CN116258785A - Ocean environment remote sensing visual expression method based on numerical feature statistics - Google Patents

Ocean environment remote sensing visual expression method based on numerical feature statistics Download PDF

Info

Publication number
CN116258785A
CN116258785A CN202310113256.2A CN202310113256A CN116258785A CN 116258785 A CN116258785 A CN 116258785A CN 202310113256 A CN202310113256 A CN 202310113256A CN 116258785 A CN116258785 A CN 116258785A
Authority
CN
China
Prior art keywords
data
value
values
pixel
numerical
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
Application number
CN202310113256.2A
Other languages
Chinese (zh)
Inventor
赵彬如
王子珂
李艳雯
焦红波
牛思文
任晓明
张峰
邢喆
谷祥辉
杨晓彤
赵现仁
郭丽
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NATIONAL MARINE DATA AND INFORMATION SERVICE
Original Assignee
NATIONAL MARINE DATA AND INFORMATION SERVICE
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by NATIONAL MARINE DATA AND INFORMATION SERVICE filed Critical NATIONAL MARINE DATA AND INFORMATION SERVICE
Priority to CN202310113256.2A priority Critical patent/CN116258785A/en
Publication of CN116258785A publication Critical patent/CN116258785A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/001—Texturing; Colouring; Generation of texture or colour
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T11/00—2D [Two Dimensional] image generation
    • G06T11/40—Filling a planar surface by adding surface attributes, e.g. colour or texture
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G06T3/4007—Scaling of whole images or parts thereof, e.g. expanding or contracting based on interpolation, e.g. bilinear interpolation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G06T3/403—Edge-driven scaling; Edge-based scaling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Multimedia (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Algebra (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Image Processing (AREA)

Abstract

The invention provides a marine environment remote sensing visual expression method based on numerical feature statistics, which comprises the following steps: the data preprocessing module is used for carrying out noise reduction processing on the data so as to improve the data resolution and weaken the edge sawtooth phenomenon; the data grouping and characteristic statistics module is used for obtaining data grouping and carrying out characteristic statistics through cluster analysis and characteristic statistics aiming at cluster phenomena existing in numerical distribution of marine environment remote sensing inversion raster data; and the group interval optimization and color mapping module establishes an index relation between the set color and the numerical value interval through the set numerical value interval to perform color mapping and optimize the visual display effect. The invention has the beneficial effects that: a visual expression method of marine environment remote sensing based on numerical feature statistics aims at the problem that inverted raster data inevitably contains noise and errors, a data mode and characteristics of the data mode are easy to hide, and a data preprocessing module is utilized to improve data resolution and weaken edge sawtooth phenomenon.

Description

Ocean environment remote sensing visual expression method based on numerical feature statistics
Technical Field
The invention belongs to the field of marine environment remote sensing, and particularly relates to a marine environment remote sensing visual expression method based on numerical feature statistics.
Background
The marine remote sensing inversion data is quantitative inversion of marine environment and phenomenon information, and has the characteristics of obvious difference of marine and land environment properties, large scale of a common coverage area of a marine satellite and the like, and the information expression of the satellite remote sensing inversion data has difference with land. At present, marine environment data is mainly stored in a grid data form, and the conventional marine element visual rendering method has single-value rendering, grading rendering, stretching rendering and the like, and is subject to the problems of reasonable grading of color bands due to the characteristics of complex distribution rule, uneven information distribution and the like of marine environment information in a large-scale area. According to the conventional method, the ocean time sequence elements are visually expressed by adopting fixed grades, so that pixel values are often excessively concentrated in individual cells, the resolution effect of a high-color concentration area is seriously affected, the problems of small visual contrast, large discrete degree of individual data, serious edge sawtooth phenomenon and the like during rendering are caused, and the image forming information expression capability and visual attractiveness are further affected.
At present, the analysis work of the visualization of the marine inversion data is not paid enough attention, and ESRI provides a method for increasing the visual contrast of the grid display by stretching, but on the principle of the method, the color bands are classified by uniform density segmentation, the data is not analyzed and specially processed, and the problem of poor visualization effect of the marine environment remote sensing inversion information is not solved substantially as a result of the method.
Disclosure of Invention
In view of the above, the present invention aims to provide a marine environment remote sensing visual expression method based on numerical feature statistics, so as to at least solve at least one problem in the background art.
In order to achieve the above purpose, the technical scheme of the invention is realized as follows:
a marine environment remote sensing visual expression method based on numerical feature statistics comprises the following steps:
the data preprocessing module is used for carrying out noise reduction processing on the data so as to improve the data resolution and weaken the edge sawtooth phenomenon;
the data grouping and characteristic statistics module is used for obtaining data grouping and carrying out characteristic statistics through cluster analysis and characteristic statistics aiming at cluster phenomena existing in numerical distribution of marine environment remote sensing inversion raster data;
and the group interval optimization and color mapping module establishes an index relation between the set color and the numerical value interval through the set numerical value interval to perform color mapping and optimize the visual display effect.
Further, in the data preprocessing module, data preprocessing is performed on marine environment remote sensing inversion raster data, abnormal values and abrupt values are removed, and conversion processing is performed on special abundance distribution type data, specifically as follows:
a1, outlier rejection: according to the element attribute, the region and the climatology characteristic, carrying out value range and landing inspection on the data, and eliminating abnormal values;
a2, removing mutation values: searching for isolated peaks in the data and removing the peaks;
a3, preprocessing special abundance distribution type data: aiming at the grid data visual expression of the large gradient change process elements facing the estuary/near shore-sea basin/ocean mixing area, the logarithmic preprocessing is firstly carried out to preliminarily and uniformly distribute the abundance.
Further, in step A2, the method specifically includes the following steps:
(1) Traversing to calculate average value D of abrupt change of lattice points and neighborhood lattice points ij ：
Figure BDA0004077698890000021
wherein ,gij For the target lattice point value g pq The number of the neighbor grid points is the number of the neighbor grid points, and n is the number of the neighbor grid points effective values;
(2) Calculating the spatial standard deviation sigma of the abrupt mean D ：
Figure BDA0004077698890000031
Wherein N is the number of valid values of raster data, r is the total number of rows of raster data, r is the total number of columns of raster data,
Figure BDA0004077698890000032
mean value of grid data mutation;
(3) Traversing the lattice point mutation mean value D ij And the space standard deviation sigma D Relationship, determination D ij >3σ D The lattice point of (2) is a mutation abnormal value, and an invalid value noDataValue is assigned to the mutation abnormal value to finish the elimination processing.
Further, in step A3, the method specifically includes the following steps:
(1) Calculating a raster data value range;
(2) Performing histogram analysis of not less than 10 groups of raster data, and performing logarithmic processing on the raster data when the cumulative group spacing of the first n maximum frequency groups with the cumulative frequency greater than 90% is less than 10% of the range of the value range;
(3) And storing the logarithmic processing mapping relation, and determining the color bar coordinates by using inverse mapping after finishing the subsequent hierarchical optimization flow.
Furthermore, in the data preprocessing module, interpolation algorithm is adopted for the discontinuous space or the lower resolution data to improve the density, supplement missing data, eliminate and weaken internal and boundary mosaic phenomena, and more smoothly and finely describe the real distribution rule of remote sensing elements of the marine environment, and the method is as follows:
b1, mask construction: overlapping the raster data with a land vector, setting the position of the non-DataValue of the land part as-1, setting the positions of the non-DataValue of other sea parts as 0, setting the positions of the other effective numerical values as 1, and acquiring an inversion raster data mask;
b2, hole filling: selectively filling the small-range internal missing value with the mask gray value of 0;
b3, internal interpolation: in order to prevent data distortion, the interpolation proportion of the internal data is set to be 0.5-3, and the internal continuous data is processed by using a DINOF or bilinear interpolation method according to the abundance of the time-space sequence of the data;
b4, edge interpolation: the bilinear interpolation cannot solve the problem of edge jaggies, and for edge data, the transition between the edge data and an internal effective value is natural.
Further, in step B2, the method specifically includes the following steps:
(1) Extracting connected domain S for internal missing value with mask gray value of 0 i I=1, 2,3, … …, m, m is the total number of connected domains, and is divided into two types of small and large according to the size of the connected domains, wherein the area of the connected domains is less than or equal to 5 pixels and is set as small connected domains, and the rest is large connected domains;
(2) Dividing each small connected domain into time sequence rule filling and adjacent value filling according to whether the time sequence data is the time sequence data, and reconstructing inversion information containing time sequence data by an empirical orthogonal function interpolation method; and for single inversion information which does not contain time sequence data, assigning values to the missing values by adopting a method of counting adjacent value characteristics, and sequentially assigning and filling according to the sequence from outside to inside, wherein the filling values are average values of all adjacent values, and filling processing is not carried out on each large communication domain.
Further, in step B4, the following steps are used to smooth the edges, weakening the mosaic phenomenon:
(1) Edge finding: traversing the four adjacent domains of the pixel with the mask value of 1, and if any pixel in the four adjacent domains is-1, the pixel is an edge pixel;
(2) Reconstructing an optimal edge broken line according to edge shape fitting, dividing each edge into long side and short side combinations according to the number of pixels, dividing the edges into Z types, U types and L types according to an MLAA algorithm, and reconstructing an optimal edge line;
(3) Assigning each non-DataValue pixel penetrated by the reconstructed edge folding line, wherein the pixel value of each pixel marked by the reconstructed edge line is the length ratio mixing result of the reconstructed edge line penetrating through the adjacent pixels, and the value of the pixel to be assigned is as follows:
Figure BDA0004077698890000041
wherein T is the number g of adjacent edge pixels of the pixel to be assigned i Is the value of the ith adjacent pixel, r i To reconstruct the length of the edge line across the ith edge pixel, r is the pixel length, g a And (5) assigning the value of the pixel to be assigned.
Further, in the data grouping and feature statistics module, the method specifically includes the following steps:
c1, determining the grouping number n, and grouping continuous raster data, wherein the data in the group has obvious clustering characteristics;
c2, carrying out intra-group feature statistics according to the grouping determined in the step C1;
c3, aiming at the steps C1 and C2, adjusting the group spacing;
wherein, in step C1, the number of packets should be determined according to the following method:
(1) Histogram statistics is carried out on the raster data, and a pixel value range [ g ] is obtained min ，g max ]The number of pixels corresponding to each value;
(3) Extracting a histogram contour line and smoothing;
(3) Searching a histogram peak value through a findpeaks function in matlab, setting the minimum interval number between two peak values to be 5, and obtaining the total number of peak values meeting the condition as the grouping number S;
in step C2, the specific steps are as follows:
value field [ g ] min ，g max ]Equally dividing into S groups, and counting the corresponding pixel number n in each group k (k∈[1,S]) Each group of corresponding value ranges is [ str ] k ,end k ]；
In step C3, the specific steps are as follows:
(1) From the data histogram, calculate [ str ] k ,end k ]The probability density sum corresponding to all values in the interval is used for obtaining the probability density P of the values in the group k ：
Figure BDA0004077698890000051
Wherein n is the total number of pixels, n k Is of the value [ str ] k ,end k ]All the pixel numbers in the interval;
(2) The degree of dispersion of the numerical space distribution is balanced by the specific weight of the covariance matrix norm of all the numerical space coordinates in the groupQuantity search [ str ] k ,end k ]Space coordinates corresponding to all values in the interval
Figure BDA0004077698890000052
Figure BDA0004077698890000061
Calculating the numerical distribution dispersion D in the group k The formula is as follows:
COV(X k ,Y k )＝E[(X k -E(X k ))(Y k -E(Y k ))]
Figure BDA0004077698890000062
/>
Figure BDA0004077698890000063
Figure BDA0004077698890000064
wherein ,
Figure BDA0004077698890000065
E(X k ) Is X k Desirably, E (Y k ) Is Y k Desirably, C is the covariance matrix; />
Figure BDA0004077698890000066
Singular values of C; f (F) k Covariance norms for the k-th group;
(3) Obtain each interval [ str ] k ,end k ]The number of pixel values and the next adjacent interval str k+1 ,end k+1 ]Gradient V of variation of pixel values of (2) k ：
Figure BDA0004077698890000067
Figure BDA0004077698890000068
wherein ,nk+1 Is of the value [ str ] k+1 ,end k+1 ]All the pixel numbers in the interval;
in step C4, the specific steps are as follows:
(1) According to the numerical characteristic parameters of each group counted in the step C3, forming characteristic weights A in each group k ：
Figure BDA0004077698890000069
(2) Optimizing the interval between each group by adjusting the characteristic weight, and obtaining the value range [ g ] min ，g max ]Readjusting according to the characteristic weights in the groups, and then the kth group spacing is as follows:
F k ＝(g max -g min )×A k 。
further, in the group interval optimization and color mapping module, the method specifically includes the following steps:
establishing a numerical value-color band mapping table: and selecting a color model for the determined optimal S group of numerical values, and performing data mapping according to the following data mapping formula:
IMG(x,y)→{[str 1 ,end 1 ],[str 2 ,end 2 ]……[str s ,end s ]}→ColorBar
wherein IMG is raster data, [ x, y ] is space abscissa and color bar is color bar.
Further, the scheme discloses a computer readable storage medium which stores a computer program, wherein the computer program realizes a marine environment remote sensing visual expression method based on numerical feature statistics when being executed by a processor.
Compared with the prior art, the marine environment remote sensing visual expression method based on numerical feature statistics has the following beneficial effects:
(1) According to the marine environment remote sensing visual expression method based on numerical feature statistics, aiming at the problems that inverted raster data inevitably contains noise and errors, a data mode and features of the data mode are easy to be hidden, a data preprocessing module is utilized to improve data resolution and weaken edge sawtooth phenomenon;
(2) According to the marine environment remote sensing visual expression method based on numerical feature statistics, the group spacing optimization and color mapping module is utilized to ensure that numerical probability density distribution of each grouping interval is uniform and reasonable, group differences are obvious, visual display effects are optimized, expression of marine environment phenomena is highlighted, and the problem that the visual effects of marine remote sensing inversion information raster data are not ideal is solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention. In the drawings:
FIG. 1 is a schematic diagram of an overall process flow according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a sequential assignment filling process according to an outside-in sequence in an interpolation calculation process according to an embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating an edge interpolation process according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating the creation of a numeric-to-color bar map according to an embodiment of the present invention.
Detailed Description
It should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other.
The invention will be described in detail below with reference to the drawings in connection with embodiments.
The invention aims at marine environment remote sensing inversion raster data, and mainly comprises the following modules:
(1) And (5) preprocessing data. Because the inverted raster data inevitably contains noise and errors, the data pattern and its features are easily hidden, and thus a preprocessing operation is required for the data. Firstly, eliminating abnormal values and invalid data and performing necessary conversion; then carrying out interpolation calculation on the data, wherein the interpolation calculation mainly comprises four steps of mask construction, hole filling, internal interpolation and edge interpolation, so that holes caused by some numerical value missing are eliminated, the data resolution is improved, and the edge sawtooth phenomenon is weakened;
(2) Data packet and feature statistics. The distribution of the numerical values of the marine environment remote sensing inversion raster data has a cluster phenomenon, and reasonable grouping can be obtained through cluster analysis and feature statistics. Firstly, carrying out statistical analysis on the histogram, determining a raster data clustering center, determining the grouping number, and then carrying out characteristic statistical analysis on the data in the group, wherein the characteristic statistical analysis mainly comprises probability density, distribution dispersion and variation gradient of numerical values.
(3) Group spacing optimization and color mapping. The marine environment remote sensing inversion raster data establishes an index relation between a set color and a numerical value interval by setting a reasonable numerical value interval so as to carry out color mapping. Firstly, determining weights of all groups through characteristic statistical values in the groups, optimizing group spacing, and then carrying out grouping rendering on grouping data according to designed color bands, so that the numerical probability density distribution of all grouping intervals is ensured to be uniform and reasonable, the difference among the groups is obvious, the visual display effect is optimized, the expression of marine environment phenomenon is highlighted, and the problem that the visual effect of the marine remote sensing inversion information raster data is not ideal is solved. The general flow is shown in figure 1.
The steps are as follows:
1. and (5) preprocessing data. And carrying out data preprocessing on the marine environment remote sensing inversion raster data, removing abnormal values and mutation values, and carrying out conversion processing on special abundance distribution type data.
1-1: outlier rejection: and (3) carrying out value range and landing inspection on the data according to the element attribute, the region and the climatology characteristic, and eliminating abnormal values.
1-2: mutation value elimination: searching for isolated peak values in data and removing the peak values
(1) Traversing to calculate average value D of abrupt change of lattice points and neighborhood lattice points ij ：
Figure BDA0004077698890000091
wherein ,gij For the target lattice point value g pq And n is the number of the neighbor lattice point effective values. (2) Calculating the spatial standard deviation sigma of the abrupt mean D ：
Figure BDA0004077698890000092
Wherein N is the number of valid values of raster data, r is the total number of rows of raster data, r is the total number of columns of raster data,
Figure BDA0004077698890000093
mean value of grid data mutation.
(3) Traversing the lattice point mutation mean value D ij And the space standard deviation sigma D Relationship, determination D ij >3σ D The lattice point of (2) is a mutation abnormal value, and an invalid value noDataValue is assigned to the mutation abnormal value to finish the elimination processing.
1-3: pretreatment of special abundance distribution type data: aiming at the grid data visual expression of large gradient change process elements (such as chlorophyll, suspended matters and the like) facing the estuary/coastal-sea basin/ocean mixing area, the logarithmic pretreatment is carried out firstly to preliminarily and uniformly distribute the abundance.
(1) Calculating a raster data value range;
(2) Performing histogram analysis of not less than 10 groups of raster data, and performing logarithmic processing on the raster data when the cumulative group spacing of the first n maximum frequency groups with the cumulative frequency greater than 90% is less than 10% of the range of the value range;
(3) And storing the logarithmic processing mapping relation, and determining the color bar coordinates by using inverse mapping after finishing the subsequent hierarchical optimization flow.
2. And (5) interpolation calculation. And an interpolation algorithm is adopted for the spatially discontinuous or lower resolution data to improve the density, supplement missing data, eliminate and weaken internal and boundary mosaic phenomena, and more smoothly and finely describe the real distribution rule of the remote sensing elements of the marine environment. Because the marine inversion remote sensing raster data volume is large, the traditional interpolation algorithm consumes a long time and cannot consider internal and edge data, the invention adopts the following method to carry out interpolation calculation on the data:
2-1: and (5) mask construction. Superposing the raster data with the land vector, and setting the NoDataValue position of the land part as
-1, the other sea noDataValue positions are set to 0 and the remaining valid value positions are set to 1, obtaining an inversion raster data mask.
2-2: and (5) filling the holes. And selectively filling the small-range internal missing value with the mask gray value of 0.
(1) Extracting connected domain S for internal missing value with mask gray value of 0 i I=1, 2,3, … …, m, m is the total number of connected domains. The connected domains are classified into small and large ones according to the size (the connected domain area is less than or equal to 5 pixels, which are small connected domains, and the rest are large connected domains).
(2) Dividing each small connected domain into time sequence rule filling and adjacent value filling according to whether the time sequence data is the time sequence data, and reconstructing inversion information containing the time sequence data by an empirical orthogonal function interpolation method (DINOF); for single inversion information which does not contain time series data, the missing values are assigned by adopting a method of counting the characteristics of the adjacent values, and filling is sequentially assigned according to the sequence from outside to inside (see figure 2), wherein the filling value is the average value of all the adjacent values. And (5) not performing filling processing on each large communication domain.
2-3: and (5) internal interpolation. For internal data, to prevent data distortion, the interpolation ratio is set to 0.5-3, and the internal continuous data is processed by using a DINOF or bilinear interpolation method according to the abundance of the data time-space sequence.
2-4: edge interpolation. The bilinear interpolation cannot solve the problem of edge jaggies, and for edge data, the transition between the edge data and an internal effective value is natural.
(1) Searching for the edge, traversing the four adjacent domains of the pixel with the mask value of 1, and if any pixel in the four adjacent domains is-1, determining the pixel as an edge pixel;
(2) Reconstructing an optimal edge broken line according to edge shape fitting, dividing each edge into long side and short side combinations according to the number of pixels, dividing the edges into Z type, U type and L type according to MLAA algorithm, and reconstructing an optimal edge line (see figure 3, wherein a white broken line is the reconstructed optimal edge line);
(3) Assigning each non-DataValue pixel penetrated by the reconstructed edge folding line, wherein the pixel value of each pixel marked by the reconstructed edge folding line is the length ratio mixing result (see figure 3) of the reconstructed edge line penetrating through adjacent pixels, and the value of the pixel to be assigned is as follows:
Figure BDA0004077698890000111
wherein T is the number g of adjacent edge pixels of the pixel to be assigned i Is the value of the ith adjacent pixel, r i To reconstruct the length of the edge line across the ith edge pixel, r is the pixel length, g a And (5) assigning the value of the pixel to be assigned.
3. Raster data packets.
3-1: the number of packets n is determined. The consecutive raster data are grouped, the data in the group should have obvious clustering characteristics, and the grouping number should be determined according to the following method.
(1) Histogram statistics is carried out on the raster data, and a pixel value range [ g ] is obtained min ，g max ]The number of pixels corresponding to each value.
(2) Extracting a histogram contour line and smoothing;
(3) Searching the peak value of the histogram through a findpeaks function in matlab, setting the minimum interval number between two peak values to be 5, and obtaining the total number of peak values meeting the condition as the grouping number S.
3-2: and (5) counting the characteristics in the group.
(1) Value field [ g ] min ，g max ]Equally dividing into S groups, and counting the corresponding pixel number n in each group k (k∈[1,S]) Each group of corresponding value ranges is [ str ] k ,end k ]；
3-3: the values in the group are subjected to characteristic statistical analysis,obtaining probability density P of values in each group k Distribution dispersion D k Gradient V of variation k 。
(1) From the data histogram, calculate [ str ] k ,end k ]The probability density sum corresponding to all values in the interval is used for obtaining the probability density P of the values in the group k 。
Figure BDA0004077698890000121
Wherein n is the total number of pixels, n k Is of the value [ str ] k ,end k ]All the pixels in the interval.
(2) The degree of dispersion of the numerical space distribution is measured by the proportion of the covariance matrix norms of all numerical space coordinates in the group, and the [ str ] is searched k ,end k ]Space coordinates corresponding to all values in the interval
Figure BDA0004077698890000122
Figure BDA0004077698890000123
Calculating the numerical distribution dispersion D in the group k The formula is as follows:
COV(X k ,Y k )＝E[(X k -E(X k ))(Y k -E(Y k ))]
Figure BDA0004077698890000124
/>
Figure BDA0004077698890000125
Figure BDA0004077698890000126
wherein ,
Figure BDA0004077698890000127
E(X k ) Is X k Desirably, E (Y k ) Is Y k Desirably, C is the covariance matrix; />
Figure BDA0004077698890000128
Singular values of C; f (F) k Is the covariance norm of the k-th group.
(3) Obtain each interval [ str ] k ,end k ]The number of pixel values and the next adjacent interval str k+1 ,end k+1 ]Gradient V of variation of pixel values of (2) k 。
Figure BDA0004077698890000129
Figure BDA00040776988900001210
wherein ,nk+1 Is of the value [ str ] k+1 ,end k+1 ]All the pixels in the interval.
3-4: and (5) adjusting the group spacing.
(1) According to the numerical characteristic parameters of each group counted in the step 3-3, forming characteristic weights A in each group k 。
Figure BDA0004077698890000131
(2) And optimizing the interval between each group of groups through characteristic weight adjustment. Value field [ g ] min ，g max ]Readjusting according to the characteristic weights in the groups, and then the kth group spacing is as follows:
F k ＝(g max -g min )×A k
4. a value-to-band mapping table is established. And selecting a color model for the determined optimal S group numerical values, and performing data mapping, see figure 4. The data mapping formula is as follows:
IMG(x,y)→{[str 1 ,end 1 ],[str 2 ,end 2 ]……[str s ,end s ]}→ColorBar
wherein IMG is raster data, [ x, y ] is space abscissa and color bar is color bar.
Those of ordinary skill in the art will appreciate that the elements and method steps of each example described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the elements and steps of each example have been described generally in terms of functionality in the foregoing description to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the several embodiments provided in this application, it should be understood that the disclosed methods and systems may be implemented in other ways. For example, the above-described division of units is merely a logical function division, and there may be another division manner when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted or not performed. The units may or may not be physically separate, and components shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiment of the present invention.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit of the invention, and are intended to be included within the scope of the appended claims and description.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.

Claims (10)

1. The marine environment remote sensing visual expression method based on numerical feature statistics is characterized by comprising the following steps of:
the data preprocessing module is used for carrying out noise reduction processing on the data so as to improve the data resolution and weaken the edge sawtooth phenomenon;
the data grouping and characteristic statistics module is used for obtaining data grouping and carrying out characteristic statistics through cluster analysis and characteristic statistics aiming at cluster phenomena existing in numerical distribution of marine environment remote sensing inversion raster data;
and the group interval optimization and color mapping module establishes an index relation between the set color and the numerical value interval through the set numerical value interval to perform color mapping and optimize the visual display effect.
2. The marine environment remote sensing visual expression method based on numerical feature statistics according to claim 1, wherein in the data preprocessing module, data preprocessing is performed on marine environment remote sensing inversion raster data, abnormal values and abrupt values are removed, and conversion processing is performed on special abundance distribution type data, specifically as follows:
a1, outlier rejection: according to the element attribute, the region and the climatology characteristic, carrying out value range and landing inspection on the data, and eliminating abnormal values;
a2, removing mutation values: searching for isolated peaks in the data and removing the peaks;
a3, preprocessing special abundance distribution type data: aiming at the grid data visual expression of the large gradient change process elements facing the estuary/near shore-sea basin/ocean mixing area, the logarithmic preprocessing is firstly carried out to preliminarily and uniformly distribute the abundance.
3. The marine environment remote sensing visual expression method based on numerical feature statistics according to claim 2, wherein in step A2, the method specifically comprises the following steps:
(1) Traversing to calculate average value D of abrupt change of lattice points and neighborhood lattice points ij ：
Figure FDA0004077698860000011
wherein ,gij For the target lattice point value g pq The number of the neighbor grid points is the number of the neighbor grid points, and n is the number of the neighbor grid points effective values;
(2) Calculating the spatial standard deviation sigma of the abrupt mean D ：
Figure FDA0004077698860000021
Wherein N is the number of valid values of raster data, r is the total number of rows of raster data, r is the total number of columns of raster data,
Figure FDA0004077698860000022
mean value of grid data mutation;
(3) Traversing the lattice point mutation mean value D ij And the space standard deviation sigma D Relationship, determination D ij >3σ D The lattice point of (2) is a mutation abnormal value, and an invalid value noDataValue is assigned to the mutation abnormal value to finish the elimination processing.
4. The marine environment remote sensing visual expression method based on numerical feature statistics according to claim 2, wherein in step A3, the method specifically comprises the following steps:
(1) Calculating a raster data value range;
(2) Performing histogram analysis of not less than 10 groups of raster data, and performing logarithmic processing on the raster data when the cumulative group spacing of the first n maximum frequency groups with the cumulative frequency greater than 90% is less than 10% of the range of the value range;
(3) And storing the logarithmic processing mapping relation, and determining the color bar coordinates by using inverse mapping after finishing the subsequent hierarchical optimization flow.
5. The visual expression method of marine environment remote sensing based on numerical feature statistics according to claim 1, wherein in the data preprocessing module, interpolation algorithm is adopted for spatial discontinuous or lower resolution data to improve the density, supplement missing data, eliminate and weaken internal and boundary mosaic phenomena, and more smoothly and finely describe the real distribution rule of marine environment remote sensing elements, specifically as follows:
b1, mask construction: overlapping the raster data with a land vector, setting the position of the non-DataValue of the land part as-1, setting the positions of the non-DataValue of other sea parts as 0, setting the positions of the other effective numerical values as 1, and acquiring an inversion raster data mask;
b2, hole filling: selectively filling the small-range internal missing value with the mask gray value of 0;
b3, internal interpolation: in order to prevent data distortion, the interpolation proportion of the internal data is set to be 0.5-3, and the internal continuous data is processed by using a DINOF or bilinear interpolation method according to the abundance of the time-space sequence of the data;
b4, edge interpolation: the bilinear interpolation cannot solve the problem of edge jaggies, and for edge data, the transition between the edge data and an internal effective value is natural.
6. The visual expression method of marine environmental remote sensing based on numerical feature statistics according to claim 5, wherein in step B2, the method specifically comprises the following steps:
(1) Extracting connected domain S for internal missing value with mask gray value of 0 i I=1, 2,3, … …, m, m is the total number of connected domains, and is divided into two types of small and large according to the size of the connected domains, wherein the area of the connected domains is less than or equal to 5 pixels and is set as small connected domainsThe rest is large connected domain;
(2) Dividing each small connected domain into time sequence rule filling and adjacent value filling according to whether the time sequence data is the time sequence data, and reconstructing inversion information containing time sequence data by an empirical orthogonal function interpolation method; and for single inversion information which does not contain time sequence data, assigning values to the missing values by adopting a method of counting adjacent value characteristics, and sequentially assigning and filling according to the sequence from outside to inside, wherein the filling values are average values of all adjacent values, and filling processing is not carried out on each large communication domain.
