Local Binary Patterns for JavaScript

Pattern detection using Local Binary Patterns algorithm

http://en.wikipedia.org/wiki/Local_binary_patterns

Installation

npm install lbp

or

// put library in your javascript root directory
<script src="{js directory}/lbp/lib/LBP.js"></script>
<script>
    var lbp = new LBP();
    ...
</script> 

Examples

// image data (two dimensional array of pixel values in gray scale)
var data = [
    [6,5,2, ...],
    [7,6,1, ...],
    [9,8,7, ...],
    ...
];
var lbp = new LBP(data, 8, 1); // 8 is number of sampling points, 1 is radius
 
// calculate LBP, contrast and variance for pixel at position [1,1]
console.log(lbp.calculate(1, 1));
 
// calculate LBP distribution, contrast and variance for whole image
console.log(lbp.distribution());

Rotation invariant binary patterns

By default, library is using rotation independent binary patterns. It means, all following patterns will be rotated to one uniform pattern:

1100 0000 -> 0000 0011
0000 1100 -> 0000 0011
1000 0001 -> 0000 0011

If you want to desable this behaviour, please set LBP.rotationInvariantLBPs to false.

Rotation invariant histogram

TODO

Memory

For better performance library uses local memory for storing calculation results. If you want to use your own memory adapter, it should implement following interface:

MyAdapter.prototype.set = function(key, value) {};
MyAdapter.prototype.get = function(key) {}; // returns null if key is not found
 
// assign custom memory adapter
LBP.memory = new MyAdapter();

Known issues

• Does not support rotation invariant histogram (yet)
• Need more testing with patterns different from (8,1)
• Contrast and variance values should be quantized

If you'd like to contribute, open an issue ticket or send pull request.

Test

mocha -R spec