CN104102771B - Automatic drawing method for land-sea ecotone meteorological isolines - Google Patents

Abstract

The invention discloses a method for automatically drawing meteorological contours in an interlaced zone of sea and land. It includes the following steps: 1) Read the data from the multi-dimensional marine meteorological data, obtain the sea-land boundary data, perform overlay analysis to separate the ocean area data; 2) set the threshold for judging the noise, and judge each pixel, if The pixel is a noise point, and the pixel is corrected to obtain the effective numerical range of the ocean area; 3) A division of the entire area is obtained according to the effective numerical range of the ocean area; 4) The interval tables for the ocean area and other areas are respectively constructed, and each 5) Merge the intervals of ocean areas and other areas; 6) Use MATLAB to read data in batches, and apply the intervals to MATLAB contour drawing commands to realize automatic drawing. The isolines drawn by the invention can effectively highlight the changes of the elements in the ocean area, and take into account the performance of the meteorological elements in the land area, and can be applied to the fields of making meteorological thematic maps of the sea-land transition zone and the like.

Description

A Method for Automatically Drawing Meteorological Contours in the Sea-land Ecotone

technical field

The invention relates to the technical field of marine meteorological contour drawing. In particular, it relates to a method for automatically drawing meteorological contours in the sea-land ecotone.

Background technique

Contour is a smooth curve formed by connecting points with equal values in the mapping object. It has very important uses in many fields such as hydrology, geology, ocean, atmosphere, etc., such as isobar maps in pressure fields and geodetic mapping. Contour maps of ore body contour maps in geology, etc. At present, the methods of contour drawing and filling are relatively mature, but there are problems in the field of marine meteorology. Oceanographic meteorological data includes land and ocean areas, and element values vary gradually over oceans but significantly over land. Using the traditional method of contour interval setting will make the contours concentrated in the land area, and there is a lack of distinguishable contours in the ocean area, which is the area that marine meteorologists focus on. Therefore, how to automatically set the interval of marine meteorological contours, so as to increase the information content of the drawn elements and highlight the regions of interest, is an urgent problem to be solved.

For contour lines with unequal intervals, there are currently two methods for setting intervals: one is manual setting by the drawer according to specific data; the other is interval setting based on area statistics. The manual setting method is to use the software to draw the contour line, and then adjust the interval according to the effect until a satisfactory result is obtained. This method can often get good results, but the drawing process is cumbersome, time-consuming and labor-intensive. The method of area statistics solves the problem of manual drawing to a certain extent, and can realize business batch processing, but when it is applied to special data such as marine meteorology, the drawn contour effect cannot meet professional requirements. For example, the interval of atmospheric isobars usually needs to be set to a specific value, such as 2.5 hPa or 4 hPa, and the method of area statistics is only divided according to the principle of equal area, which cannot solve such special needs.

In view of the above problems, the present invention proposes a method for automatically drawing meteorological contours in the sea-land transition zone. First, the original elements and the sea-land boundary data are superimposed and analyzed, and the ocean area is extracted, and then the numerical range of the ocean area is obtained by performing spot denoising calculations, and then Combined with the mapping table to select the appropriate interval value, merge the ocean and land areas to obtain unequal interval intervals, and finally apply the interval to the MATLAB contour drawing command to realize automatic drawing. Based on this method, the automatic drawing of meteorological contours in the sea-land ecotone can be realized, and the information of the ocean area can be highlighted.

Contents of the invention

The purpose of the present invention is to provide a method for automatically drawing meteorological contours in the sea-land transition zone in order to overcome the problems in the prior art.

A method for automatically drawing meteorological contours in the sea-land ecotone includes the following steps:

1) Read out the layer of data to be drawn from the multi-layer and multi-variable marine meteorological raw data files, and then obtain the sea-land boundary data consistent with the data format, and use the sea-land boundary data to perform overlay analysis on the original data, Separate out ocean area data;

2) Set the threshold for judging the noise, traverse each pixel in the data, if the value of the current pixel and the variance of the eight surrounding pixels minus the overall variance is greater than the given threshold, then the pixel is considered is the noise point, and the pixel is corrected to obtain the effective numerical range of the ocean area;

3) Obtain a division of the entire area according to the effective numerical range of the ocean area;

4) Construct the interval table of the ocean area and other areas respectively, and set the intervals of different densities in each area according to the interval table;

5) Merge intervals between oceanic areas and other areas;

6) Use MATLAB to read data in batches, and apply the merged interval in step 5) to the MATLAB contour drawing command to realize automatic drawing.

