File ColorSampling.str.pre
Version 1.0

This file contains functions that allow one to downsample from 4:4:4 to 4:2:0 or upsample from 4:2:0 or 4:2:2 to 4:4:4.

ChannelUpsample_Vert_and_Horz

int->int ChannelUpsample_Vert_and_Horz(int sourcewidth, int sourceheight)

Upsamples from 4:2:0 to 4:4:4

Input:

<int> - A 4:2:0 downsampled image

Output:

<int> - A 4:4:4 upsampled image

Parameters:

sourcewidth - The width of the downsampled image

sourceheight - The height of the downsampled image

ChannelUpsample_Vertical

int->int ChannelUpsample_Vertical(int sourcewidth, int sourceheight)

Upsample from 4:2:2 to 4:4:4 (in the vertical direction)

Input:

<int> - A 4:2:2 downsampled in the vertical direction

Output:

<int> - A 4:4:4 upsampled image

Parameters:

sourcewidth - The width of the downsampled image

sourceheight - The height of the downsampled image

ChannelUpsample_Horizontal

int->int ChannelUpsample_Horizontal(int sourcewidth, int sourceheight)

Upsample from 4:2:2 to 4:4:4 (in the horizontal direction)

Input:

<int> - A 4:2:2 downsampled in the horizontal direction

Output:

<int> - A 4:4:4 upsampled image

Parameters:

sourcewidth - The width of the downsampled image

sourceheight - The height of the downsampled image

ChannelUpsample_1D

int->int ChannelUpsample_1D

Performs a one dimensional upsampling, doubling the pixel count Each pixel is computed from a weighted average of its two neighbors with the closer neighbor getting weight1 and the further neighbor getting weight2.

Input:

<int> - The downsampled data stream

Output:

<int> - The reconstructed upsampled data stream

Parameters:

sourcelen - The length of the downsampled data stream

weight1 - The weight of the closer neighbor

weight2 - The weight of the further neighbor

DropSecond

int->int DropSecond

This drops every second value from a data stream, essentially performing a very simple downsampling (used for the horizontal downsampling in color channel downsampling.

Input:

<int> - A data stream

Output:

<int> - A downsampled data stream

ChannelDownsample_1D

int->int ChannelDownsample_1D(float weight1, float weight2)

This cuts a data stream in half by combining two pixels into one. Each new pixel's value is based on a weighting.

Input:

<int> - A data stream

Output:

<int> - A data stream

Parameters:

weight1 - The weighting used for the antecedent pixel

weight2 - The weighting used for the subsequent pixel

ChannelDownsample

int->int ChannelDownsample()

Downsample from 4:4:4 to 4:2:0

Input:

<int> - A 4:4:4 upsampled image

Output:

<int> - A 4:2:0 downsampled image