Types of Input Data

Intel IPP operations are divided into groups according to the data types on which the operation is performed. by supporting a variety of data types and layouts for each function, the Intel IPP library delivers a rich set of options for developers to choose from when designing and optimizing an application. Each group has its own prefix in the function name (see Function Naming and Parameters). The input data types are:

One-Dimensional Arrays and Signals

This group includes most functions operating on one-dimensional arrays of data. Generally these arrays represent signals, and many of the operations are signal-processing operations. Examples of one-dimensional array operations include:

• vectorized scalar arithmetic, logical, statistical operations

• digital signal processing

• data compression

• audio processing and audio coding

• speech coding

• cryptography and data integrity

• string operations

Images

An image is a two-dimensional array of pixel values. Examples of image operations include:

• arithmetic, logical, statistical operations

• color conversion

• image filtering

• image linear and geometric transformations

• morphological operations

• computer vision

• image compression

• video coding

Many primitive functions operating on image data work in a region of interest (ROI). Several primitives also assume that edge pixels exist - simply running them on a full image will produce an error. These images may be processed with an adjusted anchor point and ROI. For more information please see the KB article Processing an Image from Edge to Edge available at http://software.intel.com/en-us/articles/intel-integrated-performance-primitives-intel-ipp-processing-an-image-from-edge-to-edge.

Matrices

This group includes functions operating on matrices and vectors that are one- and two-dimensional arrays, and on arrays of matrices and vectors. These arrays are treated as linear equations or data vectors and subjected to linear algebra operations. Examples of matrix operations include:

• vector and matrix algebra

• solving systems of linear equations

• solving least squares problem

• computing eigenvalue problem

3D objects

This group includes functions operating with 3D objects. In this case input data depends on the used techniques. Examples of 3D operations include:

• realistic rendering

  Note

  Rendering functionality is deprecated in Intel® IPP 7.1.

• resizing and affine transforming

  Note

  This functionality is deprecated in Intel® IPP 7.1.