内に含ま
その他のドキュメント
サポート リソース
|  PDF 文書ファイルをダウンロードする
XIL Acceleration on SX
4
- This Appendix covers commonly-used XIL functions which have been accelerated on the SX for byte and short images. There are certain restrictions which have been placed on the functions in order for acceleration to occur. If the function is not accelerated on SX, it is performed using the memory driver of XIL.
- Two functions, xil_lookup(3) and xil_rotate(3), require contiguous memory (up to 128k bytes) for lookup, and the size of the source image for xil_rotate().
- The following functional features are supported for the XIL port to SX. All other parameters will cause an XIL_FAILURE to be returned, prompting the memory code to finish the operation.
- Note that rotate and lookup require SXDRAM (for which you'll need over 32MBytes on your machine.)
4.1 SX XIL Features
- Dyadic functions:
-
- 1,3,4 banded byte images.
- 3 banded child of 4 banded byte image
- child images with x,y offsets ok; no band offsets
-
-
XIL Function |
Byte |
Short | Byte
Child |
-
1 band
-
-
3 bands 4 bands n bands 3/4
xil_add x x x x x
xil_and x x x x x
xil_multiply x x x x x
xil_or x x x x x
xil_subtract x x x x x
xil_xor x x x x x
- Other functions:1
-
- band short images
- 1,3,4 band byte images
- 3-banded child of a 4-banded image
- child images w/ x,y offsets, no band offsets
- full support for ROIs
-
XIL Function |
Byte |
Short | Byte
Child |
-
1 band 3 bands
-
4 bands 1 band
-
3/4
-
|
|
|
|
|
|
| x | x | x | x | x |
| x | x | x | x | x |
| x | x | x | x | x |
| x | x | x | x | x |
| x | x | x | x | x |
| x | x | x | x | x |
| x | x | x | x | x |
| x | x | x | x | x |
-
-
xil_add_const
xil_and_const
xil_divide_const
xil_multiply_const
xil_not
xil_or_const
xil_subtract_const
xil_subtract_from_const
-
XIL Function |
Byte |
Short | Byte
Child |
-
1 band 3 bands
-
4 bands 1 band
-
3/4
-
-
xil_xor_const x x x x x
xil_blend x x x x x
xil_paint x x x o x
xil_scale x x x x x
xil_rescale x x x x x
xil_set_value x x x o x
xil_extrema x x x x x
xil_convolve (see Note 4 below) x x x x x
xil_rotate (Nearest Neighbor only, no
- x....x
- x....x
- x
- ROI)
-
xil_lookup (see Note 1 below)
-
xil_color_convert (see Note 2 below)
-
-
xil_copy (see Note 3 below) x x x x x
xil_threshold x o o x o
xil_transpose x x x x x
xil_translate x x x x x
xil_get_pixel x x x x x
xil_put_pixel x x x x x
-
xil_band_combine (see Note 5 below)....................x (3-band)
-
xil_decompress (see Note 6 below)
-
xil_cast (see Note 7 below)
-
- The following variations of xil_lookup are implemented.
-
| XIL Function | From-To | Bands |
|
| xil_lookup | 8-8 | 1 band only |
| xil_lookup | 16_16 | 1 band only |
| xil_lookup | 8_16 | 1 band only |
| xil_lookup | 16_8 | 1 band only |
| xil_lookup | 8-24 | 1 band to 3 bands |
-
- The following variations of xil_color_convert are implemented:
-
Source Colorspace | t
o |
Dest. Colorspace |
|
| rgblinear | to | rgb709 |
| rgblinear | to | ycc709 |
| rgblinear | to | ycc601 |
| rgblinear | to | ylinear |
| rgblinear | to | cmyk |
| ycc601 | to | rgb709 |
| rgb709 | to | rgblinear |
| rgb709 | to | ycc601 |
| rgb709 | to | photoycc |
| photoycc | to | rgb709 |
| cmyk | to | rgblinear |
-
- Copy function also allows the insertion and extraction of one band from 3/4 banded byte and 3-banded short images.
-
-
xil_convolve is implemented for the following kernels with central key pixels:
-
| XIL Function | Kernel |
|
| xil_convolve | 3x3 |
| xil_convolve | 5x5 |
| xil_convolve | 7x7 |
| xil_convolve | 3x1/1x3 (molecule) |
| xil_convolve | 5x1/1x5 (molecule) |
| xil_convolve | 7x1/1x7 (molecule) |
-
-
xil_band_combine has been implemented for 3-banded short images.
-
xil_decompress has been implemented for JPEG.
-
xil_cast has been implemented for conversion between 3-banded byte and 3-banded short images.
4.2 Molecules
- The following molecules are implemented:
-
| Function | Bands (bytes only) |
|
| 1 3 4 |
|
|
| xil_copy+display | x x x |
| xil_rotate+display | x x x |
| xil_scale+display | x x x |
| xil_set_value+display | x x x |
| xil_translate+display | x x x |
| xil_transpose+display | x x x |
| xil_convolve+xil_convolv e | x x x (for separable convolution) |
|
|
