194
7: Hua-mei Chen, Pramod K. Varshney
Table 7.4. Registration results for Landsat TM 1995 and 1997 images. T denotes the transformation parameters (vertical and horizontal displacements)
Algorithm
95 ---+ 97 T95,97
97 ---+ 95 T97,95
2-date
[pixel,pixel]
[pixel,pixel]
consistency
Linear
[2.9109,62.4256]
[-3.0757, -62.4665]
0.1698
1st order
[3.0000,62.0000]
[-3.0000, -62.0000]
0.0000
2nd order
[2.9812,62.3717]
[-2.9819, -62.3739]
0.0024
3rd order
[2.9837,62.3907]
[-2.9838, -62.3930]
0.0023
MSD
[3.2801,62.4177]
[-3.0000, -62.3163]
0.2979
Nee
[3.1412,62.3719]
[-3.1039, -62.3604]
0.0390
Table 7.5. Registration results for Landsat TM 1995 and 1998 images. T denotes the transformation parameters (vertical and horizontal displacements)
Algorithm
95 ---+ 98 T95,98
98 ---+ 95 T98,95
2-date
[pixel,pixel]
[pixel,pixel]
consistency
Linear
[33.3183, -13.5464]
[-33.3238, 13.5576]
0.0125
1st order
[33.0001, -13.9990]
[-33.0000,14.0000]
0.0010
2nd order
[33.2304, -13.6165]
[-33.2307,13.6169]
0.0005
3rd order
[33.2504, -13.5969]
[-33.2522,13.5975]
0.0020
MSD
[33.3777, -13.6033]
[-33.3438, 13.6517]
0.0592
Nee
[33.3507, -13.6290]
[-33.3489, 13.6486]
0.0197
Table 7.6. Registration results for Landsat TM 1997 and 1998 images. T denotes the transformation parameters (vertical and horizontal displacements)
Algorithm
97 ---+ 98 T97•98
98 ---+ 97 T 98.97
2-date
[pixel,pixel]
[pixel,pixel]
consistency
Linear
[30.3784, -76.0752]
[-30.3745,76.0577]
0.0179
1st order
[30.0000, -76.0000]
[-30.0010,76.0003]
0.001l
2nd order
[30.2734, -75.9989]
[-30.2723,76.0115]
0.0127
3rd order
[30.2953, -75.9996]
[-30.2938,76.0000]
0.0015
MSD
[30.3637, -75.9647]
[-30.4245,75.7822]
0.1923
Nee
[30.3818, -75.8891]
[-30.3867,76.1127]
0.2237
transformation parameters (i. e. displacements in x and y directions) along
with the corresponding two-date registration consistency. Table 7.7 shows
the corresponding three-date registration consistencies of each algorithm to
register these images.
A glance at the first four rows of Table 7.4 to 7.6 shows no significant
difference between the performances of each algorithm as all of them perform
fairly well with the linear interpolation resulting in the worst registration
consistency. Further, on close inspection of transformation parameters for the
7: Hua-mei Chen, Pramod K. Varshney
Table 7.4. Registration results for Landsat TM 1995 and 1997 images. T denotes the transformation parameters (vertical and horizontal displacements)
Algorithm
95 ---+ 97 T95,97
97 ---+ 95 T97,95
2-date
[pixel,pixel]
[pixel,pixel]
consistency
Linear
[2.9109,62.4256]
[-3.0757, -62.4665]
0.1698
1st order
[3.0000,62.0000]
[-3.0000, -62.0000]
0.0000
2nd order
[2.9812,62.3717]
[-2.9819, -62.3739]
0.0024
3rd order
[2.9837,62.3907]
[-2.9838, -62.3930]
0.0023
MSD
[3.2801,62.4177]
[-3.0000, -62.3163]
0.2979
Nee
[3.1412,62.3719]
[-3.1039, -62.3604]
0.0390
Table 7.5. Registration results for Landsat TM 1995 and 1998 images. T denotes the transformation parameters (vertical and horizontal displacements)
Algorithm
95 ---+ 98 T95,98
98 ---+ 95 T98,95
2-date
[pixel,pixel]
[pixel,pixel]
consistency
Linear
[33.3183, -13.5464]
[-33.3238, 13.5576]
0.0125
1st order
[33.0001, -13.9990]
[-33.0000,14.0000]
0.0010
2nd order
[33.2304, -13.6165]
[-33.2307,13.6169]
0.0005
3rd order
[33.2504, -13.5969]
[-33.2522,13.5975]
0.0020
MSD
[33.3777, -13.6033]
[-33.3438, 13.6517]
0.0592
Nee
[33.3507, -13.6290]
[-33.3489, 13.6486]
0.0197
Table 7.6. Registration results for Landsat TM 1997 and 1998 images. T denotes the transformation parameters (vertical and horizontal displacements)
Algorithm
97 ---+ 98 T97•98
98 ---+ 97 T 98.97
2-date
[pixel,pixel]
[pixel,pixel]
consistency
Linear
[30.3784, -76.0752]
[-30.3745,76.0577]
0.0179
1st order
[30.0000, -76.0000]
[-30.0010,76.0003]
0.001l
2nd order
[30.2734, -75.9989]
[-30.2723,76.0115]
0.0127
3rd order
[30.2953, -75.9996]
[-30.2938,76.0000]
0.0015
MSD
[30.3637, -75.9647]
[-30.4245,75.7822]
0.1923
Nee
[30.3818, -75.8891]
[-30.3867,76.1127]
0.2237
transformation parameters (i. e. displacements in x and y directions) along
with the corresponding two-date registration consistency. Table 7.7 shows
the corresponding three-date registration consistencies of each algorithm to
register these images.
A glance at the first four rows of Table 7.4 to 7.6 shows no significant
difference between the performances of each algorithm as all of them perform
fairly well with the linear interpolation resulting in the worst registration
consistency. Further, on close inspection of transformation parameters for the