pnmconvol

Apply convolution filters to PNM images

TLDR

Apply convolution matrix

$ pnmconvol [matrix.pgm] [input.pnm] > [output.pnm]

Create blur effect

$ pnmconvol -matrix=[1,1,1;1,1,1;1,1,1] [input.pnm] > [output.pnm]

Edge detection

$ pnmconvol -matrix=[-1,-1,-1;-1,8,-1;-1,-1,-1] [input.pnm] > [output.pnm]

SYNOPSIS

pnmconvol [options] convolution input

pnmconvol applies convolution matrix filters to PNM images for effects like blurring, sharpening, and edge detection. The convolution kernel can be specified as a PGM file or inline with the -matrix option.
Each output pixel is computed as the weighted sum of surrounding input pixels according to the kernel values. Part of the Netpbm toolkit for image processing.

PARAMETERS

CONVOLUTION

Convolution matrix as a PGM file.

INPUT

Input PNM image file.

-matrix SPEC

Specify convolution matrix inline (e.g., "1,1,1;1,1,1;1,1,1").

-matrixfile FILE

Read convolution matrix from a file.

-normalize

Scale weights to sum to one, preserving overall image brightness.

-bias n

Amount to add to each convolved sample value, useful for handling negative results.

-nooffset

Normalize matrix values by dividing by maxval instead of using offset calculations.

CAVEATS

Part of Netpbm. Matrix defines effect. Slow for large kernels.

HISTORY

pnmconvol is part of Netpbm for convolution operations.