IMAGE_DIFFUSE
Black/White Version of Grayscale Image

IMAGE_DIFFUSE is a MATLAB function which uses diffusion to smooth the colors in a picture.

Usage:

function g2 = image_diffuse ( g, c, k )

• g is an m by n uint8 (short integers, 0 to 255) array containing the grayscale image data; MATLAB's imread() command can be used to create such a dataset from a graphics image file.
• c is a value, typically between 0 and 1, which controls the diffusion. C = 0 means the picture is unchanged. C = 1 means each pixel is completely replaced by a neighborhood average.
• k indicates the number of steps of diffusion to carry out. After k steps, a single black pixel will be diffused with pixels up to k pixels away in all directions.
• g2 is an m by n uint8 array containing the diffused version of the image.

Licensing:

The computer code and data files described and made available on this web page are distributed under the GNU LGPL license.

Related Data and Programs:

IMAGE_CONTRAST, a MATLAB program which applies image processing techniques to increase the contrast in an image.

IMAGE_DENOISE, a MATLAB library which applies image processing techniques to remove noise from an image.

IMAGE_EDGE, a MATLAB program which demonstrates a simple technique for edge detection in an image.

IMAGE_NOISE, a MATLAB library which adds noise to an image.

IMAGE_QUANTIZATION, a MATLAB library which demonstrates how the KMEANS algorithm can be used to reduce the number of colors or shades of gray in an image.

IMAGE_RGB_TO_GRAY, a MATLAB function which creates a grayscale version of an RGB image.

IMAGE_THRESHOLD, a MATLAB library which makes a black and white version of a grayscale image by setting all pixels below or above a threshold value to black or white.

Reference:

MathWorks documentation for the Image Processing Toolbox is available at http://www.mathworks.com/access/helpdesk/help/pdf_doc/images/images_tb.pdf.

• Jonas Gomes, Luiz Velho,
  Image Processing for Computer Graphics,
  Springer, 1997,
  ISBN: 0387948546,
  LC: T385.G65.
• William Pratt,
  Digital Image Processing,
  Second Edition,
  Wiley, 1991,
  ISBN13: 978-0471857662,
  LC: TA1632.P7.

Source Code:

• image_diffuse.m, applies diffusion to a gray scale image.

Examples and Tests:

FOOL is an image which was originally a true black and white image; that is, as a gray scale image, every pixel was set to 0 or 255.

• fool.png, the original image.
• fool_c0.05_k20.png, the result for C = 0.05, K = 20;
• fool_c0.50_k01.png, the result for C = 0.50, K = 1.
• fool_c0.50_k20.png, the result for C = 0.50, K = 20.
• fool_c1.00_k01.png, the result for C = 1.00, K = 1

You can go up one level to the MATLAB source codes.