Discussion Measurement Models and Algorithms of the Wind Vector Field Based on Satellite Images

Transcription

1 pplied Matheatics, 2013, 4, Published nline ugust 2013 ( Discussion Measureent Models and lgoriths of the Wind Vector Field Based on Satellite Iages Ting Lou *, Liang Lin, Ni Zhan College of Science, Guilin University of Technology, Guilin, China Eail: * @qq.co Received pril 7, 2013; revised May 7, 2013; accepted May 14, 2013 Copyright 2013 Ting Lou et al. This is an open access article distributed under the Creative Coons ttribution License, which perits unrestricted use, distribution, and reproduction in any ediu, provided the original work is properly cited. BSTRCT This article ais to discuss the strike two-diensional wind vector on geostationary satellite iageries. The agnitude and direction of the wind vector are decided by the oving speed of the clouds. First, based on the features of the cloud ap, we extract the characteristics of clouds and establish atching odel for the clouds iage. Maxiu correlation coefficient between the target odules and tracking odule is obtained by using infrared brightness teperature crosscorrelation coefficient ethod. Then, the beginning and end of the wind vector can be ascertained. Using the spherical triangles of the law of cosines, we deterine the agnitude and direction of the wind vector. Keywords: Meteorological Satellite; Cloud Motion Winds; Related Coefficient; Styling; Iage Matching 1. Introduction variety of detection tools provide various eans to obtain inforation for basic theoretical research. Wind vector generated due to atospheric otion can also be obtained through a variety of detection tools. To detect wind vector in addition to the use of traditional wind vane, aneoeter (such as the windill aneoeter, theral aneoeter, acoustic aneoeter and laser aneoeter), obtained directly, but also through other eans, such as weather balloons, radars and satellites sounding data. The eteorological satellite reote sensing since the 1960s is a ajor breakthrough for eteorological observation. The satellite iageries inforation it provides cannot be copared to any previous detection eans on tie and space continuity. The eteorological satellite iageries have played an extreely iportant role in the research about astering the atospheric circulation, ediu and long-ter weather forecasts and severe weather science. ethod for observing atospheric circulation is to ark wind vector on the satellite iage. The agnitude and direction of the wind vector is decided by oving speed of the cloud. The wind vectors where there are no clouds or clouds unstable are defined as zero wind vectors. Defined by the oveent of the clouds, the wind vector is called cloud track wind. The eteorologi- * Corresponding author. cal departent already has soe ethods to coputing cloud track wind based on the changes of the satellite cloud. Such ethod is also called cloud otion wind. Izawa and Fujita [1], Hubert and Whitney [2] took advantage of ovie aniation technology to intuitive judgent the oveent of clouds by the visible screen, resulting in the cloud otion wind vector. This anual identification ethod is obvious anthropogenic factors greatly, not very accurate, tie-consuing and laborintensive. End lich [3] and Wolf [4], using pattern recognition techniques respectively, calculate cloud track wind according to find soe characteristic quantities of the clouds. Leese [5] and Sith [6] use infrared brightness teperature cross-correlation coefficient ethod to cloud track wind. The ain purpose of this paper is as follows: t first, obtain Infra-red satellite wind iage by using the geostationary eteorological satellite. Then, according to the adjacent ties iage data and its characteristics obtained by the infra-red satellite wind iage, establish easureent odels and algoriths to describe the actual wind vector field as accurately as possible. Deterine a two-diensional wind vector need four indicators: The latitude and longitude of the starting point and direction of the wind vector (unit: ass angle fro clockwise direction of true north) and size (unit: /s). This article gets the two-diensional gray-scale atrix data obtained by the geostationary eteorological satel-

