CN112669236B - Distortion correction method based on foundation cloud picture - Google Patents
Distortion correction method based on foundation cloud picture Download PDFInfo
- Publication number
- CN112669236B CN112669236B CN202011621505.1A CN202011621505A CN112669236B CN 112669236 B CN112669236 B CN 112669236B CN 202011621505 A CN202011621505 A CN 202011621505A CN 112669236 B CN112669236 B CN 112669236B
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- sky
- angle
- radial distance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- 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
- Image Processing (AREA)
Abstract
The invention relates to the field of digital image processing, and provides a distortion correction method based on a foundation cloud picture, which comprises the following steps of firstly, determining the radial distance from a pixel in an image to a zenith; then, determining the zenith angle of a pixel in the image according to an image radial distance-zenith angle formula obtained by fitting; then, determining the sky radial distance from the pixels in the space to the zenith based on the conversion model; determining the radial distance of the sky image and the coordinates of pixels in the sky image and completing mapping; and finally, interpolating the sky image with incomplete information to obtain a complete distortion correction image. The invention is mainly applied to the digital image processing occasions.
Description
Technical Field
The invention relates to the field of digital image processing, in particular to a distortion correction method based on a foundation cloud picture.
Background
The weather image is one of important data in the weather related field, can represent local atmospheric state, and becomes a key research target in the fields of weather analysis, weather monitoring and the like. The collection of the weather image is completed by the foundation equipment, and the collected weather image is called a foundation cloud picture. The ground equipment monitors local sky in real time, and presents information such as cloud distribution, illumination distribution and the like in a high space-time resolution mode. In the research of a power prediction model of a distributed photovoltaic power station, the motion and the distribution of a cloud cluster are one of important factors causing the change of photovoltaic power. In order to realize accurate cloud cluster motion prediction and photovoltaic power prediction, distortion correction of a foundation cloud picture needs to be carried out, the foundation cloud picture acquired by foundation equipment is corrected, and finally a real sky state is presented.
The ground equipment adopted by the invention is a Total Sky Imager (TSI), and an imaging system of the TSI is suspended above a hemispherical mirror for mapping the Sky by a supporting arm, so that the ground cloud picture has not only the distortion of a simple fisheye lens, but also the distortion caused by a complex optical path brought by the hemispherical mirror and a dome-shaped cloud layer. There are two existing distortion correction methods. The first method first adjusts the pixel position based on empirical analysis of radial distortion to account for geometric distortion along the radial direction. Further, solid angle correction is performed to correct the problem of pixels in the image corresponding to different solid angles. Features calculated from the image can be corrected by this method, but the distortion-corrected image cannot be directly generated. Another method is to transform the ground cloud map to sky coordinates based on geometric transformation, first fit the zenith angle-radial distance curve in radial distortion, and then perform the transformation of the ground cloud map from spherical coordinates to rectangular coordinates network. The method is the current mainstream method, and has high efficiency and good accuracy, however, the accuracy of the model limits further improvement of the accuracy of the method, and therefore, a distortion correction method based on a higher-accuracy model is urgently needed to be provided.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide a distortion correction method based on a foundation cloud picture, and the method comprises the following steps of firstly, determining the radial distance from a pixel in an image to the zenith; then, determining the zenith angle of a pixel in the image according to an image radial distance-zenith angle formula obtained by fitting; then, determining a sky radial distance from the pixel to the zenith in the space based on the conversion model; determining the radial distance of the sky image and the coordinates of pixels in the sky image and completing mapping; and finally, interpolating the sky image with incomplete information to obtain a complete distortion correction image.
