Mohamed Alkanhal and B. V. K. Vijaya Kumar, "Polynomial distance classifier correlation filter for pattern recognition," Appl. Opt. 42, 4688-4708 (2003)
We introduce what is to our knowledge a new nonlinear shift-invariant classifier called the polynomial distance classifier correlation filter (PDCCF). The underlying theory extends the original linear distance classifier correlation filter [Appl. Opt. 35, 3127 (1996)] to include nonlinear functions of the input pattern. This new filter provides a framework (for combining different classification filters) that takes advantage of the individual filter strengths. In this new filter design, all filters are optimized jointly. We demonstrate the advantage of the new PDCCF method using simulated and real multi-class synthetic aperture radar images.
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
When trained on all training images of serial numbers BMP2 (c21), BTR70 (c71), and T72 (s7). The filter was tested on all serial numbers.
Table 5
Confusion Matrix Showing the Results of the PDCCFa
Images
BMP2
BTR70
T72
BMP2 (sn-9563)
194
0
1
BMP2 (sn-9566)
193
2
1
BMP2 (sn-c21)
195
0
1
BTR70 (sn-c71)
0
196
0
T72 (sn-132)
7
0
189
T72 (sn-812)
0
0
195
T72 (sn-s7)
0
0
191
With the best ϕ’s from Table
3 when trained on all training images of serial numbers BMP2 (c21), BTR70 (c71) and T72 (s7). The filter was tested on all serial numbers.
When trained on all training images of serial numbers BMP2 (c21), BTR70 (c71), and T72 (s7). The filter was tested on all serial numbers. Every third from the training set was selected to design the filters.
Table 8
Confusion Matrix Showing the Results of the PDCCFa
Images
BMP2
BTR70
T72
BMP2 (sn-9563)
178
8
9
BMP2 (sn-9566)
174
9
13
BMP2 (sn-c21)
188
0
8
BTR70 (sn-c71)
0
196
0
T72 (sn-132)
13
0
183
T72 (sn-812)
8
1
186
T72 (sn-s7)
1
0
190
With the best ϕ’s from Table
6 when trained on all training images of serial numbers BMP2 (c21), BTR70 (c71), and T72 (s7). The filter was tested on all serial numbers. Every third from the training set was selected to design the filters.
When trained on all training images of serial numbers BMP2 (c21), BTR70 (c71), and T72 (s7). The filter was tested on all serial numbers.
Table 5
Confusion Matrix Showing the Results of the PDCCFa
Images
BMP2
BTR70
T72
BMP2 (sn-9563)
194
0
1
BMP2 (sn-9566)
193
2
1
BMP2 (sn-c21)
195
0
1
BTR70 (sn-c71)
0
196
0
T72 (sn-132)
7
0
189
T72 (sn-812)
0
0
195
T72 (sn-s7)
0
0
191
With the best ϕ’s from Table
3 when trained on all training images of serial numbers BMP2 (c21), BTR70 (c71) and T72 (s7). The filter was tested on all serial numbers.
When trained on all training images of serial numbers BMP2 (c21), BTR70 (c71), and T72 (s7). The filter was tested on all serial numbers. Every third from the training set was selected to design the filters.
Table 8
Confusion Matrix Showing the Results of the PDCCFa
Images
BMP2
BTR70
T72
BMP2 (sn-9563)
178
8
9
BMP2 (sn-9566)
174
9
13
BMP2 (sn-c21)
188
0
8
BTR70 (sn-c71)
0
196
0
T72 (sn-132)
13
0
183
T72 (sn-812)
8
1
186
T72 (sn-s7)
1
0
190
With the best ϕ’s from Table
6 when trained on all training images of serial numbers BMP2 (c21), BTR70 (c71), and T72 (s7). The filter was tested on all serial numbers. Every third from the training set was selected to design the filters.