MimTransform
| Board | Supported |
|---|---|
| Host System | Yes |
| V4L2 | Yes |
| Clarity UHD | Yes |
| Concord PoE | No |
| GenTL | Yes |
| GevIQ | Yes |
| GigE Vision | Yes |
| Indio | No |
| Iris GTX | Yes |
| Radient eV-CL | Yes |
| Rapixo CL | Yes |
| Rapixo CoF | Yes |
| Rapixo CXP | Yes |
| USB3 Vision | Yes |
Perform a Fast Fourier transform (FFT), a Discrete Cosine transform (DCT), or a polar-cartesian coordinates transformation.
Syntax
void MimTransform(
AIL_ID SrcImageRBufId, //in
AIL_ID SrcImageIBufId, //in
AIL_ID DstImageRBufId, //out
AIL_ID DstImageIBufId, //out
AIL_INT64 TransformType, //in
AIL_INT64 ControlFlag //in
)
Description
This function performs a forward or reverse FFT or DCT transformation on an image. It also perfoms a cartesian-to-polar or polar-to-cartesian point-to-point coordinates transformation.
Parameters
SrcImageRBufId (in, AIL_ID)
Specifies the identifier of the source buffer for the real component of the image.
SrcImageIBufId (in, AIL_ID)
Specifies the identifier of the source buffer for the imaginary component of the image.
DstImageRBufId (out, AIL_ID)
Specifies the identifier of the destination buffer for the real component of the image.
DstImageIBufId (out, AIL_ID)
Specifies the identifier of the destination buffer for the imaginary component of the image.
TransformType (in, AIL_INT64)
Specifies the type of transform to perform on the image.
ControlFlag (in, AIL_INT64)
Specifies if the transform is a forward transform or a reverse transform.
Parameter Associations
For selecting the transformation type.
M_DCT8X8
Board availability: Clarity UHD, GenTL, GevIQ, GigE Vision, Host System, Iris GTX, Radient eV-CL, Rapixo CL, Rapixo CXP, Rapixo CoF, USB3 Vision, V4L2
Performs a discrete cosine transform on each 8x8 pixel block in the image.
M_DEFAULT
Same as M_FORWARD.
M_FORWARD
Performs a forward transform on the image.
| Value | Description |
|---|---|
8-bit signed/unsigned image buffer ID | Performs calculations in fixed-point format and returns to the destinations in 23.9 fixed point format. |
16-bit signed/unsigned image buffer ID | Performs calculations in fixed-point format and returns to the destinations in 25.7 fixed point format. |
32-bit signed image buffer ID | Conserves the fixed point format between the destination buffer and the source buffer. |
Floating-point image buffer ID | Specifies that processing will be performed in floating-point arithmetic. The destination buffer should also be a floating-point buffer. |
M_REVERSE
Performs a reverse transform on the image.
| Value | Description |
|---|---|
8-bit signed/unsigned image buffer ID | The format of the source buffer is assumed to be in 23.9 fixed point format. |
16-bit signed/unsigned image buffer ID | The format of the source buffer is assumed to be in 25.7 fixed point format. |
32-bit signed image buffer ID | The fixed point format of the destination buffer is the same as that of the source. |
Floating-point image buffer ID | If SrcImageRBufId is a floating-point buffer, this must also be a floating-point buffer and processing will be performed in floating-point arithmetic. |
M_FFT
Performs a Fast Fourier transform.
M_DEFAULT
Same as M_FORWARD.
