the cumulative ratio is greater than or equal to the preset high ratio;
step 23: and generating a low tide synthetic image by using the spectral reflectance mean value of each pixel corresponding to the lower ratio or equal to the preset low ratio in all the wave bands, and generating a high tide synthetic image by using the spectral reflectance mean value of each pixel corresponding to the higher ratio or equal to the preset high ratio in all the wave bands.
Further optionally, the remote sensing image is a Sentinel-2 image.
In a second aspect, the invention provides a system based on the intertidal zone extraction method, which comprises:
the remote sensing image acquisition module is used for acquiring a remote sensing image of a target area in a target period;
the low tide and high tide synthetic image generating module is used for generating a low tide synthetic image and a high tide synthetic image by adopting a quantile synthetic method based on the remote sensing image;
the tidal flat index calculation module is used for calculating the tidal flat index corresponding to each pixel position by utilizing the low tide and high tide synthetic images;
and the intertidal zone beach extraction module is used for performing object-oriented multi-scale segmentation by using the computation result of the tidal flat index to obtain a plaque object, and classifying the plaque object by using a classification model to obtain the distribution result of the intertidal zone beach.
In a third aspect, the present invention provides an electronic terminal, comprising:
one or more processors;
and a memory storing one or more computer programs;
the processor calls the computer program to perform:
a step of intertidal zone mudflat extraction method based on high and low tide synthetic images.
In a fourth aspect, the present invention provides a readable storage medium storing a computer program for execution by a processor to:
a step of intertidal zone mudflat extraction method based on high and low tide synthetic images.
Advantageous effects
1. The invention provides a tidal zone mudflat extraction method based on a tidal zone index, aiming at the problem that the difficulty of extracting the mudflat by using a remote sensing technology is high due to the influence of tide fluctuation on the tidal zone, the remote sensing image in a target period is fully utilized to synthesize a low tide synthetic image and a high tide synthetic image, and meanwhile, the characteristics of the low tide synthetic image and the high tide synthetic image are taken into the calculation of the tidal zone index, so that the calculated tidal zone index has the characteristics of low tide and high tide, the influence of periodical flood of the tide is effectively overcome, the blank of the mudflat extraction technology considering the periodical flood of the tide is filled, the mudflat extraction precision is effectively improved, the technical scheme of the invention is easy to realize, the supplement and perfection of the existing mudflat extraction method is realized, and the distribution of the intertidal zone mudflat can be extracted quickly, simply and accurately.
2. In a further preferred scheme of the invention, low tide and high tide characteristics are integrated from tide fluctuation characteristics in the field, a brand new tidal flat index calculation formula is provided, and the tidal flat index calculated by the formula can be used for more effectively and accurately distinguishing the types of the tidal flats and other ground objects, so that the final precision of the tidal flat extraction is improved. According to the tidal flat index in the preferred scheme, the difference of the spectral reflectivity between the tidal flat and other background ground objects is enhanced through square operation, and the spectral reflectivity of the background ground objects such as a water body is inhibited through product operation, so that the spectral reflectivity characteristic of the tidal flat is highlighted, and the remarkable difference between the tidal flat and other ground object types is increased.
3. In a further preferred scheme of the invention, in order to obtain the distinguishing characteristic wave bands for identifying the tidal flat and other ground object types, a DNA coding theory is creatively applied to the field for the first time, an improved DNA coding method suitable for the field is provided, and the selection precision of the distinguishing characteristic wave bands is effectively improved, so that the reliability and the accuracy of the tidal flat index are finally improved, and a foundation is laid for accurately extracting the intertidal zone tidal flat.
Drawings
FIG. 1 is a Sentinel-2 image at high and low tide levels, wherein (a) is a high tide synthetic image and (b) is a low tide synthetic image;
FIG. 2 is a graph of the mean spectral reflectance of a culture pond, background water, buildings, land vegetation, and mudflats in a Sentinel-2 image at high and low tide levels;
FIG. 3 is a schematic representation of the results of the intertidal zone beach index (TFRI) calculation for the study area;
FIG. 4 is a box diagram statistically generated from the results of the beach index (TFRI) calculations for different land features;
FIG. 5 is the final identification of intertidal zone mudflat in the research area;
fig. 6 is a schematic flow chart of a intertidal zone mudflat extraction method based on a high-low tide synthetic image according to an embodiment of the present invention.
Detailed Description
The invention aims to improve the beach extraction precision and overcome the technical obstacle brought to beach extraction by the periodical flood of tides, thereby providing an intertidal zone beach extraction method based on a high and low tide synthetic image, filling the blank of the beach extraction technology considering the periodical flood of tides, and further explaining the invention by combining with an embodiment.
Example 1:
the embodiment provides a tidal flat index-based intertidal zone tidal flat extraction method, which comprises the following steps:
step 1: and acquiring a remote sensing image of a target area in a target period.
In this embodiment, the target period is set to one year, and the remote sensing image is a Sentinel-2 image. Selecting all the Sentinel-2 images in one year to form a Sentinel-2 image set, and then preprocessing each Sentinel-2 image in the Sentinel-2 image set, wherein the preprocessing technology comprises radiometric calibration, atmospheric correction and orthometric correction. Since the preprocessing is a conventional technology, it is not specifically stated, and different preprocessing means can be selected according to actual needs and precision requirements in different embodiments, which are not specifically limited by the present invention. It should also be understood that in other possible embodiments, the target period may be adaptively adjusted according to actual requirements, and other remote sensing images suitable for the technical idea of the present invention also belong to the protection scope of the present invention.
Step 2: and generating a low tide synthetic image and a high tide synthetic image by adopting a quantile synthetic method based on the remote sensing image. In this example, the quantile synthesis method was chosen to generate the low-tide and high-tide Sentinel-2 synthetic images. The specific process is as follows:
step 21: and (3) counting the spectral reflectivity corresponding to each pixel in each wave band in the Sentinel-2 image set, and further generating a histogram corresponding to each pixel in each wave band.
Wherein, the Sentinel-2 image comprises 12 bands, and each pixel corresponds to a spectral reflectance in each band. In this embodiment, spectral reflectivity is counted for each pixel (same pixel position) in each band of the Sentinel-2 image set, and the spectral reflectivities are arranged from small to large according to the magnitude of the values, so that a histogram is generated from the queued values.
Step 22: and respectively determining the spectral reflectances of which the cumulative ratio reaches a preset low ratio and a preset high ratio aiming at the histogram corresponding to each pixel in each wave band, and then calculating the spectral reflectivity mean value of which the cumulative ratio is less than or equal to the preset low ratio and the spectral reflectivity mean value of which the cumulative ratio is greater than or equal to the preset high ratio.
Step 23: and generating a low tide synthetic image by using the spectral reflectance mean value of each pixel corresponding to the lower ratio or equal to the preset low ratio in all the wave bands, and generating a high tide synthetic image by using the spectral reflectance mean value of each pixel corresponding to the higher ratio or equal to the preset high ratio in all the wave bands.
In the embodiment, the preset low-to-high ratio is set to 10%, and the preset high-to-low ratio is set to 90%; therefore, in the embodiment, for the histogram corresponding to each band of each pixel, the positions of the spectral reflectances with the cumulative percentage of 10% and 90% are determined and marked, and then the average value of the spectral reflectances less than or equal to 10% and greater than or equal to 90% in the histogram is calculated. The calculation is repeated for 12 wave bands corresponding to each pixel, and finally, the mean value of the spectral reflectivity, which is obtained by correspondingly less than or equal to 10 percent of each pixel in the 12 wave bands, is utilized to synthesize a low tide Sentinel-2 synthetic image; and synthesizing the high-tide Sentiniel-2 synthetic image by using the average value of the spectral reflectivity of each pixel, which is obtained within 12 wave bands and is greater than or equal to 90%.
It should be understood that in other possible embodiments, the preset low-duty ratio and the preset high-duty ratio can be adaptively adjusted according to the application requirements and the precision requirement.
And step 3: and calculating the tidal flat index corresponding to each pixel by using the low tide and high tide synthetic images.
The preferred tidal flat index in this embodiment is calculated as follows:
in the formula, TFRI is the tidal flat index, and is the spectral reflectance of the i-th band on the low tide synthetic image, and is the spectral reflectance of the j-th band and the k-th band on the high tide synthetic image, respectively.
The tidal flat index of the Sentinel-2 image enhances the difference of spectral reflectances between the tidal flat and other background ground objects through square operation, and inhibits the spectral reflectances of the background ground objects such as a water body and the like through product operation, thereby highlighting the spectral reflectivity characteristics of the tidal flat.