The method comprises the following specific steps:
step 1, determining the radial distance and polar angle of an image:
the position of a pixel in an image is represented by a pixel coordinate system, the origin of the coordinate system is positioned at the upper left corner of the image, the horizontal axis is a column u, the vertical axis is a row v, the pixel coordinate refers to the row and the column of the pixel in the image, the unit is the pixel, in order to realize the mapping of an original image to a distortion corrected sky image, a radial mapping method is adopted, the polar coordinate of the pixel in the image, namely the radial distance of the image and the polar angle of the image, is required to be determined, and the detailed steps are as follows:
1) The polar coordinate system has its pole set as zenith and is located in the center of sky area in the image and has pixel coordinate (u)o,vo) The polar diameter of the polar coordinates of the pixel, i.e. the radial distance from the pixel in the image to the zenith, is calculated by the following formula:
wherein (u)I,vI) Is the pixel coordinates of the image pixel. The radial distance of the image characterizes the pixel in the original imageDistance of zenith position;
2) And the polar angle phi of the polar coordinate of the pixel, namely the angle of the pixel in the image deviating from the positive direction, takes the vertical direction in the south in the representation geographical position as the initial direction and takes the clockwise direction as the positive direction, and is calculated by the following formula:
the calculation of the polar angle accords with the theoretical description of the celestial body coordinate in the space, and is convenient for understanding and subsequent mapping;
step 2, determining zenith angles of pixels in the image:
the position of an object in a space is represented by a celestial body coordinate system, in order to construct a mapping relation between a pixel coordinate in an image and a celestial body coordinate in the space, a relation between an image radial distance of the image pixel coordinate and a zenith angle of the space celestial body coordinate is established, and the zenith angle of the pixel in the image is determined according to the relation, and the method specifically comprises the following steps:
1) The sun is one of the entities located in the sky area in the image, and the real-time moving sun has a changed spatial position, so that the mapping relation between the pixel coordinate system of the image and the celestial body coordinate system of the space can be constructed by taking the sun as an object.
For a certain specific moment, the coordinates of the sun in the pixel coordinate system of the image can be detected from the original foundation cloud picture, the sun is positioned on the shading band of the foundation cloud picture, when the condition that the cloud shades the sun does not exist, the position of the sun shows high brightness and is presented as a sun disc, and the central coordinate of the sun disc is extracted as the pixel coordinate of the sun;
the horizontal coordinate system uses a local horizon of an observer as a base plane, a visible object above the horizon is represented by horizontal coordinates, namely a zenith angle alpha and an azimuth angle beta, the unit is DEG, the zenith angle is an angle between a zenith and the object based on the observer, the azimuth angle is a clockwise relative angle of the object taking a south as an initial direction, and the zenith angle of the sun is calculated by an astronomical formula:
wherein, delta is the solar declination angle calculated according to the date,
for geographic latitude, τ is the solar time angle calculated from the date, time and longitude:
2) The radial image distance of the image pixel coordinate and the zenith angle of the space celestial body coordinate have the same meaning and both represent the distance between the pixel and the zenith angle, so that foundation cloud pictures with different dates and times and without sun shielding are randomly selected, the radial image distance and the zenith angle of the sun are calculated, the transverse axis is the radial image distance, the longitudinal axis is the zenith angle, and the radial image distance and the zenith angle data of the sun are distributed in a curve manner, so that a cubic polynomial is fitted to obtain an image radial distance-zenith angle formula;
3) After the radial distance-zenith angle formula of the image is determined, substituting the radial distance of the image obtained by the step 1 into the formula, and finally finishing the calculation of the zenith angle of the pixel in the image;
step 3, determining the radial sky distance:
in order to correct and obtain the radial distance from a pixel in the sky to the zenith, namely the radial distance of the sky, a geocentric angle is used as an indirect variable, the relation between a zenith angle under a fixed azimuth angle and the arc length of a corresponding cloud layer is constructed, namely a conversion model of the zenith angle and the radial distance of the sky is constructed, and the calculation of the radial distance of the sky is realized;
step 4, determining the radial distance of the sky image, realizing coordinate transformation and finishing mapping:
converting the sky radial distance with a larger numerical value into a sky image radial distance, further converting the polar coordinates of the sky image into pixel coordinates, and finishing the mapping from the pixel to the distortion-corrected sky image;
step 5, interpolation is carried out on the sky image:
adopting a radial interpolation method:
searching pixels B and C which are adjacent to a pixel A to be interpolated in the radial direction and contain pixel information, and if two adjacent pixels exist in the radial direction, giving the average value of the two pixels to A; if there is one adjacent pixel in the radial direction, the pixel value is given to a. And traversing all pixels to be interpolated in the sky image to obtain a final distortion correction result.
