Mercurial > hg > nsaunier > traffic-intelligence

comparison scripts/learn-motion-patterns.py @ 734:1d4dcb5c8708 dev

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
first example script to learn prototypes
author Nicolas Saunier <nicolas.saunier@polymtl.ca>
date Tue, 11 Aug 2015 12:55:09 -0400
parents
children 0e875a7f5759
comparison
equal deleted inserted replaced
733:c35e4a4b199d 734:1d4dcb5c8708
1 #! /usr/bin/env python
2
3 import sys, argparse
4
5 #import matplotlib.pyplot as plt
6 import numpy as np
7
8 import ml, utils, storage
9
10 parser = argparse.ArgumentParser(description='The program learns prototypes for the motion patterns') #, epilog = ''
11 #parser.add_argument('--cfg', dest = 'configFilename', help = 'name of the configuration file')
12 parser.add_argument('-d', dest = 'databaseFilename', help = 'name of the Sqlite database file', required = True)
13 parser.add_argument('-t', dest = 'trajectoryType', help = 'type of trajectories to display', choices = ['objectfeatures', 'feature', 'object'], default = 'objectfeatures')
14 parser.add_argument('-n', dest = 'nTrajectories', help = 'number of the object or feature trajectories to load', type = int, default = None)
15 parser.add_argument('-e', dest = 'epsilon', help = 'distance for the similarity of trajectory points', type = float, required = True)
16 parser.add_argument('--metric', dest = 'metric', help = 'metric for the similarity of trajectory points', default = 'cityblock') # default is manhattan distance
17 parser.add_argument('-s', dest = 'minSimilarity', help = 'minimum similarity to put a trajectory in a cluster', type = float, required = True)
18 parser.add_argument('-c', dest = 'minClusterSize', help = 'minimum cluster size', type = int, default = None)
19 parser.add_argument('--display', dest = 'display', help = 'display trajectories', action = 'store_true') # default is manhattan distance
20
21 args = parser.parse_args()
22
23 # TODO parameters (random init?) and what to learn from: objects, features, longest features from objects
24 # TODO add possibility to cluter with velocities
25
26 trajectoryType = args.trajectoryType
27 if args.trajectoryType == 'objectfeatures':
28 trajectoryType = 'object'
29
30 #features = storage.loadTrajectoriesFromSqlite(databaseFilename, args.trajectoryType)
31 objects = storage.loadTrajectoriesFromSqlite(args.databaseFilename, trajectoryType, withFeatures = (args.trajectoryType == 'objectfeatures'), objectNumbers = args.nTrajectories)
32
33 if args.trajectoryType == 'objectfeatures':
34 features = []
35 for o in objects:
36 tmp = utils.sortByLength(o.getFeatures(), reverse = True)
37 features += tmp[:min(len(tmp), 3)]
38 objects = features
39
40 trajectories = [o.getPositions().asArray().T for o in objects]
41
42 lcss = utils.LCSS(metric = args.metric, epsilon = args.epsilon)
43 nTrajectories = len(trajectories)
44
45 similarities = np.zeros((nTrajectories, nTrajectories))
46 for i in xrange(nTrajectories):
47 for j in xrange(i):
48 similarities[i,j] = lcss.computeNormalized(trajectories[i], trajectories[j])
49 similarities[j,i] = similarities[i,j]
50
51 prototypeIndices, labels = ml.prototypeCluster(trajectories, similarities, args.minSimilarity, args.minClusterSize)
52
53 if args.display:
54 for i,o in enumerate(objects):
55 if i not in prototypeIndices:
56 o.plot(utils.colors[labels[i]])
57 for i in prototypeIndices:
58 objects[i].plot(utils.colors[i]+'o')
59
60 # TODO store the prototypes (if features, easy, if objects, info must be stored about the type)