2 T. LU ET L. 123 lite scanning. Zhang Qingshan [7], Wang Guanghui [8] and Ni Zhan [9] have done a lot of research for the actual positioning of the satellite cloud- the conversion of visual field coordinates and the ground coordinates. Those provide a great help for researching the cloud track wind in this article. ccording to analyze the feature of the clouds in the iage, we can binarization processing to the cloud pictures using threshold ethod and extract the characteristics of clouds. Then, establish atching odel for the clouds iage, use Infrared brightness teperature cross-correlation coefficient ethod to get the axiu correlation coefficient between the target odule and the tracking odule, and get the start and end of the wind vector. Then we can take advantage of the spherical triangles of the law of cosines to deterine the agnitude and direction of the wind vector. Finally, we analyze and discuss the feasibility of the odel according to a case. 2. Extraction of the Feature Clouds Use by the oveent of the clouds to strike a wind vector, we ust first extract the clouds. Satellite cloud picture is a picture which is ade up of the covered clouds and the earth s surface features obtained by eteorological satellite s top-down observations. It is drawn with grayscale data which are converted fro the teperature data over the earth detected by infrared detector on geosynchronous satellite. Earth is the sole source of the detected teperature. If the sky is cloudless, the detected teperature can be regarded as the teperature of the surface of the earth; where there are clouds, the detected teperature is relatively low. nd the cloud is higher the teperature is lower; the detected teperature can be seen as the teperature of the clouds area. ccording to the conversion between the gradation value and the teperature we can know that the gray value is greater the teperature is lower. Therefore, we can know where the grey value is larger is the clouds covered. Extract orphological characteristics of satellite clouds iage, we need to separate clouds fro satellite iage. There are any ethods to extract the clouds [10,11], the ost coonly is threshold ethod a gray value is selected as the threshold value to binarization-processing the original iage. For easy and fast calculation, we can de fine the ean value of the gray value as the threshold, defined as follows: R 0 n i1 j1 bij, n where in the su of the gray values is total nuber of the gray value is n. n i1 j1 b i, j (1). The In this article, we extract cloud ass in accordance with pixels for a region. When the ean of the gray values of the selected area is greater than the threshold, we think there are clouds and we can calculate the wind vectors continue. However, when the ean of the gray values of the selected area is less than the selected threshold, we can treat this area as cloudless regions. Consider this place as zero wind vectors area. When all the clouds are extracted, the next step is to atch clouds. 3. The Cloud Iage Module Matching 3.1. Correlation Coefficient In the tracer cloud object odule s tracing algorith of the ventilation syste, which odule identification factor is of ten used is the correlation coefficient between the target odule and the tracking odule, and satisfactory results has been achieved. ccording to the research of Wang Zhenhui et al. [12], this paper utilizes the correlation coefficient C of the original dot pitch to achieve the odule atching, which is defined as follows: C I, J, M, N S0i, js1i, j S0i, j S1i, j In the expression, S and 0 S1 respectively ean pixel intensity atrix which coe fro two different odules to be copared. nd C I, J, M, N is the correlation coefficient of the original dot pitch between S 0 whose reference position is I, J and S1 whose reference position is M, N. Use C I, J, M,N as recognition factor in the producing process of the wind vectors. In addition, we can use it as a control condition in the quality control of the wind vector to ake the wind vector in a higher reliability on the continuity and uniqueness when we conduct the ultiple-factor general editing for the generated cloud otion wind Matching lgorith In this case, we set the Matching window of the Pixel block size to pixels and the search range is liited to pixels. The steps are as follows: 1) Take two consecutive geostationary satellite cloud iages, interval between 0.5 h to 1 h. 2) We deterine a sapling point in wind vector field of the first cloud ap. The point is the starting position for the wind vector. Then we can get a saple odule T it is a pixels odule with as the center (Figure 1). 3) Find the probed area S it is a pixels odule with as the center in the second cloud iage. 4) In the second cloud iage, scan the target teplate T over the probed area S, pixel by pixel. Calculate the (2)