The step 3 comprises the following steps:
1) And the ground equipment observes the top sky by an earth surface observation point, H is cloud height, H is the earth radius, and theta is the geocentric angle. Assuming that the earth is a standard sphere and the cloud layer height is fixed, the difference caused by these assumptions can be ignored for the current distortion correction research target, and according to the geometric model, the conversion formula of the zenith angle and the geocentric angle is derived as follows:
therefore, the zenith angle based on the earth surface observation point is substituted, and the geocentric angle based on the earth center is calculated.
2) According to the geometric model, a conversion formula of the geocentric angle and the sky radial distance is obtained through derivation as follows:
rS＝θ×(H+h) (5)
wherein r isSIs the sky radial distance of the pixel. Two derived conversion formulas are the conversion models of the zenith angle and the sky radial distance based on the geometric models of the zenith angle and the sky radial distance of the pixel, and the calculation of the sky radial distance is realized.
Step 4 detailed steps are as follows:
1) Setting the radius of a sky area in the sky image as r, and calculating the radial distance of the sky image by the following formula:
wherein r isSmaxIs rSIs measured. Thereby obtaining the pixel in the sky imageThe polar angle of the pixel in the sky image is equal to the polar angle of the pixel in the sky image, so that the polar coordinate of the pixel in the sky image is uniquely determined;
2) The mapping of pixels in the distortion correction process needs to determine the pixel coordinates of the pixels in the sky image, and is expressed as:
uSI＝uSIo-rSI×sinφ (7)
vSI＝vSIo-rSI×cosφ (8)
wherein u isSIoAnd vSIoThe poles are rows and columns of pixel coordinates of poles in the sky image, and the poles are set to be the centers of the sky image. After the image coordinates of the pixels and the sky image coordinates are determined, the information of the pixels in the image can be given to the pixels of the corresponding coordinates in the sky image, and the mapping of the pixels is completed.
The invention has the characteristics and beneficial effects that:
the method carries out distortion correction on the foundation cloud picture, has high precision, simple and convenient algorithm and good universality, and is suitable for various foundation equipment. The sky image to be interpolated is different from a common image to be interpolated, the farther the sky image is from the zenith position, namely the center of the image, the less the number of pixels containing information, and the provided radial interpolation method is suitable for the image acquired by the foundation equipment and has better interpolation effect compared with the traditional interpolation algorithm.
Description of the drawings:
fig. 1 pixel coordinate system.
Fig. 2 pixel coordinates of the sun.
Fig. 3 a horizontal coordinate system.
Figure 4 image radial distance and zenith angle fitted curves for the sun.
Figure 5 geometric model of zenith angle versus radial distance of the sky.
Fig. 6 is a mapped sky image.
Fig. 7 radial interpolation.
Fig. 8 distortion correction results.
Detailed Description
The invention is further illustrated with reference to specific embodiments below.
The present invention proposes a more accurate method to correct distortion in an image than other distortion correction methods. Firstly, determining the radial distance from a pixel in an image to the zenith; then, determining the zenith angle of a pixel in the image according to an image radial distance-zenith angle formula obtained by fitting; then, determining a sky radial distance from the pixel to the zenith in the space based on the conversion model; determining the radial distance of the sky image and the coordinates of pixels in the sky image and completing mapping; and finally, interpolating the sky image with incomplete information to obtain a complete distortion correction image.
