NVIDIA 2D Image And Signal Performance Primitives (NPP)
Version 11.8..*
|
Linear and non-linear image filtering functions. More...
Modules | |
1D Linear Filter | |
The set of 1D linear filtering functions available in the library. |
|
1D Window Sum | |
The set of 1D window sum functions available in the library. |
|
1D Window Sum with Border Control | |
The set of 1D window sum functions with border control available in the library. |
|
Convolution | |
The set convolution functions available in the library. |
|
2D Fixed Linear Filters | |
The set of 2D fixed linear filtering functions available in the library. |
|
Rank Filters | |
The set of functions providing min/max/median values for rectangular mask region with/without border available in the library. |
|
Fixed Filters | |
Fixed filters perform linear filtering operations (such as convolutions) with predefined kernels of fixed sizes. |
|
Computer Vision | |
The set of computer vision functions available in the library. |
|
Linear and non-linear image filtering functions.
Filtering functions are classified as Neighborhood Operations. It is the user's responsibility to avoid Sampling Beyond Image Boundaries.
These functions can be found in the nppif library. Linking to only the sub-libraries that you use can significantly save link time, application load time, and CUDA runtime startup time when using dynamic libraries.