3 124 T. LU ET L. correlation coefficient C once T oves one pixel. So we need to calculate a total of correlation coefficients. It indicates that the siilarity of those two areas is the largest when the value of C is the axiu of all. Therefore, we choose the position as the ending of the vector where C is the axiu in the second iage. By now, the position of a wind vector is deterined (Figure 2). 5) Match as described above for the next sapling point until the end of the atching. We can use the MTLB software for the specific operation. 4. Wind Vectors Calculation Deterining a two-diensional vector requires four indicators: The latitude, longitude of the starting point, direction (unit: easured angle fro true north clockwise) of the wind vector and size. fter the above odule atch, we have coe to the beginning and end of the wind vector. Next, we have to calculate the size and direction of the wind vector. Set the initial position of the tracer cloud as and the end position as B. Point and B is arked on the sphere (Figure 3). Because points, B, and vertex C copose a spherical triangle, so we can seek the agnitude and direction of the wind vector in accordance with the law of cosines of the spherical traingles. Spherical triangles of the law of cosines [13]: ny side of the spherical triangle cosine equal to the other on both sides of the cosine of the product of plus this on both sides of the sine and cosine of the angle even ultiplied. Calculated as follows: First of all, the row and colun coordinates of, B can be converted into Latitude and longitude coordinates through the conversion of visual field coordinates and the ground coordinates. It was ade a detailed description in reference [9] (see Figure 4). s shown above: Infrared detector is located at S. nd P is an arbitrary sapling point on the Earth. Q(I, J) is sub-satellite point coordinates. Q = a and N = b are the sei-ajor and sei-inor axes of the ellipsoid. We can obtain the conversion Forula (3). S (, i j) 1 B 2 B 0 Sij (, 0 ) (a) (b) (a) S 0 (b) S 1 Figure 1. Module grabs. Figure 2. Matchingprocess. C B J B C N E Figure 3.The displaceent of the scheatic of and B. M 2 acda addc 1) nc 2 ad1 arctan 2 ca a d dc 1) a cd a a d dc 1) ad1 ad1 ca a d dc 1 0 arctan 2 acd addc 1 2 (3)

4 T. LU ET L. 125 N FX, 360, (8) where 0 l k Figure 4.Coordinatetransforation [9]. tan, n tan J j I i, n d. a b So we can get that the longitude and latitude of the point are 1 and 1, and the longitude and latitude of the point B are 2 and 2. We can see the earth as a unifor sphere. Earth s radius is R. So in the Spherical triangle CB, according to the law of cosines we have: cos B sin Csin CB cos EJ cos Ccos CB coscb sin Csin B cosmn cos Ccos B nd considering B, C 90, CB 90, EJ, MN 2 1 Then we can get 1 2 cos sin 901 sin 902 cos 2 cos901cos902 cos902sinsin 901cos coscos901 This is equivalent to 1 (4) (5) 1 cos sin1sin2 cos1cos2cos (6) cos sin2 cossin1 sin cos1 In addition, B R, wind speed for the wind vector can be obtained as FS R t (7) ong the, t is the tie difference through the point to the point B. nd wind direction is 5. Nuerical Exaple Two gray-scale atrix obtained by the existing Chinese FY-2 satellite which is located at 0 N latitude and 86.5 E longitude, IR1_2100.at and IR1_2130.at, respectively indicated ne day at 21:00 and 21: gray value atrix. Given the distance between synchronous satellite and the center of earth is 42,164,000. Earth can be regarded as an ideal sphere. The radius of the earth is The detection scans by the stepping angle of 140 icro-arcs. Calculation range is: 40 degrees south latitude to 40 degrees north latitude, 46 degrees to 126 degrees east longitude. t first, the satellite cloud iage is drawn with the data of gray value atrix IR1_2100.at and IR1_2130.at in MTLB, as shown in Figure 5. Then, according to the ethod described herein, we add the wind vectors into the satellite cloud ap, shown in Figures 6 and 7. Figure 5. Satellite cloud ap. Figure 6. The wind vector ap.