The distortion correction based on the foundation cloud atlas mainly comprises the following steps:
step 1, determining the radial distance and polar angle of an image:
the position of the pixel in the image is represented by a pixel coordinate system, as shown in fig. 1, with the origin of the coordinate system located in the upper left corner of the image, the horizontal axis being the column u and the vertical axis being the row v. Pixel coordinates refer to the rows and columns of pixels in an image, in pixels. In order to realize the mapping from an original image to a distortion corrected sky image, a radial mapping method is adopted, the polar coordinates of pixels in the image, namely the radial distance of the image and the polar angle of the image, are required to be determined, and the specific steps are as follows:
1. the polar coordinate system has its pole set as zenith and is located at the center of the sky region in the image, such as the pixel coordinate (u) in FIG. 1o,vo) The indicated position. The polar diameter of the polar coordinates of the pixels, i.e., the radial distance from the pixel in the image to the zenith (image radial distance), is calculated by the following formula:
wherein (u)I,vI) Is the pixel coordinates of the image pixel. The image radial distance characterizes how far and how close the pixel is to the zenith position in the original image.
2. The polar angle of the polar coordinates of the pixels, i.e., the angle at which the pixels in the image deviate from the positive direction (image polar angle), is calculated by taking the vertical direction in the southern area as the starting direction and the clockwise direction as the positive direction in the geographic position, as follows:
the calculation of the polar angle accords with the theoretical description of the celestial body coordinate in the space, and is convenient to understand and map subsequently.
Step 2, determining zenith angles of pixels in the image:
the position of an object in space is represented by a celestial body coordinate system, and in order to construct a mapping relation between pixel coordinates in an image and celestial body coordinates in space, a relation between the image radial distance of the image pixel coordinates and the zenith angle of the space celestial body coordinates is established, and the zenith angle of the pixels in the image is determined accordingly. The method comprises the following specific steps:
1. the sun is one of the entities in the sky area in the image, and the real-time moving sun has a changing spatial position, so that the mapping relation between the pixel coordinate system of the image and the celestial body coordinate system of the space can be constructed by taking the sun as an object.
For a particular time, the coordinates of the sun in the pixel coordinate system of the image may be detected from the original ground based cloud map. As shown in fig. 2, the sun is located on the shade band of the ground-based cloud map, and when there is no case where the cloud shades the sun, the sun position shows high brightness and appears as a sun disk, and the central coordinates of the sun disk are extracted as the pixel coordinates of the sun.
As shown in fig. 3, the horizontal coordinate system uses the local horizon of the observer as a base plane, and the visible objects above the horizon are represented by horizontal coordinates — a zenith angle α and an azimuth angle β, in degrees. The zenith angle is an angle between the zenith and the object with the observer as a reference, and the azimuth angle is a relative angle of the object clockwise with the south as a starting direction. The zenith angle of the sun is calculated by an astronomical formula:
wherein, delta is the solar declination angle calculated according to the date,
τ is the solar time angle calculated from the date, time and longitude for the geographic latitude.
2. The image radial distance of the image pixel coordinate has the same meaning with the zenith angle of the space celestial body coordinate, and both represent the distance between the pixel and the zenith. Therefore, foundation cloud pictures of different dates and times without sun shading are randomly selected, and the radial distance and the zenith angle of the image of the sun are calculated. As shown in fig. 4, the horizontal axis is the radial distance of the image, the vertical axis is the zenith angle, and the radial distance of the image of the sun and the data of the zenith angle are distributed in a curve, so that a cubic polynomial is fitted to obtain the formula of radial distance of the image and zenith angle.
3. And (3) after the radial distance-zenith angle formula of the image is determined, substituting the radial distance of the image obtained by the step (1) into the formula, and finally finishing the calculation of the zenith angle of the pixel in the image.
