peaks-similarity

[![build status][travis-image]][travis-url] [![David deps][david-image]][david-url]

Peaks similarity - calculate a similarity between 0 and 1 of 2 arrays of [x,y] points.

Description

The idea is to calculate a similarity value between 0 and 1 between two arrays of peaks ([x,y] points).

const peaks1 = [
  [1, 1],
  [2, 2],
];
const peaks2 = [
  [1, 2],
  [2, 1],
];

import Comparator from 'peaks-similarity';
const comparator = new Comparator({ widthBottom: 4, widthTop: 2 });
const result = comparator.getSimilarity(peaks1, peaks2);

console.log(result);

options

• widthBottom : bottom width of the trapezoid (default: 2)
• widthTop : top width of the trapezoid (default: 1)
• trapezoid : Should we use overlapping of trapezoids surface (default: false). If trapezoid is false it is based only on peaks and not surface.
• common: Only take the common part of the array based on the widthBottom of the trapezoid (default: false)
  • true: take the common part between the 2 array
  • first: keep first array and take the common part of the second one
  • second: keep second array and take the common part of the first one

Methods

• setOptions(newOptions)
• setPeaks1(anArray)
• setPeaks2(anArray)
• getExtract1()
• getExtract2()
• getExtractInfo1()
• getExtractInfo2()
• setTrapezoid(newWidthBottom, newWidthTop)
• setFromTo(newFrom, newTo)
• getOverlap(x1, y1, x2, y2)
• getOverlapTrapezoid(x1, y1, x2, y2)
• calculateDiff()
• checkArray(points)
• getSimilarity(newPeaks1, newPeaks2)
• fastSimilarity(newPeaks2, from, to)
  • newPeaks2 has to be normalized (sum to 1)
  • fastSimilarity will use the common parameter. This parameter may be 'second' and will take the peaks1 that are at a distance less than widthBottom of any peak of peaks2

License

MIT