5 126 T. LU ET L. Figure 7. Local direction. s can be seen fro the above clear that it basically draw the wind vector at the nonzero wind vector area and the directions of the arrows roughly agree with the clouds oving trend. bout the size of the wind speed, for instance, the wind speeds of typhoon located at the lower right corner are around 20 (/s). It is generally consistent with the weather of the tie. 6. Conclusion Cloud otion wind ap reflects the strength and direction of wind very well. It can be generalized as the iportant reference for the long-ter weather forecasts in eteorology. It is also playing an iportant role on astery atospheric circulation and indicating disastrous weather. This article has copletely described the whole process of calculating cloud otion wind by the use of the theory of analytic geoetry and MTLB coputing software. The ethod is siple, fast, and easy to operate and proote. Using relevant eteorological data, we present intuitive and basically consistent with the actual cloud otion wind ap. However, the window size of the clouds this algorith selected is liited to pixels. nd the search scope is liited to pixels. But in practical application, fixed window size and search scope do not copletely extract the feature inforation of iages. Missing inforation reduce the quality of the pixel block atching. Therefore, it has liitations. In order to iprove the quality of the pixel block atching, we can consider abolishing the liited window size and search scope. Design the effective ethod which can adaptively deterine the window size and the search range. Because errors exist in both the cloud ap atching process and the cloud otion wind calculating process, how to reduce these errors and ake the accuracy of the results to a further iproveent and enhanceent is our work that we need to continue research. REFERENCES [1] T. Izawa and T. Fujita, Relationship between Winds and Cloud Velocities Deterined fro Pictures btained by the ESS 3, ESS 5 and ST-I Satellites. North-Holland Publishing Co., Space Research IX, sterda, 1969, pp [2] L. F. Hubert and L. F. Whitney, Wind Estiation fro Geostationary-Satellite Picture, Monthly Weather Review, Vol. 99, No. 9, 1971, pp [3] R. M. Endlich, D. E. Wolf, D. J. Hall and. E. Brain, Use of a Pattern Recognition Technique for Deterining Cloud Motions fro Sequences of Satellite Photographs, Journal of pplied Meteorology, Vol. 10, No. 1, 1971, pp [4] D. E. Wolf, D. J. Hall and R. M. Endlich, Experients in utoatic Cloud Tracking Using SMS-GES Data, Journal of pplied Meteorology, Vol. 16, No. 11, 1977, pp [5] J.. Leese, C. S. Novak and B. B. Clark, n utoated Technique for btaining Cloud Motion fro Geosynchronous Satellite Data Using Cross Correlation, Journal of pplied Meteorology, Vol. 10, No. 1, 1971, pp [6] E.. Sith, The McIDS Syste, IEEE Transactions on Geoscience Electronics, Vol. 13, No. 3, 1975, pp [7] Q. S. Zhang, Method of Static Meteorological Tatellite Cloud Iage Landark Navigation, China cadeic Journal Electronic Publishing House, Beijing, [8] G. H. Wang, B. Chen and W. P. He, Research of Longitude and Latitude of VHRR 2048 Based on Polar Coordinate Transforation, Journal of Qingdao University, Vol. 16, No. 3, 2003, pp [9] Z. Ni, L. Liang ND L. Ling, Interconverting Models of Gray Matrix and Geographic Coordinates Based on Space nalytic Geoetry, pplied Matheatics, Vol. 4, No. 1, 2013, pp [10] Y. B. Li, Y. X. Li and J. Q. Qu, Method of Extracting and Representing Morphological Features of Satellite Cloud Iages, Journal of Nanjing Institute of Meteorology, Vol. 29, No. 5, 2006, pp [11] H. Li, Y. Wang and F. Y. Liao, btaining the Mean Scale of Cloud ggloerate by Correlation nalytical Method fro Infrared Satellite Cloud Picture, Scientia Meteorologica Sinica, Vol. 23, No. 3, 2003, pp [12] Z. H. Wang, D. Z. Tang and D. S. Fu, nanalytical and Experiental Coparision aong the Pattern-Recognition Paraenters in Iage Cross-Correlation nalysis, Scientia Meteorologica Sinica, Vol. 12, No. 1, 1992, pp [13] X. J. Zhou and S. D. Chu, Forula of Spherical Triangle and Its pplication, Journal of Zhejiang International Maritie College, Vol. 4, No. 1, 2008, pp