7. The visual expression method of marine environmental remote sensing based on numerical feature statistics according to claim 5, wherein in step B4, the following steps are adopted to smooth edges and weaken mosaic phenomenon:
(1) Edge finding: traversing the four adjacent domains of the pixel with the mask value of 1, and if any pixel in the four adjacent domains is-1, the pixel is an edge pixel;
(2) Reconstructing an optimal edge broken line according to edge shape fitting, dividing each edge into long side and short side combinations according to the number of pixels, dividing the edges into Z types, U types and L types according to an MLAA algorithm, and reconstructing an optimal edge line;
(3) Assigning each non-DataValue pixel penetrated by the reconstructed edge folding line, wherein the pixel value of each pixel marked by the reconstructed edge line is the length ratio mixing result of the reconstructed edge line penetrating through the adjacent pixels, and the value of the pixel to be assigned is as follows:
Figure FDA0004077698860000041
wherein T is the number g of adjacent edge pixels of the pixel to be assigned i Is the value of the ith adjacent pixel, r i To reconstruct the length of the edge line across the ith edge pixel, r is the pixel length, g a And (5) assigning the value of the pixel to be assigned.
8. The marine environment remote sensing visual expression method based on numerical feature statistics according to claim 1, wherein in the data grouping and feature statistics module, the method specifically comprises the following steps:
c1, determining the grouping number n, and grouping continuous raster data, wherein the data in the group has obvious clustering characteristics;
c2, carrying out intra-group feature statistics according to the grouping determined in the step C1;
c3, aiming at the steps C1 and C2, adjusting the group spacing;
wherein, in step C1, the number of packets should be determined according to the following method:
(1) Histogram statistics is carried out on the raster data, and a pixel value range [ g ] is obtained min ，g max ]The number of pixels corresponding to each value;
(2) Extracting a histogram contour line and smoothing;
(3) Searching a histogram peak value through a findpeaks function in matlab, setting the minimum interval number between two peak values to be 5, and obtaining the total number of peak values meeting the condition as the grouping number S;
in step C2, the specific steps are as follows:
value field [ g ] min ，g max ]Equally dividing into S groups, and counting the corresponding pixel number n in each group k (k∈[1,S]) Each group of corresponding value ranges is [ str ] k ,end k ]；
In step C3, the specific steps are as follows:
(1) From the data histogram, calculate [ str ] k ,end k ]The probability density sum corresponding to all values in the interval is used for obtaining the probability density P of the values in the group k ：
Figure FDA0004077698860000051
Wherein n is the total number of pixels, n k Is of the value [ str ] k ,end k ]All the pixel numbers in the interval;
(2) The discrete degree of the numerical value space distribution adopts all the components in the groupThe specific gravity of covariance matrix norm of numerical space coordinates is measured, find [ str ] k ,end k ]Space coordinates corresponding to all values in the interval
Figure FDA0004077698860000058
Figure FDA0004077698860000059
Calculating the numerical distribution dispersion D in the group k The formula is as follows:
COV(X k ,Y k )＝E[(X k -E(X k ))(Y k -E(Y k ))]
Figure FDA0004077698860000052
Figure FDA0004077698860000053
Figure FDA0004077698860000054
wherein ,
Figure FDA0004077698860000055
E(X k ) Is X k Desirably, E (Y k ) Is Y k Desirably, C is the covariance matrix; />
Figure FDA0004077698860000056
Singular values of C; f (F) k Covariance norms for the k-th group;
(3) Obtain each interval [ str ] k ,end k ]The number of pixel values and the next adjacent interval str k+1 ,end k+1 ]Gradient V of variation of pixel values of (2) k ：
Figure FDA0004077698860000057
Figure FDA0004077698860000061
wherein ,nk+1 Is of the value [ str ] k+1 ,end k+1 ]All the pixel numbers in the interval;
in step C4, the specific steps are as follows:
(1) According to the numerical characteristic parameters of each group counted in the step C3, forming characteristic weights A in each group k ：
Figure FDA0004077698860000062
/>
(2) Optimizing the interval between each group by adjusting the characteristic weight, and obtaining the value range [ g ] min ，g max ]Readjusting according to the characteristic weights in the groups, and then the kth group spacing is as follows:
F k ＝(g max -g min )×A k 。
9. the marine environment remote sensing visual expression method based on numerical feature statistics according to claim 8, wherein in the group interval optimization and color mapping module, the method specifically comprises the following steps:
establishing a numerical value-color band mapping table: and selecting a color model for the determined optimal S group of numerical values, and performing data mapping according to the following data mapping formula:
IMG(x,y)→{[str 1 ,end 1 ],[str 2 ,end 2 ]……[str s ,end s ]}→ColorBar
wherein IMG is raster data, [ x, y ] is space abscissa and color bar is color bar.
10. A computer-readable storage medium storing a computer program, characterized in that: the computer program, when executed by a processor, implements a marine environment remote sensing visual expression method based on numerical feature statistics as set forth in any one of claims 1-9.
CN202310113256.2A 2023-02-14 2023-02-14 Ocean environment remote sensing visual expression method based on numerical feature statistics Pending CN116258785A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310113256.2A CN116258785A (en) 2023-02-14 2023-02-14 Ocean environment remote sensing visual expression method based on numerical feature statistics

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310113256.2A CN116258785A (en) 2023-02-14 2023-02-14 Ocean environment remote sensing visual expression method based on numerical feature statistics

Publications (1)

Publication Number Publication Date
CN116258785A true CN116258785A (en) 2023-06-13

Family

ID=86678855

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310113256.2A Pending CN116258785A (en) 2023-02-14 2023-02-14 Ocean environment remote sensing visual expression method based on numerical feature statistics

Country Status (1)

Country Link
CN (1) CN116258785A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117082474A (en) * 2023-10-17 2023-11-17 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)) System for acquiring marine environment forecast data in real time by scientific investigation ship

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100798413B1 (en) * 2006-10-24 2008-01-28 (주)한국해양과학기술 System and method for visualizing surveyed data in 3D form at sea
CN102855662A (en) * 2012-07-25 2013-01-02 中国科学院对地观测与数字地球科学中心 Ocean environment visualization method
US20190073534A1 (en) * 2015-11-08 2019-03-07 Agrowing Ltd. Method for aerial imagery acquisition and analysis
CN112884675A (en) * 2021-03-18 2021-06-01 国家海洋信息中心 Batch remote sensing image color matching engineering realization method
CN113505858A (en) * 2021-08-24 2021-10-15 中煤科工集团重庆研究院有限公司 Method for identifying underground illegal operation area of coal mine based on massive activity track inversion
US20220383633A1 (en) * 2019-10-23 2022-12-01 Beijing University Of Civil Engineering And Architecture Method for recognizing seawater polluted area based on high-resolution remote sensing image and device

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100798413B1 (en) * 2006-10-24 2008-01-28 (주)한국해양과학기술 System and method for visualizing surveyed data in 3D form at sea
CN102855662A (en) * 2012-07-25 2013-01-02 中国科学院对地观测与数字地球科学中心 Ocean environment visualization method
US20190073534A1 (en) * 2015-11-08 2019-03-07 Agrowing Ltd. Method for aerial imagery acquisition and analysis
US20220383633A1 (en) * 2019-10-23 2022-12-01 Beijing University Of Civil Engineering And Architecture Method for recognizing seawater polluted area based on high-resolution remote sensing image and device
CN112884675A (en) * 2021-03-18 2021-06-01 国家海洋信息中心 Batch remote sensing image color matching engineering realization method
CN113505858A (en) * 2021-08-24 2021-10-15 中煤科工集团重庆研究院有限公司 Method for identifying underground illegal operation area of coal mine based on massive activity track inversion

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
张玉娟等: "基于多要素的物理海洋数据统计与可视化表达", 海洋信息, no. 04, 15 December 2016 (2016-12-15) *
赵彬如等: "面向海岛海岸带区域的高分遥感影像智能化色彩增强方法", 自然资源遥感, 3 July 2023 (2023-07-03) *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117082474A (en) * 2023-10-17 2023-11-17 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)) System for acquiring marine environment forecast data in real time by scientific investigation ship
CN117082474B (en) * 2023-10-17 2024-02-02 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)) System for acquiring marine environment forecast data in real time by scientific investigation ship

Similar Documents

Publication Publication Date Title
CN109102477A (en) A kind of high-spectrum remote sensing restoration methods based on the constraint of non-convex low-rank sparse
CN110264484A (en) A kind of improvement island water front segmenting system and dividing method towards remotely-sensed data
CN109558806A (en) The detection method and system of high score Remote Sensing Imagery Change
CN105335972B (en) Knitted fabric defect detection method based on small echo contourlet transform and vision significance
CN116258785A (en) Ocean environment remote sensing visual expression method based on numerical feature statistics
AU2020103047A4 (en) Crop Distribution Mapping
CN107437068A (en) Pig individual discrimination method based on Gabor direction histograms and pig chaeta hair pattern
CN111738954B (en) Single-frame turbulence degradation image distortion removal method based on double-layer cavity U-Net model
CN109784401A (en) A kind of Classification of Polarimetric SAR Image method based on ACGAN
CN114881861B (en) Unbalanced image super-division method based on double-sampling texture perception distillation learning
CN108710862A (en) A kind of high-resolution remote sensing image Clean water withdraw method
CN105550682B (en) Bronze object stone inscription inscription rubbing method
CN107392887A (en) A kind of heterogeneous method for detecting change of remote sensing image based on the conversion of homogeneity pixel
CN115810149A (en) High-resolution remote sensing image building extraction method based on superpixel and image convolution
CN104899592B (en) A kind of road semiautomatic extraction method and system based on circular shuttering
CN109241369B (en) Rainfall isopleth construction method based on grid stretching method
CN112989940B (en) Raft culture area extraction method based on high-resolution third satellite SAR image
CN109919843A (en) Skin image texture evaluation method and system based on adaptive quartering method
CN102903104B (en) Subtractive clustering based rapid image segmentation method
CN106971402B (en) SAR image change detection method based on optical assistance
CN109300115A (en) A kind of multispectral high-resolution remote sensing image change detecting method of object-oriented
CN116740579A (en) Intelligent collection method for territorial space planning data
CN118113919A (en) GDP spatialization method
CN110276270A (en) A kind of high-resolution remote sensing image building area extracting method
CN106934836A (en) A kind of haze image is based on the air light value computational methods and system of automatic cluster

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