×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
Mapping without the sun
Author
Zhang, Jixian

153
Table 4. Residue of tie points-comer
The image coordinates of gravity center tie points are listed in
table 5. These tie points are used to register the PAN image and
the MS image. The residual is listed in table 6. The average
residual is 1.1 m.
No
MS
PAN
Column
Row
Column
Row
1
93
506
655
738
2
99
509
681
750
3
91
497
648
702
4
96
500
669
714
5
101
503
689
726
6
94
493
659
687
7
95
490
665
672
8
101
493
687
685
9
106
496
707
696
10
1704
815
7088
1978
11
1639
830
6827
2037
12
1673
843
6965
2088
13
1819
387
7545
263
14
1819
456
7547
539
15
2705
553
11084
930
16
2730
570
11186
999
17
2752
1243
11272
3691
18
2561
1690
10509
5477
19
1335
1778
5614
5827
20
1134
1771
4812
5800
21
1130
1799
4797
5911
22
1091
1747
4639
5706
23
106
793
707
1886
24
98
787
676
1859
25
152
776
892
1817
26
135
770
822
1796
27
239
515
1237
772
28
651
557
2884
943
29
632
553
2807
926
30
1423
460
5966
555
31
1414
461
5930
562
32
1397
454
5861
533
33
2903
934
11873
2455
34
2905
933
11882
2453
Table 5. Tie points-gravity center
No
DX (m)
DY (m)
DH (m)
Lon (°)
Lain
H (m)
1
0.614062
0.099311
0.000898
66.5038
45.84216
12.23018
2
-1.23508
0.065404
-0.00722
66.5036
45.84209
12.48353
3
-0.27595
0.091463
-0.00295
66.5038
' 45.84237
11.75083
4
-1.21428
0.07187
-0.0067
66.5037
45.8423
11.97572
5
-1.2488
0.065249
-0.0069
66.5035
45.84223
12.20407
6
0.606496
0.741034
-0.01208
66.5037
45.84246
11.59733
7
-1.20592
-1.20093
0.01889
66.5037
45.84253
11.44491
8
0.559579
-0.54401
0.013131
66.5035
45.84246
11.67833
9
0.525857
-1.18904
0.026498
66.5033
45.8424
11.89462
10
-0.4202
0.983144
-0.00463
66.4516
45.83573
28.89208
11
0.921315
0.416142
-0.00265
66.4537
45.83536
29.39594
12
-1.11013
-0.28598
0.002354
66.4526
45.83508
29.36405
13
1.399646
-0.90938
0.016064
66.4478
45.84565
15.48529
14
-0.38934
-0.94789
0.010687
66.4478
45.84406
17.661
15
-0.06905
0.08487
-0.00031
66.4191
45.84219
27.80031
16
-2.02742
0.676073
0.001319
66.4183
45.84181
27.84866
17
-0.11628
0.443255
-0.01782
66.4177
45.8263
44.73466
18
0.43671
-0.92311
-0.01952
66.4239
45.81594
58.05384
19
0.506976
-1.0099
0.000351
66.4636
45.81337
29.73939
20
0.04247
-0.17446
-0.00049
66.4701
45.81344
27.52945
21
-0.82709
-0.84148
-0.00469
66.4702
45.8128
27.26041
22
1.228146
1.179983
0.006289
66.4715
45.81397
27.44878
23
0.529867
-0.0066
0.007545
66.5033
45.83556
22.20843
24
-0.32284
-1.9317
-0.00016
66.5036
45.8357
21.77115
25
-0.6946
-0.71485
-0.00695
66.5018
45.83598
24.42901
26
1.23475
1.259045
0.012387
66.5024
45.8361
23.5107
27
1.417718
-1.31574
0.044876
-66.499
45.84203
14.94287
28
-0.51233
0.14652
-0.00337
66.4857
45.84124
20.88123
29
0.520512
-0.45821
0.012267
66.4863
45.84132
20.69716
30
-0.39113
-0.5687
0.010798
66.4606
45.8438
17.90377
31
-0.32515
1.358252
-0.0225
66.4609
45.84376
18.05702
32
0.689816
0.749536
-0.01364
66.4615
45.84392
17.88244
33
1.446471
0.451103
0.0109
66.4128
45.83348
37.87234
34
0.536908
1.712487
-0.03244
66.4127
45.8335
37.93841
Mean error=l. 118235 m
Table 6. Residue of tie points-gravity center
4. ANALYSIS AND CONCLUSION
Compare the average residual of comer and the average residual
of gravity center, it is not difficult to draw a conclusion that the
gravity center tie points are more suitable for registration of
different resolution images. The problem is that, as analysis in
the introduction, most of the automatic interest point extraction
algorithms can not extract gravity center points. Therefore, how
to extract gravity center points is our future work.
REFERENCE
Baker, S., Nayar, S.K., and Murase, H. 1998. Parametric feature
detection. International Journal of Computer Vision, 27(1 ):27-
50.
Beaudet, P.R. 1978. Rotationally invariant image operators. In
Proceedings of the 4th International Joint Conference on
Pattern Recognition, Tokyo, pp. 579-583.
Beus H.L. and Tiu S.S.H. 1987. An improved comer detection
algorithm based on chain-coded plane curves. Pattern
Recognition, 20:291-296.
Chen C.H., Lee J.S. and Sun Y. N., 1995, Wavelet
transformation for grey-level comer detection, 28(6), pp853-
861, Pattern Recognition, 1995.