Difference between revisions of "Function:FindTransformECC"

From CUVI Wiki
 
(2 intermediate revisions by the same user not shown)
Line 5: Line 5:
|style="font-size:100%;"|
|style="font-size:100%;"|
<syntaxhighlight lang="cpp">
<syntaxhighlight lang="cpp">
CuviStatus cuvi::computerVision::findTransformECC(CuviImage& templateImage, CuviImage& inputImage, Cuvi64f t[2][3], Cuvi32s iterations = 50, Cuvi64f threshold = 0.001, const CuviStream& stream = CuviStream());
CuviStatus cuvi::computerVision::findTransformECC(CuviImage& templateImage,  
                                                  CuviImage& inputImage,
                                                  Cuvi64f t[2][3],
                                                  Cuvi32s iterations = 50,
                                                  Cuvi64f threshold = 0.001,
                                                  const CuviStream& stream = CuviStream());
</syntaxhighlight>
</syntaxhighlight>
|}
====Parameters====
{|
|style="font-size:75%;"|
{|class="wikitable"
|-
! Name
! Type
! Description
|-
| templateImage
| CuviImage&
| Reference image
|-
| inputImage
| CuviImage&
| Image to be alignment
|-
| t
| Cuvi64f[][]
| resultant warp matrix
|-
| iterations
| Cuvi32s
| number of iterations to diverge the result
|-
| threshold
| Cuvi64f
| threshold criterion
|-
| stream
| const CuviStream&
| GPU stream ID for execution
|}
|}
====Image Type Support====
{|
|style="font-size:75%;"|
{| class="wikitable"
|-
! Input
! Output
|-
| 8uC1
| 8uC1
|}
|}
|}

Latest revision as of 20:00, 19 October 2022

Finds the geometric transform (warp) between two images in terms of the ECC criterion and generates a floating-point 2×3 mapping matrix. The matrix can then be fed to warpAffineBack to align the two images.

Function

CuviStatus cuvi::computerVision::findTransformECC(CuviImage& templateImage, 
                                                  CuviImage& inputImage,
                                                  Cuvi64f t[2][3],
                                                  Cuvi32s iterations = 50,
                                                  Cuvi64f threshold = 0.001,
                                                  const CuviStream& stream = CuviStream());

Parameters

Name Type Description
templateImage CuviImage& Reference image
inputImage CuviImage& Image to be alignment
t Cuvi64f[][] resultant warp matrix
iterations Cuvi32s number of iterations to diverge the result
threshold Cuvi64f threshold criterion
stream const CuviStream& GPU stream ID for execution

Image Type Support

Input Output
8uC1 8uC1