In this embodiment, the ith wavelength band is the 8 th wavelength band, the jth wavelength band is the 6 th wavelength band, and the kth wavelength band is the 7 th wavelength band. The bands 6(0.740), 7(0.783) and 8(0.842) have separability, the near-infrared band of the band 8(0.842) is sensitive to water, the bands 6(0.740) and 7(0.783) are red-edge bands, mudflats can be extracted according to the spectral reflectivity difference, and the tidal flat index TFRI (intertidal zone mudflat extraction index) of the Sentinel-2 image is constructed according to the selected reflectivity characteristic band. In the embodiment, the band 9 is an inflection point (an upward or downward trend inflection point) of many surface feature categories, if the inflection point band 9 is selected, a great uncertainty is brought, and the effect is not as good as that of the above band, so the above setting of the embodiment is an optimal value, and in other possible embodiments, the above calculation can be performed according to the following method. In this embodiment, in order to obtain more accurate distinctive characteristic bands, a DNA encoding method is preferably adopted for extraction, and specific implementation processes will be highlighted below. In other possible embodiments, other existing methods can be adopted, and in short, the identification criterion for distinguishing the characteristic wave bands is the wave band with the most significant difference between the mudflat and the data of other ground object types.
And 4, step 4: carrying out object-oriented multi-scale segmentation by utilizing the computation result of the tidal flat index to obtain a plaque object, and then classifying the plaque object by utilizing a classification model to obtain the distribution result of intertidal zones of a target area.
When the tidal flat index is used for segmentation, the tidal flat and other ground object types can be effectively separated, and therefore more accurate tidal flat patches can be obtained. However, the technical means for performing object-oriented multi-scale segmentation based on the mudflat extraction index is the prior art, and therefore, no specific statement is made on the technical means. In this embodiment, the classification model is set as a remote sensing classification method such as an expert decision tree, a neural network, a support vector machine, a nearest neighbor algorithm, and the like, and is trained by using training samples of different ground feature types and adopting the remote sensing classification method. The input of the classification model is the characteristics of the plaque, and the output result is the type of the plaque; or the input of the classification model is the feature of the plaque, the output result is the classification result of whether the plaque is intertidal zone tidal flat, and the selected classification algorithm and the input feature refer to the prior art, so that the specific statement is not made on the classification algorithm and the input feature. In addition, the training samples of the classification model are preferably samples generated by the same type of remote sensing images, manual marking or other high-precision marking algorithms can be adopted to determine beach plaques and other non-beach plaques on the samples, and when different requirements (lowering) are made on the model precision, the sources of the training samples of the classification model can be adjusted adaptively.
According to the embodiment, the method can effectively improve the beach extraction precision, and obtain a more accurate intertidal zone beach distribution result.
Regarding the selection of the distinctive characteristic bands, the DNA coding method is preferably adopted for extraction, and the specific implementation process is as follows:
s1: and respectively collecting samples of the known ground object types on the generated low tide synthetic image and the high tide synthetic image.
Among them, the ground feature type samples on the same type of remote sensing images with the same target period length are preferable. The surface feature types in this embodiment include: culture ponds, background water, buildings, land vegetation and mudflats.
S2: respectively calculating the spectral reflectance mean value and/or the first order differential spectral reflectance mean value of all ground object types on the low tide synthetic image and the high tide synthetic image based on the respective spectral reflectance curves of the low tide synthetic image and the high tide synthetic image and/or based on the first order differential spectral reflectance curve of the spectral reflectance curve, and taking the mean value as the corresponding intermediate threshold value T of the low tide synthetic image and the high tide synthetic image mid 。
In this embodiment, it is preferable to select a typical sample point from each feature type, form a mean spectral reflectance curve and a first-order differential spectral reflectance curve based on the mean spectral reflectance curve of the feature type on the low-tide synthesized image, and perform subsequent processing on both the mean spectral reflectance curve and the first-order differential spectral reflectance curve according to a method that obtains two intermediate thresholds T of the spectral reflectance curve and the first-order differential spectral reflectance curve pair for the low-tide synthesized image mid (ii) a Obtaining two intermediate threshold values T of spectral reflectivity curve and first-order differential spectral reflectivity curve pair aiming at climax synthetic image mid 。
In other possible embodiments, selecting a typical sample point or selecting all image points of each surface feature type to calculate a mean spectral reflectance curve of each surface feature type or calculate a first-order differential spectral reflectance curve of each surface feature type is a technical means protected by the present invention. In other feasible embodiments, mean value calculation is not selected, a spectral emissivity curve and a first-order differential spectral reflectivity curve of a typical sample point in each ground object type are generated by taking an image point as a unit, each curve is subjected to subsequent processing according to the following method, and finally, final results are obtained in a comprehensive mode, and only the calculation data amount is greatly increased.
In other possible embodiments, the subsequent process may be performed to identify the distinctive bands by using only the mean spectral reflectance curve of the above feature types on the low tide synthetic image or the first order differential spectral reflectance curve based on the mean spectral reflectance curve, which is weaker than the combination scheme of the two types of curves in this embodiment.
S3: respectively obtaining the spectral reflectivity curve of the low tide and the high tide synthetic images and/or the spectral reflectivity curve of all pixels on the first order differential spectral reflectivity curve and/or the minimum value T of the first order differential spectral reflectivity min Maximum value T max And based on said minimum value T min The maximum value T max And the intermediate threshold value T mid Form a region [ T min ,T mid ]And [ T mid ,T max ]。
In this embodiment, a set of intervals [ T ] is obtained for the mean spectral reflectance curve and the first-order differential spectral reflectance curve of the low tide min ,T mid ]And [ T mid ,T max ](ii) a Respectively obtaining a group of intervals [ T ] aiming at the mean spectral reflectivity curve and the first-order differential spectral reflectivity curve of climax min ,T mid ]And [ T mid ,T max ]。
Other possible embodiments, the corresponding processing is performed according to the curve generated in the previous step S2 to obtain a respective set of intervals T min ,T mid ]And [ T mid ,T max ]。
S4: then, the spectral reflectivity curve and/or the first order differential spectral reflectivity curve of the low tide and high tide synthetic images respectively falling into the corresponding two intervals are averaged to obtain 1/4 average threshold values T corresponding to the spectral reflectivity curve and/or the first order differential spectral reflectivity curve respectively 1/4 Average threshold T with 3/4 3/4 Further form the interval [ T ] corresponding to the low tide and the high tide min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ]；
Similarly, in this embodiment, a group of intervals [ T ] is obtained for the mean spectral reflectance curve and the first-order differential spectral reflectance curve of the low tide min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ](ii) a Respectively obtaining a group of intervals [ T ] aiming at the mean spectral reflectivity curve and the first-order differential spectral reflectivity curve of climax min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ]。
Other possible embodiments are to perform the corresponding processing according to the curve generated in the previous step S2 to obtain a respective set of intervals T min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ]。
S5: according to the spectral reflectivity and/or first-order differential spectral reflectivity corresponding to each wave band of the pixel of each ground feature type on the low tide and high tide synthetic image and the interval [ T min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ]According to the wave band sequence, coding to form a DNA code chain of each ground object type corresponding to the low tide and high tide synthetic images; wherein, the DNA code chain of each ground object type is composed of codes of all wave bands.
Wherein the interval [ T min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ]Corresponding to four character codes, e.g. four intervals [ T ] min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ]The four intervals correspond to four coding modes of A, T, C and G respectively.
In this embodiment, the data and corresponding interval [ T ] of the mean spectral reflectance curve and the first order differential spectral reflectance curve for the low tide min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ]According to the wave band sequence, coding is carried out to form a reflectivity code chain and a waveform code chain of each ground feature type corresponding to the climax, and then the reflectivity code chain and the waveform of the same ground feature type are codedCombining the code chains to obtain the low-tide DNA code chain of the ground object type; data and corresponding interval [ T ] for the mean spectral reflectance curve and first order differential spectral reflectance curve of climax min ,T 1/4 )、[T 1/4 ,T mid )、[T mid ,T 3/4 ) And [ T 3/4 ,T max ]According to the relation of (1), coding is carried out according to the wave band sequence to form a reflectivity code chain and a waveform code chain of each ground object type corresponding to the climax, and then the reflectivity code chain and the waveform code chain of the same ground object type are combined to obtain a DNA code chain of the climax of the ground object type.
In other possible embodiments, if a plurality of mean spectral reflectance curves and/or first-order differential spectral reflectance curves exist for each feature type in the low-tide and high-tide composite images, a plurality of DNA code chains of low tide and DNA code chains of high tide are obtained correspondingly.
S6: and comparing DNA code chains generated on the low tide and high tide synthetic images by each ground category, and respectively selecting the code segment with the largest difference relative to the mudflat, wherein the obtained code segments corresponding to the low tide and the high tide are the distinguishing characteristic wave segments corresponding to the low tide and the high tide.
In this embodiment, the difference between DNA code chains of mudflat and other ground object categories is analyzed according to the principle of similarity within the largest categories and difference between the largest categories, and the most different band is selected.
It should be understood that the present embodiment applies the DNA encoding method to the art for the first time to determine the distinctive characteristic bands, compared to the conventional extraction methods, such as the envelope method, the characteristic parameter method, and the differential method, which all form a spectrum curve with certain characteristics, visually identify the characteristic bands, and then perform the band extraction, with subjective and speculative colors. And by adopting a DNA coding method, the characteristic wave band can be quantized and extracted, the trend of the whole waveform can be researched and judged on the whole through a DNA coding chain, the sensitivity of detecting changes caused by slight differences in the wave band is realized, the construction of the beach index based on the extracted wave band in the later period is facilitated, and the classification result of the tidal flat is more reliable.
When the difference characteristic Band is obtained, calculating the intertidal zone beach index corresponding to the sample by using a Band Math tool in the ENVI software, as shown in FIG. 3; and then quantitatively counting the mudflat index calculation results of the sample points in the culture pond (pond), the background water body, the building, the land vegetation and the mudflat, generating a box-type diagram as shown in figure 4, judging the separability of the mudflat and other ground objects, and if the separability meets the requirement, determining that the obtained distinguishing characteristic wave band meets the requirement.
According to the invention, the northern gulf sea area of Zhanjiang province of Guangdong province is selected as a research area, and all the Sentinel-2 images of the research area of 2021 year are selected to form a Sentinel-2 image set in total 249 scenes. The images were processed by fractional synthesis to obtain the low and high tide Sentinel-2 synthetic images shown in FIG. 1. Secondly, selecting typical sample points from the culture pond (pond), the background water body, the buildings, the land vegetation and the mud flat, and analyzing the mean spectral reflectance curves of the ground objects in the low tide and high tide Sentinel-2 synthetic images as shown in figure 2 respectively. Finally, the intertidal zone mudflat distribution diagram shown in figure 5 is obtained.
Example 2:
the embodiment provides a system based on the intertidal zone beach extraction method, which comprises the following steps: the tidal flat index extraction system comprises a remote sensing image acquisition module, a low tide and high tide synthetic image generation module, a distinctive characteristic waveband acquisition module, a tidal flat index calculation module and an intertidal zone beach extraction module.
The remote sensing image acquisition module is used for acquiring a remote sensing image of a target area in a target period.
And the low tide and high tide synthetic image generating module is used for generating a low tide synthetic image and a high tide synthetic image by adopting a quantile synthetic method based on the remote sensing image. The technical details of the quantile synthesis method can be referred to the corresponding contents of the above methods.
And the distinguishing characteristic band acquisition module is used for acquiring distinguishing characteristic bands corresponding to the low tide synthetic image and the high tide synthetic image. In particular, the above-mentioned contents may be referred to, as the DNA coding method or other conventional methods may be used.
And the tidal flat index calculation module is used for calculating the tidal flat index corresponding to each pixel position by utilizing the low tide and high tide synthetic images. The specific formula is described in the above.
And the intertidal zone beach extraction module is used for performing object-oriented multi-scale segmentation by using the computation result of the tidal flat index to obtain a plaque object, and classifying the plaque object by using a classification model to obtain the distribution result of the intertidal zone beach. The classification model is a classification model directly quoted from the existing classification model or a classification model trained again, and if the classification model is retrained, the corresponding system further comprises a classification model building module used for building the classification model.
For the implementation process of each module, please refer to the content of the above method, which is not described herein again. It should be understood that the above described division of functional blocks is merely a division of logical functions and that in actual implementation, there may be other divisions, for example, where multiple units or components may be combined or integrated into another system or where some features may be omitted or not implemented. Meanwhile, the integrated unit can be realized in a hardware form, and can also be realized in a software functional unit form.
Example 3:
the present embodiments provide an electronic terminal comprising one or more processors and a memory storing one or more computer programs, wherein the processors invoke the computer programs to perform: a tidal flat index-based intertidal zone beach extraction method. Specifically, the method comprises the following steps:
step 1: acquiring a remote sensing image of a target area in a target period;
step 2: generating a low tide synthetic image and a high tide synthetic image by adopting a quantile synthetic method based on the remote sensing image;
and step 3: calculating the tidal flat index corresponding to each pixel by utilizing the low tide and high tide synthetic images;
and 4, step 4: carrying out object-oriented multi-scale segmentation by utilizing the computation result of the tidal flat index to obtain a plaque object, and then classifying the plaque object by utilizing a classification model to obtain the distribution result of intertidal zones of a target area.
In some implementations, steps 21-23 are also performed when the processor invokes the computer program to implement step 2.
The memory may include high speed RAM memory, and may also include a non-volatile defibrillator, such as at least one disk memory.
