Intel® IPP implements image processing functions that perform linear and non-linear filtering operations on an image.
Filtering can be used in a variety of image processing operations; for example, edge detection, blurring, noise removal, and feature detection.
Table Filtering Functions lists the Intel IPP filtering functions described in more detail later in this chapter:
Function Base Name | Operation |
---|---|
Blurs an image using a box filter. |
|
Sums pixel values in the row mask applied to the image. |
|
Sums pixel values in the column mask applied to the image. |
|
Filters an image using a min filter. |
|
Filters an image using a max filter. |
|
Computes the size of the working buffer for the minimum filter. |
|
Computes the size of the working buffer for the maximum filter. |
|
Filters an image using a min filter with border replication. |
|
Filters an image using a max filter with border replication. |
|
Computes the size of the buffer for the bilateral filter. |
|
Initializes the bilateral filter structure. |
|
Performs bilateral filtering of the image. |
|
Decimates an image using the specified filter. |
|
Median Filters |
|
Compute the size of the external buffer. |
|
Applies a median filter to an image. |
|
Filters an image using a median filter with a horizontal mask. |
|
Filters an image using a median filter with a vertical mask. |
|
Filters an image using a cross median filter. |
|
Filters an image using a median filter with weighted center pixel. |
|
Applies a color median filter to an image. |
|
General Linear Filters |
|
Filters an image using a general rectangular convolution kernel. |
|
Compute the size of working buffer for 64f general linear filter. |
|
Filters an image with integer data using a floating-point rectangular convolution kernel. |
|
Perform filtering an image and rounding the result. |
|
FilterRoundGetBufSize16s, FilterRoundGetBufSize32s, FilterRoundGetBufSize32f |
Computes the size of the working buffer for filters with rounding mode. |
Separable Filters |
|
Filters an image using an integer column convolution kernel. |
|
Filters an image using a floating-point column convolution kernel. |
|
Filters an image using an integer row convolution kernel. |
|
Filters an image using a floating-point row convolution kernel. |
|
FilterRowBorderPipelineGetBufferSize, FilterRowBorderPipelineGetBufferSize_Low |
Compute the size of working buffer for rows filter with border. |
FilterColumnPipelineGetBufferSize, FilterColumnPipelineGetBufferSize_Low |
Compute the size of working buffer for columns filter with border. |
Applies the filter with border to image rows. |
|
Applies the filter with border to image columns. |
|
Calculates the dot product of taps vector and columns of the specified set of rows. |
|
Wiener Filters |
|
Computes the size of the external buffer for ippiFilterWiener function. |
|
Filters an image using the Wiener algorithm. |
|
Convolution |
|
Performs full convolution of two images. |
|
Performs valid convolution of two images. |
|
Deconvolution |
|
Allocates and initializes state structure for FFT deconvolution. |
|
Frees memory allocated for the FFT deconvolution state structure. |
|
Performs FFT deconvolution of an image. |
|
Allocates and initializes state structure for LR deconvolution. |
|
Frees memory allocated for the LR deconvolution state structure. |
|
Performs LR deconvolution of an image. |
|
Fixed Filters |
|
Filters an image using a horizontal Prewitt operator. |
|
Filters an image using a vertical Prewitt operator. |
|
Filters an image using a horizontal Scharr operator. |
|
Filters an image using a vertical Scharr operator. |
|
Filters an image using a horizontal Sobel operator. |
|
Filters an image using a vertical Sobel operator. |
|
Filters an image using a second derivative horizontal Sobel operator. |
|
Filters an image using a second derivative vertical Sobel operator. |
|
Filters an image using a second cross derivative Sobel operator. |
|
Filters an image using a horizontal Roberts cross-gradient operator. |
|
Filters an image using a vertical Roberts cross-gradient operator. |
|
Filters an image using a Laplacian kernel. |
|
Filters an image using a Gaussian kernel. |
|
Applies a highpass filter to an image. |
|
Applies a lowpass filter to an image. |
|
Applies a sharpening filter to an image. |
|
Fixed Filters with Borders |
|
Computes the size of the external buffer for the horizontal Scharr filter with border. |
|
Computes the size of the external buffer for the vertical Scharr filter with border. |
|
Computes the size of the external buffer for the horizontal Sobel filter with border. |
|
FilterSobelVertGetBufferSize, FilterSobelNegVertGetBufferSize |
Computes the size of the external buffer for the vertical Sobel filter with border. |
Computes the size of the external buffer for the second derivative horizontal Sobel filter with border. |
|
Computes the size of the external buffer for the second derivative vertical Sobel filter with border. |
|
Computes the size of the external buffer for the cross Sobel filter with border. |
|
Computes the size of the external buffer for the Laplace filter with border. |
|
Computes the size of the external buffer for the Gaussian filter with border. |
|
Computes the size of the external buffer for the lowpass filter with border. |
|
Computes kernel for the Sobel filter to an image. |
|
Applies horizontal Scharr filter with border to an image. |
|
Applies vertical Scharr filter with border to an image. |
|
Applies horizontal Sobel filter with border to an image. |
|
Applies vertical Sobel filter with border to an image. |
|
Applies horizontal (second derivative) Sobel filter with border to an image. |
|
Applies vertical (second derivative) Sobel filter with border to an image. |
|
Applies second derivative cross Sobel filter with border to an image. |
|
Applies Laplacian filter with border to an image. |
|
Applies Gauss filter with border. |
|
Applies lowpass filter with border to an image. |
Most of filtering function operate with ROI (see Regions of Interest in Intel IPP). The size of the source image ROI is equal to dstRoiSize, the destination image ROI size. Most of the functions use different source and destination image buffers, that is, they are not in-place.
Copyright © 2000 - 2011, Intel Corporation. All rights reserved.