The step 1) includes: reading out a layer of data to be drawn isolines from a multi-layer and multi-variable marine meteorological raw data file, the layer data is composed of a two-dimensional regular grid of n rows and m columns , and then obtain the sea-land boundary data with n rows and m columns consistent with the data format. The sea-land boundary data contains the corresponding binary data, and each pixel refers to the same geographical location. 0 in the sea-land boundary binary data represents other , 1 represents the ocean, using the sea-land boundary data to do overlay analysis on the original data, and separate the ocean area data.

The step 2) includes: setting a threshold for judging the noise, traversing each pixel in the data, if the value f(i, j) of the current pixel is subtracted from the variance σ ² of the surrounding eight connected pixels remove the overall variance If it is greater than a given threshold, the pixel is considered to be a noise point. This method can be expressed by the following formula:

In the formula: g _l is the value of adjacent points around f(i, j) point;

When it is confirmed that a pixel is a noise point, replace the original value with the result calculated by the following formula,

In the formula: f′(i, j) is the new brightness of the pixel; P is the probability that the pixel is a noise point,

The ocean area data after the above correction can obtain the effective numerical range of the ocean area.

The step 3) includes: obtaining the overall numerical range, and then removing the calculation result of the ocean area, so as to obtain the numerical range of other areas; the numerical range of the ocean area QO=[minO, maxO] can divide the numerical range of other areas into two parts, upper bound QL ₁ =(minL ₁ , maxL ₁ ] and lower bound QL ₂ =[minL ₂ , maxL ₂ ), they are a division of the range QA=[minA, maxA] of the whole area;

QA＝QL ₁ ∪QO∪QL ₂ (3)

In the formula, the intersection of QL ₁ , QO and QL ₂ is empty;

If all other values are greater than the ocean area, then QL ₁ is an empty set, and if all other values are less than the ocean area, then QL ₂ is an empty set.

Described step 4) comprises: for the numerical range of the marine area and other areas of seeking, build interval table and set the interval of different densities in each area, interval table area field is divided into interest area and other areas, the interval setting of different areas The principles are different, the interval of the interest area is small, and the corresponding contour lines are also relatively dense, while the intervals of other areas are relatively large, and the contour lines drawn are also relatively sparse; the same type of area has different intervals due to different value ranges, and the stated value The range focuses on the interval to which the range difference belongs, and the numerical range difference of the ocean area is maxO-minO.

Described step 5) comprises: by the mapping of the interval table that step 4) builds, obtains following interval setting, marine area numerical range interval QIO=[minO, intervalO, maxO], other area numerical range upper limit QIL1=(minL ₁ , intervalL ₁ , maxL ₁ ] and the lower bound QIL2=[minL ₂ , intervalL ₂ , maxL ₂ ), combine the intervals of the ocean area and other areas to obtain the interval QIA of the overall area,

QIA _{＝ QIL1∪QIO∪QIL2} (4)

In the formula, the intersection of QIL ₁ , QIO and QIL ₂ is empty.

The step 6) includes: using MATLAB to read data in batches, calling the MATLAB contour drawing command contour or contourf, and applying the combined interval QIA to the command to realize automatic batch drawing.

Compared with the prior art, the present invention has the beneficial effects that: the present invention ensures that the ocean area has denser contours by setting unequal intervals, and can focus on showing the numerical changes in the ocean area, so that in the ocean area, the original The sparser isolines are distributed more scientifically and abundantly; while in less important land areas, the originally denser isolines tend to be relatively sparse to reduce the confusion of the isolines in this area; ocean areas and land areas, etc. The overall information reflected by the value line is more complete and can provide more meaningful reference value.