Step 3, determining the radial sky distance:
in order to correct and obtain the radial distance from a pixel in the sky to the zenith, namely the radial distance of the sky, the geocentric angle is used as an indirect variable, the relation between the zenith angle under a fixed azimuth angle and the arc length of a corresponding cloud layer is constructed, namely a conversion model of the zenith angle and the radial distance of the sky is constructed, and the calculation of the radial distance of the sky is realized. The method comprises the following specific steps:
1. the ground equipment observes the top sky from the earth surface observation point, and fig. 5 shows a geometric model of the zenith angle and the radial distance of the sky. The blue circular ring is a cloud layer, the brown circular ring is the earth surface, the red circular arc is the sky radial distance, H is the cloud height, H is the earth radius, and theta is the geocentric angle. Assuming here that the earth is a standard sphere and the cloud heights are fixed, the differences resulting from these assumptions are negligible for the current distortion correction study objective. According to the geometric model, a conversion formula of the zenith angle and the geocentric angle is obtained through derivation as follows:
therefore, the zenith angle based on the earth surface observation point is substituted, and the geocentric angle based on the earth center is calculated.
2. According to the geometric model, a conversion formula of the geocentric angle and the sky radial distance is derived as follows:
rS＝θ×(H+h) (5)
wherein r isSIs the sky radial distance of the pixel. Based on the geometric model of the zenith angle and the radial sky distance of the pixel, two derived conversion formulas are conversion models of the zenith angle and the radial sky distance, and the calculation of the radial sky distance is realized.
Step 4, determining the radial distance of the sky image, realizing coordinate transformation and finishing mapping:
the sky radial distance is the actual span of the sky, and in order to realize the sky appearing in the image with limited size, the sky radial distance with larger numerical value needs to be converted into the sky image radial distance, and further the polar coordinate of the sky image is converted into the pixel coordinate, so that the mapping from the image to the distortion corrected sky image is completed. The method comprises the following specific steps:
1. the radius of a sky area in the sky image is set as r, and the radial distance of the sky image is calculated by the following formula:
wherein r isSmaxIs rSIs measured. Therefore, the polar path of the pixel in the sky image is obtained, and the polar angle of the pixel in the sky image is equal to the polar angle of the pixel in the sky image, so that the polar coordinate of the pixel in the sky image is uniquely determined.
2. The mapping of pixels during distortion correction requires determining the pixel coordinates of the pixels in the sky image, expressed as:
uSI＝uSIo-rSI×sinφ (7)
vSI＝vSIo-rSI×cosφ (8)
wherein u isSIoAnd vSIoThe poles are rows and columns of pixel coordinates of poles in the sky image, and the poles are set to be the centers of the sky image. After the image coordinates of the pixels and the sky image coordinates are clarified, the information of the pixels in the image can be given to the pixels of the corresponding coordinates in the sky image to complete the mapping of the pixels, and the mapped sky image shown in fig. 6 is obtained.
Step 5, interpolation is carried out on the sky image:
the forward mapping mode causes part of pixels in the sky image not to be mapped to pixel information, so that an interpolation algorithm is applied to obtain a complete distortion correction result. In the traditional interpolation method, the classical interpolation method suitable for the two-dimensional image comprises nearest neighbor interpolation, bilinear interpolation and bicubic interpolation. The nearest neighbor interpolation algorithm is the most basic and simplest, but has poor effect because severe image distortion is introduced, so that the image has blockiness. Bicubic interpolation has high calculation accuracy, but a large calculation amount greatly reduces the image processing speed. The bilinear interpolation has larger calculation amount than the nearest interpolation and does not have the defect of discontinuous gray scale, and the bicubic interpolation has the defect of damaging high-frequency components, so that the image contour is slightly blurred, but the result is still satisfactory. The sky image to be interpolated is different from a common image to be interpolated, and the farther the sky image is away from the zenith position, namely the center of the image, the less the number of pixels containing information is. The bilinear interpolation cannot well process the sky image to be interpolated, and meanwhile, a special mapping mode that pixels are radially mapped according to a zenith angle under a fixed azimuth angle in the distortion correction process is considered, so that the invention provides a radial interpolation method.