If the memory and the processor are implemented independently, the memory, the processor and the communication interface may be connected to each other via a bus and perform communication with each other. The bus may be an industry standard architecture bus, a peripheral device interconnect bus, an extended industry standard architecture bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc.
Optionally, in a specific implementation, if the memory and the processor are integrated on a chip, the memory and the processor may complete communication with each other through an internal interface.
The specific implementation process of each step refers to the explanation of the foregoing method.
It should be understood that in the embodiments of the present invention, the Processor may be a Central Processing Unit (CPU), and the Processor may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The memory may include both read-only memory and random access memory, and provides instructions and data to the processor. The portion of memory may also include non-volatile random access memory. For example, the memory may also store device type information.
Example 4:
the present embodiments provide a readable storage medium storing a computer program for being invoked by a processor to perform: a tidal flat index-based intertidal zone beach extraction method. Specifically, the method comprises the following steps:
step 1: acquiring a remote sensing image of a target area in a target period;
step 2: generating a low tide synthetic image and a high tide synthetic image by adopting a quantile synthetic method based on the remote sensing image;
and step 3: calculating the tidal flat index corresponding to each pixel by utilizing the low tide and high tide synthetic images;
and 4, step 4: carrying out object-oriented multi-scale segmentation by utilizing the computation result of the tidal flat index to obtain a plaque object, and then classifying the plaque object by utilizing a classification model to obtain the distribution result of intertidal zones of a target area.
In some implementations, steps 21-23 are also performed when the processor invokes the computer program to implement step 2.
The specific implementation process of each step refers to the explanation of the foregoing method.
The readable storage medium is a computer readable storage medium, which may be an internal storage unit of the controller according to any of the foregoing embodiments, for example, a hard disk or a memory of the controller. The readable storage medium may also be an external storage device of the controller, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, provided on the controller. Further, the readable storage medium may also include both an internal storage unit of the controller and an external storage device. The readable storage medium is used for storing the computer program and other programs and data required by the controller. The readable storage medium may also be used to temporarily store data that has been output or is to be output.
Based on such understanding, the technical solution of the present invention essentially or partially contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned readable storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
It should be emphasized that the examples described herein are illustrative and not restrictive, and thus the invention is not to be limited to the examples described herein, but rather to other embodiments that may be devised by those skilled in the art based on the teachings herein, and that various modifications, alterations, and substitutions are possible without departing from the spirit and scope of the present invention.
Claims (10)
1. A tidal flat index-based intertidal zone tidal flat extraction method is characterized by comprising the following steps: the method comprises the following steps:
step 1: acquiring a remote sensing image of a target area in a target period;
step 2: generating a low tide synthetic image and a high tide synthetic image by adopting a quantile synthetic method based on the remote sensing image;
and step 3: calculating a tidal flat index corresponding to each pixel position by utilizing the low tide and high tide synthetic images;
and 4, step 4: carrying out object-oriented multi-scale segmentation by utilizing the computation result of the tidal flat index to obtain a plaque object, and then classifying the plaque object by utilizing a classification model to obtain the distribution result of intertidal zones of a target area.
2. The intertidal zone extraction method as claimed in claim 1, characterized in that: the tidal flat index is calculated according to the following formula:
wherein TFRI is the tidal flat index, Bandi low Is the spectral reflectance of the i-th band, Bandj, at a certain pixel position in the low tide synthetic image high And Bandk high The spectra of the j-th wave band and the k-th wave band at the same pixel position in the climax synthetic imageA reflectivity;
the ith wave band is a characteristic wave band for identifying the difference between the tidal flat and other ground object types based on the low tide synthetic image; the j & ltth & gt and k & ltth & gt wave bands are characteristic wave bands for distinguishing the mudflat from other ground object types based on the climax synthetic images.
3. The intertidal zone beach extraction method of claim 2, which is characterized in that: the ith waveband is an 8 th waveband, the jth waveband is a 6 th waveband, and the kth waveband is a 7 th waveband.