M_FORWARD
Performs a forward transform on the image. For floating-point destination buffers, the processing will be performed in floating-point arithmetic.
| Value | Description |
|---|---|
8-bit signed/unsigned image buffer ID | Calculations are performed in fixed-point format and returned to the destinations in 23.9 fixed point format. |
16-bit signed/unsigned image buffer ID | Calculations are performed in fixed-point format and returned to the destinations in 25.7 fixed point format. |
32-bit signed image buffer ID | The fixed point format of the destination buffer will be the same as that of the source buffer. |
Floating-point image buffer ID | The destination buffer should also be a floating-point buffer and processing will be performed in floating-point arithmetic. |
M_NULL | Specifies that a faster version of the forward transform will be performed. |
Image buffer ID | Specifies the identifier of the source buffer. |
M_REVERSE
Performs a reverse transform on the image.
| Value | Description |
|---|---|
8-bit signed/unsigned image buffer ID | The format of the source buffer is assumed to be in 23.9 fixed point format. |
16-bit signed/unsigned image buffer ID | The format of the source buffer is assumed to be in 25.7 fixed point format. |
32-bit signed image buffer ID | The fixed point format of the destination buffer is the same as that of the source. |
Floating-point image buffer ID | If SrcImageRBufId is a floating-point buffer, this must also be a floating-point buffer and processing will be performed in floating-point arithmetic. |
M_NULL | Specifies that a faster version of the reverse transform will be performed. |
Image buffer ID | Specifies the identifier of the destination buffer. |
M_POLAR
Board availability: Clarity UHD, GenTL, GevIQ, GigE Vision, Host System, Iris GTX, Radient eV-CL, Rapixo CL, Rapixo CXP, Rapixo CoF, USB3 Vision, V4L2
Performs a cartesian-to-polar or a polar-to-cartesian transform on the coordinates of the source buffer.
M_DEFAULT
Same as M_FORWARD.
M_FORWARD
Converts the real and imaginary components of cartesian coordinates to the magnitude and phase components of polar coordinates.
| Value | Description |
|---|---|
M_NULL | Specifies that a faster version of the forward transform will be performed. > Note: Only one of the destination buffers can be set to M_NULL. |
Image buffer ID | Specifies the identifier of the destination buffer. It must be a 32-bit floating-point buffer. |
M_NULL | Specifies that a faster version of the forward transform will be performed. > Note: Only one of the destination buffers can be set to M_NULL. |
Image buffer ID | Specifies the identifier of the destination buffer. The phase results are comprised in the range of 0-360. |
M_REVERSE
Converts the magnitude and phase components of polar coordinates to the real and imaginary components of cartesian coordinates.
Combination Constants — For the ControlFlag parameter
Optional.
Usage: You can add one or more of the following values to the above-mentioned values to set the transform characteristics.
Note that only M_NORMALIZE is supported with both M_FFT and M_DCT8X8. All other values are supported only with M_FFT.
| Value | Description |
|---|---|
M_1D_COLUMNS | Performs a 1D transform on all columns of the image. If M_1D_COLUMNS has been specified with the forward transform, then it must be specified for the reverse transform as well. |
M_1D_ROWS | Performs a 1D transform on all rows of the image. If M_1D_ROWS has been specified with the forward transform, then it must be specified for the reverse transform as well. |
M_CENTER | Centers the real and imaginary components of the image in the frequency domain by repositioning their respective buffers' quadrants. The DC component is positioned at (SizeX/2-1, SizeY/2-1), where SizeX and SizeY are the width and height of the source buffer, respectively. [Image: FFT_and_FFT_with_M_CENTER.png] If the size of the destination buffers is different from that of the source buffer, data that falls outside of the destination buffers is clipped; exceeding areas of the destination buffers are unaffected. See Managing data buffersfor more information. If M_CENTER has been specified for the forward transform, then it must be specified for the reverse transform as well. |