The radially connected pixels with larger radial distance in the image are mapped to the sky image, and then gaps among the pixels exist, so that the radially distributed pixels in the sky image to be interpolated have correlation unlike the common image to be interpolated, and the image is reconstructed by adopting a radial interpolation method. As shown in fig. 7, the position of the yellow pixel represents the position of the sun, the red arrow is the radial direction of the pixel a to be interpolated, pixels B and C containing pixel information, which are adjacent to the pixel a to be interpolated in the radial direction, are searched, and if there are two adjacent pixels in the radial direction, the average value of the two pixels is given to a; if there is one adjacent pixel in the radial direction, the pixel value is given to a. And traversing all pixels to be interpolated in the sky image to obtain a final distortion correction result (see fig. 8).
Claims (3)
1. A distortion correction method based on a foundation cloud picture is characterized by comprising the following steps of firstly, determining the radial distance of an image from a pixel in the image to a zenith; then, determining the zenith angle of a pixel in the image according to an image radial distance-zenith angle formula obtained by fitting; then, determining a sky radial distance from the pixel to the zenith in the space based on the conversion model; determining the radial distance of the sky image and the coordinates of pixels in the sky image and completing mapping; finally, interpolating the sky image with incomplete information to obtain a complete distortion correction image; the method comprises the following specific steps:
step 1, determining the radial distance and polar angle of an image:
the position of a pixel in an image is represented by a pixel coordinate system, the origin of the coordinate system is positioned at the upper left corner of the image, the horizontal axis is a column u, the vertical axis is a row v, the pixel coordinate refers to the row and the column of the pixel in the image, the unit is the pixel, in order to realize the mapping of an original image to a distortion corrected sky image, a radial mapping method is adopted, the polar coordinate of the pixel in the image, namely the radial distance of the image and the polar angle of the image, is required to be determined, and the detailed steps are as follows:
1) The polar coordinate system has its pole set as zenith and is located in the center of sky area in the image and has pixel coordinate (u)o,vo) The polar diameter of the polar coordinates of the pixel, i.e. the radial distance from the pixel in the image to the zenith, is calculated by the following formula:
wherein (u)I,vI) Is the pixel coordinates of the pixels of the image,the radial distance of the image represents the distance between the pixel and the zenith position in the original image;
2) And the polar angle phi of the polar coordinate of the pixel, namely the angle of the pixel in the image deviating from the positive direction, takes the vertical direction in the south in the representation geographical position as the initial direction and takes the clockwise direction as the positive direction, and is calculated by the following formula:
the calculation of the polar angle accords with the theoretical description of the celestial body coordinate in the space, and is convenient for understanding and subsequent mapping;
step 2, determining zenith angles of pixels in the image:
the position of an object in a space is represented by a celestial body coordinate system, in order to construct a mapping relation between a pixel coordinate in an image and a celestial body coordinate in the space, a relation between an image radial distance of the image pixel coordinate and a zenith angle of the space celestial body coordinate is established, and the zenith angle of the pixel in the image is determined according to the relation, and the method specifically comprises the following steps:
1) Constructing a mapping relation between a pixel coordinate system of the image and a celestial body coordinate system of the space by taking the sun as an object;
when the condition that the cloud shelters the sun does not exist, the position of the sun shows high brightness and is presented as a sun disc, and the central coordinate of the sun disc is extracted as the pixel coordinate of the sun;
the horizontal coordinate system uses a local horizon of an observer as a base plane, a visible object above the horizon is represented by horizontal coordinates, namely a zenith angle alpha and an azimuth angle beta, the unit is DEG, the zenith angle is an angle between a zenith and the object based on the observer, the azimuth angle is a clockwise relative angle of the object taking a south as an initial direction, and the zenith angle of the sun is calculated by an astronomical formula:
