A simple implementation of polyline encoding algorithm in Java.

Google Maps Platform Doc: Encoded Polyline Algorithm Format.
My blog(简体中文): 轨迹压缩算法(Polyline encoding algorithm)探究

1. Create an object trjCompressor:

TrjCompressor trjCompressor = new TrjCompressor(6);

The parameter of the constructor is the precision of this compressor. For example, if the parameter is 6, then only 6 decimal places are reserved for floating point numbers.

2. Encode List<PointTs> to a single string:

String encoding = trjCompressor.encode(points); //type of points is List<PointTs>

3. Decode the compressed code to List<PointTs>:

List<PointTs> decoding = trjCompressor.decode(trjCode) //type of trjCode is String

Since it's a compression algorithm, I did a simple experiment to calculate the compression rate(compared to json format).
The trajectory data of the experiment is downloaded from Microsoft GeoLife and 171 trajectories are used. The code of this experiment is in the file TrjCompressorTest.java, you can change the value of path which contains the trajectory data from GeoLife if you want to do the similar experiment. Here is the result of the experiment:

mean of compression rate: 10.811463326040565 %
median of compression rate: 10.643147359032803 %

The mean and median of compression rates is about 10% in this experiment. It can be concluded that, polyline encoding algorithm is an excellent compression algorithm for trajectory data.