This repository is archived and no longer maintained.

FivekoGFX (Fiveko Graphics) is a tiny Image Processing Library. The main code base of FivekoGFX is written in JavaScript and OpenGL shaders (GLSL). FivekoGFX is a web-based image processing lib and it tries to provide optimal solutions and take advantage of the GPU power (where possible).

You are also wlcome to visit fiveko.com.

• Convolution operator - 2D masks and 1D separable kernels

• Gaussian blur

• Sobel edge detection

• Mean filter

• Symmetric Nearest Neighbour filter (Symmetric NN)

• Hough transform

• Watershed trasnform (currently Meyer's flooding algorithm)

• Color space conversions (e.g. RGB <-> YCbCr, RGB to Grayscale, RGB to HSL, etc)

• Color tracking by Back Projection

• Histogram equalization

• Cartesian plane to Log-Polar image transform

• Local Binary Patterns (LBP)

• Major image files like: JPEG, PNG, WEBP, BMP, etc.
• Major Video files
• Camera support (have in mind the web browser capabilites)

The project provide an internal demo/test application you can use to test and see most fivekogfx's features.

Download the whole fivekogfx or just the /src/fivekogfx.min.js and include it into your HTML5 proect e.g.:

<script src="fivekogfx.min.js"></script>

Example of Gaussian filter over a canvas image with standart deviation of 2.0

var fivekogfx = new FivekoGFX();
fivekogfx.load(canvas);
fivekogfx.gauss(2.0); // e.g. Sigma=2.0
fivekogfx.draw(canvas);

Example of Discrete Gaussian blur approximation with standard deviation of 1.0

var fivekogfx = new FivekoGFX();
fivekogfx.load(canvas);
fivekogfx.conv1d([1, 4, 7, 4, 1]); // Blur filter
fivekogfx.draw(canvas);

Example of RGB to Grayscale conversion

var fivekogfx = new FivekoGFX();
fivekogfx.load(canvas);
fivekogfx.rgb2gray();
fivekogfx.draw(canvas);