#! /usr/bin/env python3
import os
import sys
import glob
from trafficintelligence import storage, moving
import subprocess
import numpy as np
def loadParameters(filename):
# load initial parameters from x.txt
f = open(filename, 'r+')
l = f.readline()
x = [s for s in l.strip().split(" ")]
f.close()
# create para-value list
return paraValueList(x)
def paraValueList(x):
#create para-value list
#list of the parameters and their values
p = ['--feature-quality', #]0.-0.4]
'--min-feature-distanceklt', #]0.-10]
'--block-size', #[1-10]integer
'--window-size', #[1-10]integer
'--min-feature-displacement', #[0.0001-0.1] 0.05
'--acceleration-bound', #[1.-5.] 3
'--deviation-bound', #[-1, 1] 0.6
#p[3] = '--min-tracking-error' #[0.01-0.3]
'--min-feature-time', #[2-100] integer
'--mm-connection-distance', #[0.5-100]
'--mm-segmentation-distance', #[1-100] ~mm-connection-distance / 2.5
'--min-nfeatures-group'] #[2-4]
integerParameters = [2, 3, 7]
para = []
if len(x) == 4:
for n in range(4):
if n+7 in integerParameters:
value = x[n].split('.')[0] #int(np.floor(float(x[n]))) recast to str
else:
value = x[n]
para = para + [p[-4+n],value]
else:
for n in range(len(x)):
if n in integerParameters:
value = x[n].split('.')[0]
else:
value = x[n]
para = para + [p[n], value]
return para
def process(para, intersections, recursive):
Mota = []
gtDatabaseaAbsPaths = []
configFileAbsPaths = []
cwd = os.getcwd()
# move to the location of the intersection
for intersectionPath in intersections:
intersectionAbsPath = os.path.abspath(intersectionPath)
os.chdir(intersectionAbsPath)
# iterate through all the subdirectories to find ground truth sqlite files
newPaths = [os.path.abspath(fn) for fn in glob.glob(intersectionAbsPath+'/*_gt.sqlite', recursive=recursive)]
gtDatabaseaAbsPaths.extend(newPaths)
configFilename = os.path.abspath(glob.glob(intersectionAbsPath+'/*.cfg', recursive=recursive)[0])
configFileAbsPaths.extend([configFilename]*len(newPaths))
os.chdir(cwd)
for gtDatabaseAbsPath, configFileAbsPath in zip(gtDatabaseaAbsPaths, configFileAbsPaths):
gtDatabaseBasename = gtDatabaseAbsPath[:-10]
videoFilename = gtDatabaseBasename + ".avi" # careful, it may be necessary to check video type / extension
if not os.path.exists(videoFilename):
print('Video file {} does not exist'.format(videoFilename))
databaseFilename = gtDatabaseBasename + ".sqlite"
gtDatabaseDirname = os.path.dirname(gtDatabaseAbsPath)
homographyFilename = gtDatabaseDirname + "/homography.txt"
maskFilename = gtDatabaseDirname + "/mask.png"
# Skip feature tracking if the user specified to optimize only grouping parameters
if len(para) > 8:
# Track features
trackingFeature(para, configFileAbsPath, videoFilename, databaseFilename, homographyFilename, maskFilename)
# Group features
groupFeature(para, configFileAbsPath, videoFilename, databaseFilename, homographyFilename, maskFilename)
#load trajectory
objects = storage.loadTrajectoriesFromSqlite(databaseFilename, 'object')
#load ground truth
annotations = storage.loadTrajectoriesFromSqlite(gtDatabaseAbsPath, 'object')
# Appending negative mota because nomad minimizes the output
matchingDistance = 5
inter = moving.TimeInterval.unionIntervals([a.getTimeInterval() for a in annotations])
motp, mota, mt, mme, fpt, gt, gtMatches, toMatches = moving.computeClearMOT(annotations, objects, matchingDistance, inter.first, inter.last)
Mota.append(-mota)
# Change to the previous directory
os.chdir(cwd)
return np.mean(Mota)
def trackingFeature(para, config, video, db, homo, mask):
# remove previous tracking
if os.path.exists(db):
os.remove(db)
# trackingfeature command parameters
tf = ['feature-based-tracking', config, '--tf', '--video-filename', video, '--database-filename', db, '--homography-filename', homo, '--mask-filename', mask]
# run in command line and print directly
subprocess.check_output(tf + para)
def groupFeature(para, config, video, db, homo, mask):
#remove previous grouping
storage.deleteFromSqlite(db, 'object')
#groupfeature command parameters
gf = ['feature-based-tracking', config, '--gf', '--video-filename', video, '--database-filename', db, '--homography-filename', homo, '--mask-filename', mask]
#run in command line and print directly
subprocess.check_output(gf + para) # ['--min-feature-time', 'x', '--mm-connection-distance', 'x', '--mm-segmentation-distance', 'x', '--min-nfeatures-group', 'x']
if __name__ == "__main__":
# Load args that were given with select-arguments.py
# with open('arguments.txt', 'r') as f:
# args = f.read().split('\n')
# intersections = args[0]
# optimizeGroupingOnly = eval(args[1])
# intersections = eval(intersections)
# Just write the intersections to optimize here: do not use '~' for home directory
intersections = ['/home/nicolas/Research/Data3/test-optimization/12-laurier']
recursive = False
# first and only argument should be the
if len(sys.argv) == 2:
para = loadParameters(sys.argv[1])
# run process including trackingfeature, groupfeature, load groundtruth, compute mota
print(process(para, intersections, recursive))
else:
print('Usage: site-parameters-optimization.py [initial-parameters text file]')
sys.exit(0)