wherein, delta is the solar declination angle calculated according to the date,
for geographic latitude, τ is the solar time angle calculated from the date, time and longitude:
2) Randomly selecting foundation cloud pictures with different dates and times and without sun shielding, calculating the radial distance and zenith angle of the image of the sun, wherein the horizontal axis is the radial distance of the image, the vertical axis is the zenith angle, and the radial distance and zenith angle data of the image of the sun are distributed in a curve, so that the curve is fitted into a cubic polynomial to obtain an image radial distance-zenith angle formula;
3) After the radial distance-zenith angle formula of the image is determined, substituting the radial distance of the image obtained by the step 1 into the formula, and finally finishing the calculation of the zenith angle of the pixel in the image;
step 3, determining the radial sky distance:
in order to correct and obtain the radial distance from a pixel to the zenith in the space, namely the sky radial distance, a geocentric angle is used as an indirect variable, the relation between a zenith angle under a fixed azimuth angle and the arc length of a corresponding cloud layer is constructed, namely a conversion model of the zenith angle and the sky radial distance is constructed, and the calculation of the sky radial distance is realized;
step 4, determining the radial distance of the sky image, realizing coordinate transformation and finishing mapping:
converting the sky radial distance with a larger numerical value into a sky image radial distance, further converting the polar coordinates of the sky image into pixel coordinates, and finishing the mapping from the pixel to the distortion-corrected sky image;
step 5, interpolation is carried out on the sky image:
adopting a radial interpolation method:
searching pixels B and C which are adjacent to a pixel A to be interpolated in the radial direction and contain pixel information, and if two adjacent pixels exist in the radial direction, giving the average value of the two pixels to A; if one adjacent pixel exists in the radial direction, the pixel value is given to A, all pixels to be interpolated in the sky image are processed in a traversing mode, and a final distortion correction result is obtained.
2. The method for distortion correction based on ground-based cloud atlas of claim 1, wherein the step 3 comprises the following steps:
1) The method comprises the following steps that the ground equipment observes the top sky through an earth surface observation point, H is cloud height, H is the radius of the earth, theta is the geocentric angle, the earth is assumed to be a standard sphere and the height of a cloud layer is fixed, differences caused by the assumptions can be ignored for a current distortion correction research target, and according to a geometric model, a conversion formula of the zenith angle and the geocentric angle is obtained through derivation:
therefore, substituting the zenith angle based on the earth surface observation point to calculate and obtain the geocentric angle based on the earth center;
2) According to the geometric model, a conversion formula of the geocentric angle and the sky radial distance is obtained through derivation as follows:
rS＝θ×(H+h) (5)
wherein r isSTwo derived conversion formulas are the conversion models of the zenith angle and the sky radial distance based on the geometric models of the zenith angle and the sky radial distance of the pixel, so that the calculation of the sky radial distance is realized.
3. The method for distortion correction based on ground based cloud atlas of claim 1, wherein the step 4 comprises the following steps:
1) Setting the radius of a sky area in the sky image as r, and calculating the radial distance of the sky image by the following formula:
wherein r isSIs the sky radial distance, r, of the pixelSmaxIs rSThereby obtaining a polar path of the pixel in the sky image, and the pixel in the sky imageThe polar angle is equal to the polar angle of the pixel in the image, so that the polar coordinate of the pixel in the sky image is uniquely determined;
2) And in the distortion correction process, the mapping of the pixels needs to determine the pixel coordinates of the pixels in the sky image, and the pixel coordinates are expressed as follows:
uSI＝uSIo-rSI×sinφ 7)
vSI＝vSIo-rSI×cosφ 8)
