Simplifying location points

GPS tracks are collection of latitude, longitude, elevation, accuracy, timestamp etc. But some low-end devices can not provide accurate and reliable GPS tracks. Because of noisy and error prone GPS points, plotting these location tuples or analyzing them can be hard.

We need to clean and process gps tracks before plotting them . Here, we will try to create APIs useful for cleaning, filtering, smoothing, reducing number of points, interpolating etc. these location points.

For now we have distance and snap-to-road api using googlemaps api. Some of approach to reduce to reduce teh number of points can be:

1. Removing every nth point
   
2. Removes consecutive points within certain distance
   
3. Removes consecutive points more then certain distance in lesser time (google distance or haversine library)
   
4. RDP Algorithm (https://pypi.python.org/pypi/rdp or https://pypi.python.org/pypi/simplification/)
   
5. Remove points within a radius.
   

Some useful reads:
https://www.toptal.com/gis/adventures-in-gps-track-analytics-a-geospatial-primer
https://en.wikipedia.org/wiki/Ramer%E2%80%93Douglas%E2%80%93Peucker_algorithm
https://en.wikipedia.org/wiki/Kalman_filter
https://github.com/googlemaps/google-maps-services-python

Usage

Installation

pip install lokliner

Using on waypoints

locations =  [
    [12.9956927,77.7296046 ], [ 12.9940198,77.7269958 ], [ 12.9902847,77.7264299 ], [ 12.9901569,77.7242155 ],\
    [ 12.992211,77.7190483 ],[ 12.9921783,77.719092 ],[ 12.9871973,77.7090469 ],[ 12.985359,77.7081672 ],\
    [ 12.9888217,77.7022373 ],[ 12.9891085,77.6993243 ],[ 12.9888217,77.7022373 ],[ 12.9902847,77.7264299 ],\
    [ 12.9902847,77.7264299 ],[ 12.9888279,77.687049 ], [ 12.9885003,77.6858858 ],[ 12.9904586,77.6801263 ],\
    [ 12.9886829,77.6803986 ],[ 12.9885611,77.67955], [ 12.9885611,77.67955], [ 12.9885611,77.67955]
]

# Items in time_array should be equal to points in location array
# time_array is supposed to be time. Here we have taken numbers in increasing order

time_array = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20]

# Initialize Manipulation with google api key

from lokliner.smoother import Manipulation
sm = Manipulation(key='my_google_api_key')

# Calculate distance using haversine

distance_in_meters = sm.calculate_distance(point1=locations[0], point2=locations[1])

338.3744749682981

# calculate distance using google distance matrix api

distance_in_meters = sm.calculate_distance_using_google(locations[0], locations[1])

428

# Params: location, time_array, bound In meteres, miles False
# Returns: new_locations, new_locations_time, remove_locations, remove_locations_time

results = sm.remove_consecutive_same_location_using_google(locations,
                                            time_array,
                                            bound=2,
                                            miles=False
                                            )

([[12.9956927, 77.7296046], [12.9940198, 77.7269958], [12.9902847, 77.7264299], [12.9901569, 77.7242155], [12.992211, 77.7190483], [12.9921783, 77.719092], [12.9871973, 77.7090469], [12.985359, 77.7081672], [12.9888217, 77.7022373], [12.9891085, 77.6993243], [12.9888217, 77.7022373], [12.9902847, 77.7264299], [12.9888279, 77.687049], [12.9885003, 77.6858858], [12.9904586, 77.6801263], [12.9886829, 77.6803986], [12.9885611, 77.67955]], [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 14, 15, 16, 17, 18], [[12.9902847, 77.7264299], [12.9885611, 77.67955], [12.9885611, 77.67955]], [13, 19, 20])

from lokliner.snapper import Snapper
sm = Snapper(key='my_google_api_key')
result = sm.snap_using_google_api(
            locations=locations,
            time_array=time_array,
            per_call=50,interpolate=True
        )

Result of above snap: green are four original points. Blue line is Google's snapped and interpolated points.