|
Size: 2318
Comment:
|
Size: 2373
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 3: | Line 3: |
| Trockenprogrammierung: insbesondere die Konsistenz von Bildkoordinaten von ASAR und Freibord-Footprint klären | . Trockenprogrammierung: insbesondere die Konsistenz von Bildkoordinaten von ASAR und Freibord-Footprint klären die Klasseneinteilungen von ASAR und Freibordhöhen vergleichen Frage "welche Statistik will man anwenden" beantworten |
| Line 5: | Line 5: |
| ---------- '''Am Dienstag erzeugtes Programm:''' |
|
| Line 6: | Line 8: |
| die Klasseneinteilungen von ASAR und Freibordhöhen vergleichen | {{{ #!python |
| Line 8: | Line 11: |
Frage "welche Statistik will man anwenden" beantworten ------------------------------------------------------------------------------ #Bisher erzeugtes Programm: {{{#!python |
|
| Line 31: | Line 27: |
| # the x,y positions of the freeboard footprints | # the x,y positions of the freeboard footprints |
| Line 52: | Line 48: |
| #select the class of footprints from ASAR Image | #select the class of footprints from ASAR Image |
| Line 64: | Line 60: |
| std_kl[kl-1]=std(fh_kl) | std_kl[kl-1]=std(fh_kl) |
| Line 71: | Line 67: |
| #calculate correlation coefficient | #calculate correlation coefficient |
| Line 75: | Line 71: |
| #plot ASAR image class together with freeboard heights | #plot ASAR image class together with freeboard heights |
| Line 96: | Line 92: |
| '''Ziele für Mittwoch: ''' * Vom gestern erzeugte Graphen verfeinen * Programm für ASAR-Rohdaten & Freiborddaten Vergleich |
die Besprechung ergab fuer diese Gruppe:
- Trockenprogrammierung: insbesondere die Konsistenz von Bildkoordinaten von ASAR und Freibord-Footprint klären die Klasseneinteilungen von ASAR und Freibordhöhen vergleichen Frage "welche Statistik will man anwenden" beantworten
Am Dienstag erzeugtes Programm:
1 from pylab import *
2 from scipy import *
3 import time,calendar,os,pipes,struct,string
4 import random
5 import scipy.ndimage as ndi
6
7 nk=3 #number of classes
8 ndim=[100,100] #dimension of ASAR image in pixels
9
10 fx0,fy0,fx2,fy2=[20.,100.,100.,10.] # freeboard coordinates from input file
11 nf=11 #number of freeboard footprints
12 fz_h=zeros(nf) #fz_h is the freeboard heights
13 # produce a random array of freeboard heights
14 for i in range(nf):
15 fz_h[i]=random.random()*70
16
17 # the x,y positions of the freeboard footprints
18 if fx2==fx0:
19 fx=zeros(nf)+fx0
20 fy=linspace(0,ndim[1],nf)
21 else:
22 teta=(fy2-fy0)/(fx2-fx0)
23 fx=linspace(fx0,fx2,nf)
24 fy=(fx-fx0)*teta+fy0
25 #produce a random matrix of ASAR image classes (in 3 classes)
26
27 a=zeros((ndim[0],ndim[1]))
28 for i in range(ndim[0]):
29 for j in range(ndim[1]):
30 a[i,j]=int(random.random()*(nk)+1)
31
32 #tranform from float to int
33
34 #A=ndimage.gaussian_filter(a,1)+0.5
35 #B=A.astype(int)
36 B=a.astype(int)
37
38 #select the class of footprints from ASAR Image
39 fz=zeros(nf)
40 for i in range(nf):
41 fz[i]=B[int(fx[i]-1),int(fy[i]-1)]
42
43 #calculation of mean and standard deviation of freeboard heights in each class
44 mean=zeros(nk)
45 std_kl=zeros(nk)
46 for kl in (1,2,3):
47 s=(fz==kl)
48 fh_kl=fz_h[s]
49 mean[kl-1]=fh_kl.sum()/fh_kl.shape[0]
50 std_kl[kl-1]=std(fh_kl)
51
52 #plot mean and error bar
53 kl=[1,2,3]
54 bar(kl,mean,yerr=std_kl,ecolor='r',align='center')
55
56
57 #calculate correlation coefficient
58 #corr=xcorr(fz,fz_h,normed=True)[1][nf]
59
60 figure()
61 #plot ASAR image class together with freeboard heights
62 x=arange(nf)
63 bar(x,fz,align='center')
64 ylim(0,6)
65 ax=twinx()
66 plot(x,fz_h,'r+-')
67 xlim(-2,12)
68
69 #plot ASAR image with freeboard flight
70 figure()
71 plot(fx,fy,'r+-',linewidth=2)
72 xlim(0,100)
73 ylim(0,100)
74 colorbar()
75 imshow(B,origin='lower',interpolation=None)
76 hold(True)
77
78 #figure()
79 #imshow(a)
80 show()
Ziele für Mittwoch:
- Vom gestern erzeugte Graphen verfeinen
Programm für ASAR-Rohdaten & Freiborddaten Vergleich