Analytical Feature Map

Aim: generate raster-based representations of GPS trajectories by aggregating analytical features over a regular grid.

Importing Tracklib

# Import the Tracklib library
import tracklib as tkl

Code running in a no shapely environment

Loading a collection of tracks

tkl.ObsTime.setReadFormat("2D/2M/4Y 2h:2m:2s")

chemin = '/home/md_vandamme/tracklib/trajlib/reseau/data2/'
fmt = tkl.TrackFormat({"ext": 'CSV',
    "id_E": 1,
    "id_N": 0,
    "id_U": 3,
    "id_T": 2,
    "h": 1,
    "separator": ";",
    "srid": "ENU"})

collection = tkl.TrackSource(chemin, fmt)

total = len(collection)
print ('Number of tracks: ' + str(total))

Number of tracks: 8654

Zoom in on a small area around a lake

We extracte tracks abiding by geometrical shape (a Rectangle) constraint are cut and returned in the next stage below:

xMin = 657336
yMin = 6860189
xMax = 657890
yMax = 6860625

ll = tkl.ENUCoords(xMin, yMin)
ur = tkl.ENUCoords(xMax, yMax)
bbox = tkl.Rectangle(ll, ur)

constraintBBox = tkl.Constraint(shape = bbox, mode = tkl.MODE_INSIDE, type=tkl.TYPE_CUT_AND_SELECT)
lakeCollection = constraintBBox.select(collection)
print ('Number of tracks near the lake: ' + str(lakeCollection.size()))
lakeCollection.plot('g-')

Number of tracks near the lake: 8654

Crowd density Map

The first step is to create a Raster, which defines the spatial framework of the analysis through its spatial extent and cell resolution. In this example, we build a raster with a 2-meter resolution, aligned on the bounding box of the trajectory collection and without any margin.

margin = 0.05
resolution = (2, 2)

raster = tkl.Raster(bbox=lakeCollection.bbox(), resolution=resolution, margin=margin,
                align=tkl.BBOX_ALIGN_CENTER)

Once the raster has been created, Analytical Feature Maps (AFMaps) can be added. Each AFMap is associated with a single Analytical Feature (AF) (e.g. uid, speed, or altitude) and may contain one or several Bands, each computing a different statistic over the raster cells (e.g. CountBand, CountDistinctBand, MeanBand, MedianBand, DominantBand).

In this example, we compute a crowd density map by counting the number of distinct trajectories crossing each raster cell. The selected Analytical Feature is the trajectory identifier uid, and the statistic is computed using a CountDistinctBand.

uidMap = raster.addAFMap("uid")
uidMap.addCountDistinct()

<tracklib.core.raster.CountDistinctBand at 0x7469bdb7b4f0>

Before rasterization, trajectories are spatially resampled at 1-meter intervals to ensure that every crossed cell is taken into account during the accumulation process.

cpt = 1
for trace in lakeCollection:
    trace.uid = cpt
    trace.resample(1, mode=1)
    cpt += 1

The rasterization is then performed by accumulating the trajectory observations into the raster cells and computing the requested statistics.

tkl.summarize(lakeCollection, raster)
print ('Raster created.')

Raster created.

The resulting grid can be retrieved from the corresponding AFMap and displayed.

grid = raster.getAFMap('uid')['count_distinct']

grid.plot(cmap='jet')

Write frequency of trajectories grid to Ascii Raster file

raster.writeToAscFile(grid, '/home/md_vandamme/zone_lac_2m.asc')
print ('Grid written in file.')

Grid written in file.