M_MAGNITUDE | Specifies whether to compute or use the magnitude. When used with M_FORWARD, computes the magnitude of the forward FFT (the square root of _R_ <sup>2</sup>+_I_ <sup>2</sup>). The magnitude is returned in the DstImageRBufId destination buffer, thereby overwriting the real component of the destination image. When used with M_REVERSE, computes the real component of the transform from the magnitude and phase (_R_ = _M_ cos(_P_)). The result will be used to perform the reverse FFT. For a reverse FFT, SrcImageRBufId and SrcImageIBufId must contain the magnitude and phase of a forward transform, respectively. M_MAGNITUDE must be used with M_PHASE for a reverse FFT. For a forward FFT, DstImageIBufId can be set to M_NULL if M_MAGNITUDE is used without M_PHASE. M_MAGNITUDE can only be used in transforms processed with floating-point buffers. |
M_NORMALIZE | Normalizes results (divide the final result by 8 for DCT and by (m x n) for FFT where m x _n _is the size of the image). M_NORMALIZE must be used with fixed-point arithmetic (all integer buffers) to avoid overflows. If M_NORMALIZE has been specified with the forward transform, then it must be specified for the reverse transform as well. |
M_PHASE | Specifies whether to compute or use the phase. When used with M_FORWARD, computes the phase of the forward transform (_P_ = atan(_I_/_R_)). The phase is returned in the DstImageIBufId buffer, thereby overwriting the imaginary component of the destination image. The phase is returned in the range of 0° to 360°. When used with M_REVERSE, computes the imaginary component of the transform from the magnitude and phase (_I_ = _M_ sin(_P_)). The result will be used to perform the reverse FFT. For a reverse FFT, SrcImageRBufId and SrcImageIBufId must contain the magnitude and phase of a forward transform, respectively. M_PHASE must be used with M_MAGNITUDE for a reverse FFT. For a forward FFT, DstImageRBufId can be set to M_NULL if M_PHASE is used without M_MAGNITUDE. M_PHASE can only be used in transforms processed with floating-point buffers. |
M_SATURATION | Clips (saturates) the results of a reverse FFT according to the destination buffer's data type. M_SATURATION can only be used with a reverse FFT (M_FFT+M_REVERSE) operation which uses floating-point source buffers and integer destination buffers. |
Combination Constants — For
Optional.
Usage: You can add one of the following values to the above-mentioned values to specify that the magnitude of the forward FFT is scaled logarithmically.
| Value | Description |
|---|---|
M_LOG_SCALE | Scales logarithmically the magnitude of the forward FFT to be in the range of 0-255. This value is used to scale the spectrum into a displayable range. M_LOG_SCALE can only be used in forward transforms processed with floating-point buffers. [Clarity UHD, GenTL, GevIQ, GigE Vision, Host System, Iris GTX, Radient eV-CL, Rapixo CL, Rapixo CXP, Rapixo CoF, USB3 Vision, V4L2] |
Combination Constants — For specifying the transform characteristics
Optional.
Usage: You can add one or more of the following values to the above-mentioned values to specify the transform characteristics.
Board availability: Clarity UHD, GenTL, GevIQ, GigE Vision, Host System, Iris GTX, Radient eV-CL, Rapixo CL, Rapixo CXP, Rapixo CoF, USB3 Vision, V4L2
| Value | Description |
|---|---|
M_FAST_PHASE | Performs faster phase computation. For a forward polar transformation, it uses a faster computation of the phase results but slightly less accurate. For a reverse polar transformation, it has no effect. |
M_NORMALIZE_PHASE | Rescales the phase results. For a forward polar transformation, it rescales the phase results to be in the range of 0.0-1.0. For a reverse polar transformation, SrcImageRBufId must contain the phase values in the range of 0.0-1.0. This value cannot be used in combination with M_NORMALIZE_PHASE_255. |
M_NORMALIZE_PHASE_255 | Rescales the phase results. For a forward polar conversion, it rescales the phase results to be in the range of 0-255. For a reverse polar transformation, SrcImageIBufId must contain the phase values in the range of 0-255. This value cannot be used in combination with M_NORMALIZE_PHASE. |
M_SQUARE_MAGNITUDE | Specifies that the magnitude values are returned or used squared. For a forward polar transformation, it gives the square of the magnitude results (that is, _R_ <sup>2</sup>+_I_ <sup>2</sup>). For a reverse transformation, SrcImageRBufId must contain the magnitude values squared. |