Description of drawings:

Figure 1 shows the automatic drawing method of meteorological contours in the sea-land ecotone;

Figure 2 is the numerical distribution of air pressure on the land surface;

Figure 3 is the numerical distribution of ocean surface air pressure;

Figure 4 is the contour drawing results.

detailed description:

As shown in Figure 1, the method for automatically drawing meteorological contours in the sea-land ecotone includes the following steps:

1) Read out the layer of data to be drawn from the multi-layer and multi-variable marine meteorological raw data files, and then obtain the sea-land boundary data consistent with the data format, and use the sea-land boundary data to perform overlay analysis on the original data, Separate out ocean area data;

2) Set the threshold for judging the noise, traverse each pixel in the data, if the value of the current pixel and the variance of the eight surrounding pixels minus the overall variance is greater than the given threshold, then the pixel is considered is the noise point, and the pixel is corrected to obtain the effective numerical range of the ocean area;

3) Obtain a division of the entire area according to the effective numerical range of the ocean area;

4) Construct the interval table of the ocean area and other areas respectively, and set the intervals of different densities in each area according to the interval table;

5) Merge intervals between oceanic areas and other areas;

6) Use MATLAB to read data in batches, and apply the merged interval in step 5) to the MATLAB contour drawing command to realize automatic drawing.

The step 1) includes: reading out a layer of data to be drawn isolines from a multi-layer and multi-variable marine meteorological raw data file, the layer data is composed of a two-dimensional regular grid of n rows and m columns , and then obtain the sea-land boundary data with n rows and m columns consistent with the data format. The sea-land boundary data contains the corresponding binary data, and each pixel refers to the same geographical location. 0 in the sea-land boundary binary data represents other , 1 represents the ocean, using the sea-land boundary data to do overlay analysis on the original data, and separate the ocean area data.

The step 2) includes: setting a threshold for judging the noise, traversing each pixel in the data, if the value f(i, j) of the current pixel is subtracted from the variance σ ² of the surrounding eight connected pixels remove the overall variance is greater than a given threshold, the pixel is considered to be a noise point, and this method can be expressed by the following formula:

In the formula: g _l is the value of adjacent points around f(i, j) point;

When it is confirmed that a pixel is a noise point, replace the original value with the result calculated by the following formula,

In the formula: f′(i, j) is the new brightness of the pixel; P is the probability that the pixel is a noise point,

The ocean area data after the above correction can obtain the effective numerical range of the ocean area.

The step 3) includes: obtaining the overall numerical range, and then removing the calculation result of the ocean area, so as to obtain the numerical range of other areas; the numerical range of the ocean area QO=[minO, maxO] can divide the numerical range of other areas into two parts, upper bound QL ₁ =(minL ₁ , maxL ₁ ] and lower bound QL ₂ =[minL ₂ , maxL ₂ ), they are a division of the range QA=[minA, maxA] of the whole area;

QA＝QL ₁ ∪QO∪QL ₂ (3)

In the formula, the intersection of QL ₁ , QO and QL ₂ is empty;

If all other values are greater than the ocean area, then QL ₁ is an empty set, and if all other values are less than the ocean area, then QL ₂ is an empty set.

Described step 4) comprises: for the numerical range of the marine area and other areas of seeking, build interval table and set the interval of different densities in each area, interval table area field is divided into interest area and other areas, the interval setting of different areas The principles are different, the interval of the interest area is small, and the corresponding contour lines are also relatively dense, while the intervals of other areas are relatively large, and the contour lines drawn are also relatively sparse; the same type of area has different intervals due to different value ranges, and the stated value The range focuses on the interval to which the range difference belongs, and the numerical range difference of the ocean area is maxO-minO.

Described step 5) comprises: by the mapping of the interval table that step 4) builds, obtains following interval setting, marine area numerical range interval QIO=[minO, intervalO, maxO], other area numerical range upper limit QIL1=(minL ₁ , intervalL ₁ , maxL ₁ ] and the lower bound QIL2=[minL ₂ , intervalL ₂ , maxL ₂ ), combine the intervals of the ocean area and other areas to obtain the interval QIA of the overall area,

QIA _{＝ QIL1∪QIO∪QIL2} (4)

In the formula, the intersection of QIL ₁ , QIO and QIL ₂ is empty.

The step 6) includes: using MATLAB to read data in batches, calling the MATLAB contour drawing command contour or contourf, and applying the combined interval QIA to the command to realize automatic batch drawing.

Example:

(1) Use the sea-land boundary data to do overlay analysis on the original data of the contour line, and separate the data of the ocean part

Take the numerical forecast product of apparent pressure in the Indian Ocean region at 12:00 on May 23, 2013 as an example. The longitude range is 30°E～125°E, the latitude range is 20°S～30°N, and the resolution is 476*251. Through the overlay analysis of the sea-land boundary data, the ocean and land data are separated. Figure 2 is a statistical chart of the surface air pressure numerical forecast product data in the land area. The analysis of the data shows that the air pressure value in the land area ranges from 600 to 1100 hPa, and its distribution is in a zigzag shape, with several consecutive intervals from 850 to 1020 peak, gradually decreasing towards both sides. Figure 3 is a statistical chart of the surface pressure numerical forecast product data in the ocean area. The air pressure values in the ocean area are concentrated in the range of 1000-1020, and the rest of the range has only a small amount of noise.

(2) Use the eight-neighbor point method to remove the noise of the data in the ocean area to obtain an effective value range

Set the threshold for judging noise to 1, traverse each pixel in the data, if the value f(i, j) of the current pixel is connected to the variance σ ² of the eight surrounding pixels minus the overall variance If it is greater than a given threshold, the pixel is considered to be noise. The corrected ocean area data can obtain the effective numerical range of the ocean area.

(3) Obtain a division of the entire area according to the effective numerical range of the ocean area

The calculated range of the overall value is 600-1100 hPa, and the range of the ocean area is 1000-1020 hPa, so that a division of the whole area is [600, 1000]∪(1000, 1020)∪[1020, 1100].

(4) Utilize expert experience to construct the interval table of ocean area and other areas respectively, and set the interval

According to the specific distribution of surface air pressure in the Indian Ocean region, a mapping interval table is designed, as shown in Table 2.

Table 1

(5) Interval between merged ocean area and other areas

Through the mapping of the interval table, the following interval settings can be obtained, the interval of the numerical range of the ocean area QIO=[1000,4,1020], the upper bound of the numerical range of the land area QIL1=(1020,40,1100] and the lower bound QIL2=[600, 100, 1000), and the interval QIA of the combined overall area = [600, 100, 1000)∪[1000, 4, 1020]∪(1020, 40, 1100].

(6) Use MATLAB to read data in batches, and apply the merged interval to the MATLAB contour drawing command to realize automatic drawing

Call the MATLAB contour drawing command contour, and apply the merged interval QIA to the command to realize automatic batch drawing. The result of isoline drawing is shown in Figure 4. Through the automatic setting of unequal intervals, the map has denser contour lines in the ocean area, which can show the air pressure changes in the ocean area, making the contour line more informative; while in the less important land area, the The contour is relatively sparse to reduce the confusion of the contour and provide more meaningful reference value.

This embodiment uses sea-land boundary data to process the region of interest, and combines the mapping table to automatically match the land and sea interval values, effectively highlighting changes in elements of the sea area, and can be applied to the production of marine meteorological thematic maps and other fields.

Claims (7)

1. a kind of sea land ecotone meteorological contour line automatic drafting method is it is characterised in that comprise the steps：

1) read out, from multilayer multivariable maritime meteorology raw data file, the layer data treating drawing isoline, then obtain The extra large land boundary data consistent with this data form, does overlay analysis using extra large land boundary data, separates and go to sea to initial data Oceanic province numeric field data；

2) set the threshold value judging noise, each pixel in ergodic data, if the numerical value of current pixel and eight being around connected The variance of individual pixel deducts overall variance and is more than given threshold value then it is assumed that this pixel is noise, and this pixel is corrected, Sea area Effective Numerical scope is obtained with this；

3) division of Zone Full is obtained according to the Effective Numerical scope of sea area；

4) build the spacing sheet in sea area and other regions respectively, arranged between regional different density according to spacing sheet Every；

5) merge the interval in sea area and other regions；

6) utilize MATLAB batch to read data, and by step 5) merge after interval be applied to MATLAB isoplethes drawing and refer to Order, realizes automatically drawing.

2. a kind of sea according to claim 1 land ecotone meteorological contour line automatic drafting method is it is characterised in that described Step 1) include：Treat drawing isoline one layer is read out from the multivariable maritime meteorology raw data file of multilayer Data, this layer data is to be made up of the two-dimentional regular grid of n row m row, then obtains the sea of the n row m row consistent with this data form Land boundary data, extra large land demarcates packet containing two-value data corresponding thereto, and each pixel quotes identical geographical position, In extra large land boundary two-value data, 0 represents other, and 1 represents ocean, does overlay analysis using extra large land boundary data to initial data, point Separate out sea area data.

3. a kind of sea according to claim 1 land ecotone meteorological contour line automatic drafting method is it is characterised in that described Step 2) include：One threshold value judging noise of setting, each pixel in ergodic data, if the numerical value f of current pixel (i, J) variances sigma with eight pixels being around connected²Deduct overall varianceMore than given threshold value then it is assumed that this pixel is to make an uproar Point, the method can be represented with below equation：

$\frac{1}{8}{\mathrm{\Σ}}_{l=1}^8{\[g_l-f(i,j)\]}^2-\stackrel{\‾}{{\mathrm{\σ}}^2}>\mathrm{\ϵ}---\left(1\right)$

In formula：L is the numbering of eight adjoint point pixels around f (i, j) point, g_lPut the numerical value of surrounding adjoint point for f (i, j)；

When confirming that certain pixel is noise, replace initial value with the result that below equation calculates,

$f^{\′}(i,j)=(1-P)f(i,j)+P\[\frac{1}{8}{\mathrm{\Σ}}_{l=1}^8{g}_l\]---\left(2\right)$

In formula：F ' (i, j) is the new brightness of pixel；P is the probability that this pixel is noise,

Sea area data after above correction can obtain sea area Effective Numerical scope.

4. a kind of sea according to claim 1 land ecotone meteorological contour line automatic drafting method is it is characterised in that described Step 3) include：Obtain overall number range, then remove the result of calculation of sea area, the numerical value in other regions is obtained with this Scope；Other subfield value scopes can be divided into two parts, the upper bound by sea area number range QO=[minO, maxO] QL₁=(maxO, maxL₁] and lower bound QL₂=[minL₂, minO), they are scopes QA=[minA, maxA] of Zone Full One division；

QA=QL₁∪QO∪QL₂(3)

QL in formula₁, QO and QL₂It is all empty for occuring simultaneously two-by-two；

If other numerical value are all greater than sea area, QL₁For empty set, if other numerical value are all less than sea area, QL₂For Empty set.

5. a kind of sea according to claim 1 land ecotone meteorological contour line automatic drafting method is it is characterised in that described Step 4) include：For the number range of the sea area obtained and other regions, build spacing sheet setting regional not With the interval of density, spacing sheet area field is divided into interest region and other regions, and the interval setting principle of zones of different is different, Interest region interval is less, and corresponding isopleth is also closeer, other interregional every relatively large, the isopleth of drafting is also diluter Dredge；Because the different setting different interval of number range, described number range is concerned with belonging to range differences in same class region Interval, sea area number range difference be maxO-minO.

6. a kind of sea according to claim 1 land ecotone meteorological contour line automatic drafting method is it is characterised in that described Step 5) include：By step 4) mapping of spacing sheet that builds, obtain following interval setting, between the number range of sea area Every QIO=[minO, intervalO, maxO], other subfield value upper range limit QIL1=(minL₁,intervalL₁, maxL₁] and lower bound QIL2=[minL₂,intervalL₂,maxL₂), merge the interval in sea area and other regions, obtain whole The interval QIA of body region,

QIA=QIL₁∪QIO∪QIL₂(4)

QIL in formula₁, QIO and QIL₂It is all empty for occuring simultaneously two-by-two.

7. a kind of sea according to claim 1 land ecotone meteorological contour line automatic drafting method is it is characterised in that described Step 6) include：Using MATLAB batch read data, Calling MATLAB isoplethes drawing instruction contour or Contourf, the interval QIA after merging is applied to this instruction, realizes automatic batch and draws.