4. The intertidal zone beach extraction method of claim 2, which is characterized in that: the ith wave band, the jth wave band and the kth wave band are all distinguished characteristic wave bands extracted by adopting a DNA coding method, and the specific process is as follows:
respectively collecting samples of known ground object types on the generated low tide synthetic image and the high tide synthetic image;
respectively calculating the spectral reflectance mean value and/or the first order differential spectral reflectance mean value of all ground object types on the low tide synthetic image and the high tide synthetic image based on the respective spectral reflectance curves of the low tide synthetic image and the high tide synthetic image and/or based on the first order differential spectral reflectance curve of the spectral reflectance curve, and taking the mean value as the corresponding intermediate threshold value T of the low tide synthetic image and the high tide synthetic image mid ；
Respectively obtaining the spectral reflectivity curve of the low tide and the high tide synthetic images and/or the spectral reflectivity curve of all pixels on the first order differential spectral reflectivity curve and/or the minimum value T of the first order differential spectral reflectivity min Maximum value T max And based on said minimum value T of the same curve min The maximum value T max And the intermediate threshold value T mid Form respective set of intervals [ T min ，T mid ]And [ T mid ，T max ]；
And respectively falling the spectral reflectivity curve and/or the first-order differential spectral reflectivity curve of the low tide and high tide synthetic images into corresponding intervals [ T min ，T mid ]And [ T mid ，T max ]The spectral reflectivity and/or the first order differential spectral reflectivity are averaged to obtain 1/4 average threshold values T corresponding to each 1/4 And 3/4 average threshold value T 3/4 Further, the corresponding interval [ T ] of the low tide and the high tide is formed min ，T 1/4 )、[T 1/4 ，T mid )、[T mid ，T 3/4 ) And [ T 3/4 ，T max ]；
Wherein the interval [ T min ，T 1/4 )、[T 1/4 ，T mid )、[T mid ，T 3/4 ) And [ T 3/4 ，T max ]Respectively corresponding to four character codes;
then, according to the spectral reflectivity and/or first order differential spectral reflectivity corresponding to each wave band of the pixel of each ground feature category on the low tide and high tide synthetic image and the interval [ T min ，T 1/4 )、[T 1/4 ，T mid )、[T mid ，T 3/4 ) And [ T 3/4 ，T max ]According to the wave band sequence, coding to form a DNA code chain of each ground object type corresponding to the low tide and high tide synthetic images;
wherein, the DNA code chain of each ground object type is formed by the codes of all wave bands;
and finally, comparing DNA code chains generated on the low tide and high tide synthetic images by each ground category, and respectively selecting the code segment with the largest difference relative to the mudflat, wherein the obtained code segments corresponding to the low tide and the high tide are the distinguishing characteristic wave segments corresponding to the low tide and the high tide.
5. The intertidal zone beach extraction method of claim 4, which is characterized in that: if coding is carried out based on the spectral reflectivity curve and the first-order differential spectral reflectivity curve, a reflectivity code chain is obtained based on the spectral reflectivity curve, a waveform code chain is obtained based on the first-order differential spectral reflectivity curve, and then the reflectivity code chain and the waveform code chain of the same ground object type are combined to obtain a DNA code chain.
6. The intertidal zone beach extraction method of claim 1, which is characterized in that: in step 2, the process of generating the low tide synthetic image and the high tide synthetic image by adopting a quantile synthetic method based on the remote sensing image is as follows:
step 21: counting the spectral reflectivity corresponding to each pixel in each wave band in the remote sensing image in the target period, and further generating a histogram corresponding to each pixel in each wave band;
step 22: respectively determining the spectral reflectances of which the cumulative ratio reaches a preset low ratio and a preset high ratio aiming at the histogram corresponding to each pixel in each wave band, and then calculating the spectral reflectance mean value of which the cumulative ratio is less than or equal to the preset low ratio and the spectral reflectance mean value of which the cumulative ratio is greater than or equal to the preset high ratio;
step 23: and generating a low tide synthetic image by using the spectral reflectance mean value of each pixel corresponding to the lower ratio or equal to the preset low ratio in all the wave bands, and generating a high tide synthetic image by using the spectral reflectance mean value of each pixel corresponding to the higher ratio or equal to the preset high ratio in all the wave bands.
7. The intertidal zone beach extraction method of claim 1, which is characterized in that: the remote sensing image is a Sentinel-2 image.
8. A system based on the intertidal zone beach extraction method of any one of claims 1 to 6, which is characterized in that: the method comprises the following steps:
the remote sensing image acquisition module is used for acquiring a remote sensing image of a target area in a target period;
the low tide and high tide synthetic image generating module is used for generating a low tide synthetic image and a high tide synthetic image by adopting a quantile synthetic method based on the remote sensing image;
the tidal flat index calculation module is used for calculating the tidal flat index corresponding to each pixel position by utilizing the low tide and high tide synthetic images;
and the intertidal zone beach extraction module is used for performing object-oriented multi-scale segmentation by using the computation result of the tidal flat index to obtain a plaque object, and classifying the plaque object by using a classification model to obtain the distribution result of the intertidal zone beach.
9. An electronic terminal, characterized by: the method comprises the following steps:
one or more processors;
and a memory storing one or more computer programs;
the processor calls the computer program to perform:
the steps of the intertidal zone extraction process of any one of claims 1 to 7.
10. A readable storage medium, characterized by: a computer program is stored, which is invoked by a processor to perform:
the method for intertidal zone extraction of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210690325.1A CN114998658A (en) | 2022-06-17 | 2022-06-17 | Intertidal zone beach extraction method and system based on tidal flat index |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210690325.1A CN114998658A (en) | 2022-06-17 | 2022-06-17 | Intertidal zone beach extraction method and system based on tidal flat index |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114998658A true CN114998658A (en) | 2022-09-02 |
Family
ID=83034655
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210690325.1A Pending CN114998658A (en) | 2022-06-17 | 2022-06-17 | Intertidal zone beach extraction method and system based on tidal flat index |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114998658A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115471761A (en) * | 2022-10-31 | 2022-12-13 | 宁波拾烨智能科技有限公司 | Coastal beach change monitoring method integrating multi-source remote sensing data |
| CN117036777A (en) * | 2023-07-04 | 2023-11-10 | 宁波大学 | Mud flat extraction method based on hyperspectral data |
-
2022
- 2022-06-17 CN CN202210690325.1A patent/CN114998658A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115471761A (en) * | 2022-10-31 | 2022-12-13 | 宁波拾烨智能科技有限公司 | Coastal beach change monitoring method integrating multi-source remote sensing data |
| CN115471761B (en) * | 2022-10-31 | 2023-03-24 | 宁波拾烨智能科技有限公司 | Coastal beach change monitoring method integrating multi-source remote sensing data |
| CN117036777A (en) * | 2023-07-04 | 2023-11-10 | 宁波大学 | Mud flat extraction method based on hyperspectral data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11836976B2 (en) | Method for recognizing seawater polluted area based on high-resolution remote sensing image and device | |
| Ruiz-Ruiz et al. | Testing different color spaces based on hue for the environmentally adaptive segmentation algorithm (EASA) | |
| CN114998658A (en) | Intertidal zone beach extraction method and system based on tidal flat index | |
| Sadeghi-Tehran et al. | Multi-feature machine learning model for automatic segmentation of green fractional vegetation cover for high-throughput field phenotyping | |
| Chen et al. | The application of the tasseled cap transformation and feature knowledge for the extraction of coastline information from remote sensing images | |
| CN109740631B (en) | OBIA-SVM-CNN remote sensing image classification method based on object | |
| CN107247927B (en) | Method and system for extracting coastline information of remote sensing image based on tassel cap transformation | |
| CN111008664B (en) | Hyperspectral sea ice detection method based on space-spectrum combined characteristics | |
| CN111007013B (en) | Crop rotation fallow remote sensing monitoring method and device for northeast cold region | |
| CN110889840A (en) | Effectiveness detection method of high-resolution 6 # remote sensing satellite data for ground object target | |
| CN112991351A (en) | Remote sensing image semantic segmentation method and device and storage medium | |
| CN116559111A (en) | Sorghum variety identification method based on hyperspectral imaging technology | |
| CN112488983A (en) | Defect identification network obtaining method, defect identification method and grade determining method | |
| CN117115669B (en) | Object-level ground object sample self-adaptive generation method and system with double-condition quality constraint | |
| CN111882573B (en) | Cultivated land block extraction method and system based on high-resolution image data | |
| CN117392157A (en) | Edge-aware protective cultivation straw coverage rate detection method | |
| Zhang et al. | Water quality parameters retrieval of coastal mariculture ponds based on UAV multispectral remote sensing | |
| CN116863341A (en) | Crop classification and identification method and system based on time sequence satellite remote sensing image | |
| CN116895019A (en) | Remote sensing image change detection method and system based on dynamic weighted cross entropy loss | |
| CN116452872A (en) | Forest scene tree classification method based on improved deep pavv3+ | |
| Mukhtar | Land use land cover change detection through remote sensing approach in Kano State Nigeria | |
| CN115311566A (en) | Cloud mask image generation method and pre-training network training method | |
| Nammalwar et al. | A generic framework for colour texture segmentation | |
| Indukuri et al. | Paddy Disease Classifier using Deep learning Techniques | |
| CN112800826A (en) | City change monitoring method and system based on remote sensing image classification |
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 |