wherein u isSIoAnd vSIoThe method is characterized in that the method is a method for mapping a sky image, and comprises the steps of defining the image coordinates of pixels, assigning information of the pixels in the image to the pixels of corresponding coordinates in the sky image, and completing mapping of the pixels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011621505.1A CN112669236B (en) | 2020-12-31 | 2020-12-31 | Distortion correction method based on foundation cloud picture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011621505.1A CN112669236B (en) | 2020-12-31 | 2020-12-31 | Distortion correction method based on foundation cloud picture |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112669236A CN112669236A (en) | 2021-04-16 |
| CN112669236B true CN112669236B (en) | 2022-11-01 |
Family
ID=75412112
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011621505.1A Active CN112669236B (en) | 2020-12-31 | 2020-12-31 | Distortion correction method based on foundation cloud picture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112669236B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103544679A (en) * | 2012-07-09 | 2014-01-29 | 中国电力科学研究院 | Method for correcting distortion of foundation cloud pictures of all-sky imager |
| CN107917880A (en) * | 2017-11-06 | 2018-04-17 | 中国科学院寒区旱区环境与工程研究所 | A kind of height of cloud base inversion method based on ground cloud atlas |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103353952B (en) * | 2013-05-31 | 2016-08-10 | 国家电网公司 | A kind of photovoltaic power Forecasting Methodology based on ground cloud atlas |
| CN107644416B (en) * | 2017-09-15 | 2020-04-24 | 中国科学院寒区旱区环境与工程研究所 | Real-time dynamic cloud cover inversion method based on foundation cloud picture |
-
2020
- 2020-12-31 CN CN202011621505.1A patent/CN112669236B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103544679A (en) * | 2012-07-09 | 2014-01-29 | 中国电力科学研究院 | Method for correcting distortion of foundation cloud pictures of all-sky imager |
| CN107917880A (en) * | 2017-11-06 | 2018-04-17 | 中国科学院寒区旱区环境与工程研究所 | A kind of height of cloud base inversion method based on ground cloud atlas |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112669236A (en) | 2021-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11823363B2 (en) | Infrared and visible light fusion method | |
| CN110244766B (en) | Planning method and system for unmanned aerial vehicle routing inspection route of photovoltaic power station | |
| CN107644416B (en) | Real-time dynamic cloud cover inversion method based on foundation cloud picture | |
| CN111508029A (en) | Satellite-borne segmented linear array CCD optical camera overall geometric calibration method and system | |
| CN108734685A (en) | A kind of joining method of UAV system EO-1 hyperion linear array remote sensing image | |
| CN112857356B (en) | Unmanned aerial vehicle water body environment investigation and air route generation method | |
| CN108335261B (en) | A kind of Optical remote satellite orthography garland region automatic testing method | |
| CN116182805A (en) | Homeland mapping method based on remote sensing image | |
| CN114549871A (en) | Unmanned aerial vehicle aerial image and satellite image matching method | |
| CN108509844A (en) | VPER cloud cluster recognition methods based on ground cloud atlas | |
| CN109738973B (en) | All-sky imager calibration method based on solar azimuth | |
| CN112669236B (en) | Distortion correction method based on foundation cloud picture | |
| CN116972857B (en) | Night heading determining method for sector area fitting in moonlight polarized light field | |
| CN117333412B (en) | Thermal infrared image and visible light image fusion method based on least square fitting | |
| CN113936031A (en) | Cloud shadow track prediction method based on machine vision | |
| CN113739767A (en) | Method for producing orthoimage aiming at image acquired by domestic area array swinging imaging system | |
| CN108917722A (en) | Vegetation coverage calculation method and device | |
| CN107705272A (en) | A kind of high-precision geometric correction method of aerial image | |
| CN109523498B (en) | Remote sensing image space-time fusion method for field scale crop growth monitoring | |
| CN114372945B (en) | Full-sky imager based on visible light and infrared synthetic image and cloud computing method | |
| CN116245757A (en) | Multi-scene universal remote sensing image cloud restoration method and system for multi-mode data | |
| CN115330619A (en) | Local geometric fine correction method suitable for high-resolution remote sensing image | |
| Yan et al. | Render-and-compare: Cross-view 6-DoF localization from noisy prior | |
| CN113592744A (en) | Geometric precise correction method suitable for high-resolution remote sensing image | |
| CN113379648A (en) | High-resolution seven-and-resource three-dimensional image joint adjustment method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |