Match probe data to roads by distance and bearing
npm install probematch
probematch creates an rbush index of a road network and allows you to match individual probe points to the roads by a combo of distance and bearing. Both distance and bearing filters are configurable.
var probematch = require('probematch');
var roads = {
'type': 'FeatureCollection'
'features': [
// Linestring features representing the road network
]
};
var match = probematch(roads, {
compareBearing: true, // should bearing be used to filter matches?
maxProbeDistance: 0.01, // distance filter in kilometers for probe matching
maxBearingRange: 5, // maximum bearing difference, in degrees, allowed for match filtering,
bidirectionalBearing: false // allows matching probes to roads oriented in the opposite direction.
// good if you know your data contains many 2-way roads
});
var probe = /* a point feature representing a probe */;
var probeBearing = /* a bearing in degrees */
var results = match(probe, probeBearing);probematch can also match all points in a line automatically, including bearing calculation between segments. This is ideal when working with data already assembled into traces.
var probematch = require('probematch');
var roads = //;
var match = probematch(roads, { /* options */});
var results = match.matchLine(line);When matching lines, the output is an array of the match results for each coordinate of the line, in the order of the original line coordinates.
probematch outputs an array of potential matches, sorted by distance from the probe.
[
{
"line": /* the road geometry that matched */,
"segment": /* the segment of road that matched */,
"distance":0.004666787054452857
},
{
"segment": /* ... */,
"line": /* ... */,
"distance":0.005917207788475857
}
]probematch performs best with the smallest road network possible loaded into it's index. Ideally, it is used on a tile-per-tile basis (for example, as part of a tile-reduce job)
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent