Mercurial Hosting > traffic-intelligence
changeset 672:5473b7460375
moved and rationalized imports in modules
| author | Nicolas Saunier <nicolas.saunier@polymtl.ca> |
|---|---|
| date | Tue, 26 May 2015 13:53:40 +0200 |
| parents | 849f5f8bf4b9 |
| children | 5505f9dbb28e |
| files | python/cvutils.py python/moving.py python/utils.py |
| diffstat | 3 files changed, 18 insertions(+), 49 deletions(-) [+] |
line wrap: on
line diff
--- a/python/cvutils.py Tue May 26 11:40:32 2015 +0200 +++ b/python/cvutils.py Tue May 26 13:53:40 2015 +0200 @@ -17,6 +17,7 @@ skimageAvailable = False from sys import stdout +from numpy import dot, array, append #import aggdraw # agg on top of PIL (antialiased drawing) @@ -67,15 +68,14 @@ def loadPointCorrespondences(filename): '''Loads and returns the corresponding points in world (first 2 lines) and image spaces (last 2 lines)''' - from numpy.lib.npyio import loadtxt - from numpy import float32 + from numpy import loadtxt, float32 points = loadtxt(filename, dtype=float32) return (points[:2,:].T, points[2:,:].T) # (world points, image points) def cvMatToArray(cvmat): '''Converts an OpenCV CvMat to numpy array.''' print('Deprecated, use new interface') - from numpy.core.multiarray import zeros + from numpy import zeros a = zeros((cvmat.rows, cvmat.cols))#array([[0.0]*cvmat.width]*cvmat.height) for i in xrange(cvmat.rows): for j in xrange(cvmat.cols): @@ -116,7 +116,7 @@ def computeUndistortMaps(width, height, undistortedImageMultiplication, intrinsicCameraMatrix, distortionCoefficients): from copy import deepcopy - from numpy import identity, array + from numpy import identity newImgSize = (int(round(width*undistortedImageMultiplication)), int(round(height*undistortedImageMultiplication))) newCameraMatrix = deepcopy(intrinsicCameraMatrix) newCameraMatrix[0,2] = newImgSize[0]/2. @@ -325,7 +325,6 @@ def computeHomographyFromPDTV(camera): '''Returns the homography matrix at ground level from PDTV camera https://bitbucket.org/hakanardo/pdtv''' - from numpy import array # camera = pdtv.load(cameraFilename) srcPoints = [[x,y] for x, y in zip([1.,2.,2.,1.],[1.,1.,2.,2.])] # need floats!! dstPoints = [] @@ -340,7 +339,7 @@ map1 and map2 are the mapping functions from undistorted image to distorted (original image) map1(x,y) = originalx, originaly''' - from numpy import abs, logical_and, unravel_index, dot, sum + from numpy import abs, logical_and, unravel_index, sum from matplotlib.mlab import find distx = abs(map1-x) disty = abs(map2-y) @@ -459,10 +458,6 @@ def projectArray(homography, points): '''Returns the coordinates of the projected points through homography (format: array 2xN points)''' - from numpy.core import dot - from numpy.core.multiarray import array - from numpy.lib.function_base import append - if points.shape[0] != 2: raise Exception('points of dimension {0} {1}'.format(points.shape[0], points.shape[1])) @@ -477,20 +472,18 @@ def project(homography, p): '''Returns the coordinates of the projection of the point p with coordinates p[0], p[1] through homography''' - from numpy import array return projectArray(homography, array([[p[0]],[p[1]]])) def projectTrajectory(homography, trajectory): '''Projects a series of points in the format [[x1, x2, ...], [y1, y2, ...]]''' - from numpy.core.multiarray import array return projectArray(homography, array(trajectory)) def invertHomography(homography): '''Returns an inverted homography Unnecessary for reprojection over camera image''' - from numpy.linalg.linalg import inv + from numpy import inv invH = inv(homography) invH /= invH[2,2] return invH @@ -517,7 +510,6 @@ return undistortedTrajectory def projectGInputPoints(homography, points): - from numpy import array return projectTrajectory(homography, array(points+[points[0]]).T) if opencvAvailable: @@ -527,9 +519,7 @@ img1Points are used to compute the translation TODO add diagnostic if data is all over the place, and it most likely is not a translation (eg zoom, other non linear distortion)''' - from numpy.core.multiarray import array - from numpy.lib.function_base import median - from numpy.core.fromnumeric import sum + from numpy import median, sum nextPoints = array([]) (img2Points, status, track_error) = cv2.calcOpticalFlowPyrLK(img1, img2, img1Points, nextPoints, winSize=windowSize, maxLevel=level, criteria=criteria) @@ -568,7 +558,7 @@ def createHOGTrainingSet(imageDirectory, classLabel, rescaleSize = (64, 64), orientations=9, pixelsPerCell=(8, 8), cellsPerBlock=(2, 2), visualize=False, normalize=False): from os import listdir - from numpy import array, float32 + from numpy import float32 from matplotlib.pyplot import imread inputData = []
--- a/python/moving.py Tue May 26 11:40:32 2015 +0200 +++ b/python/moving.py Tue May 26 13:53:40 2015 +0200 @@ -4,8 +4,11 @@ import utils, cvutils from base import VideoFilenameAddable -from math import sqrt -from numpy import median +from math import sqrt, atan2, cos, sin +from numpy import median, array, zeros, hypot, NaN, std +from matplotlib.pyplot import plot +from scipy.stats import scoreatpercentile +from scipy.spatial.distance import cdist try: from shapely.geometry import Polygon, Point as shapelyPoint @@ -199,7 +202,6 @@ return Point(self.x/alpha, self.y/alpha) def plot(self, options = 'o', **kwargs): - from matplotlib.pylab import plot plot([self.x], [self.y], options, **kwargs) def norm2Squared(self): @@ -230,7 +232,6 @@ return shapelyPoint(self.x, self.y) def project(self, homography): - from numpy.core.multiarray import array projected = cvutils.projectArray(homography, array([[self.x], [self.y]])) return Point(projected[0], projected[1]) @@ -295,7 +296,6 @@ The unknown of the equation is the time to reach the intersection between the relative trajectory of one road user and the circle of radius collisionThreshold around the other road user''' - from math import sqrt dv = v1-v2 dp = p1-p2 a = dv.norm2Squared()#(v1.x-v2.x)**2 + (v1.y-v2.y)**2 @@ -344,8 +344,6 @@ mode=1: cumulative distance mode=2: cumulative distance with trailing distance ''' - - from numpy import zeros ss_spline_d = [] #Prepare subsegment distances for spline in range(len(splines)): @@ -480,7 +478,6 @@ @staticmethod def fromPoint(p): - from math import atan2 norm = p.norm2() if norm > 0: angle = atan2(p.y, p.x) @@ -493,7 +490,6 @@ return NormAngle(max(self.norm+other.norm, 0), self.angle+other.angle) def getPoint(self): - from math import cos, sin return Point(self.norm*cos(self.angle), self.norm*sin(self.angle)) @@ -527,7 +523,6 @@ return FlowVector(self.position.multiply(alpha), self.velocity.multiply(alpha)) def plot(self, options = '', **kwargs): - from matplotlib.pylab import plot plot([self.position.x, self.position.x+self.velocity.x], [self.position.y, self.position.y+self.velocity.y], options, **kwargs) self.position.plot(options+'x', **kwargs) @@ -681,7 +676,6 @@ @staticmethod def _plot(positions, options = '', withOrigin = False, lastCoordinate = None, timeStep = 1, **kwargs): - from matplotlib.pylab import plot if lastCoordinate is None: plot(positions[0][::timeStep], positions[1][::timeStep], options, **kwargs) elif 0 <= lastCoordinate <= len(positions[0]): @@ -710,7 +704,6 @@ return self.positions[1] def asArray(self): - from numpy.core.multiarray import array return array(self.positions) def xBounds(self): @@ -757,7 +750,6 @@ # def add(x, y): return x+y # sq = map(add, [x*x for x in self.positions[0]], [y*y for y in self.positions[1]]) # return sqrt(sq) - from numpy import hypot return hypot(self.positions[0], self.positions[1]) # def cumulatedDisplacement(self): @@ -1083,8 +1075,6 @@ def plotCurvilinearPositions(self, lane = None, options = '', withOrigin = False, **kwargs): if hasattr(self, 'curvilinearPositions'): - from matplotlib.pylab import plot - from numpy import NaN if lane is None: plot(list(self.getTimeInterval()), self.curvilinearPositions.positions[0], options, **kwargs) if withOrigin: @@ -1179,13 +1169,11 @@ cvutils.displayTrajectories(videoFilename, [self], homography = homography, firstFrameNum = self.getFirstInstant(), lastFrameNumArg = self.getLastInstant(), undistort = undistort, intrinsicCameraMatrix = intrinsicCameraMatrix, distortionCoefficients = distortionCoefficients, undistortedImageMultiplication = undistortedImageMultiplication) def speedDiagnostics(self, framerate = 1., display = False): - from numpy import std - from scipy.stats import scoreatpercentile speeds = framerate*self.getSpeeds() coef = utils.linearRegression(range(len(speeds)), speeds) print('min/5th perc speed: {} / {}\nspeed diff: {}\nspeed stdev: {}\nregression: {}'.format(min(speeds), scoreatpercentile(speeds, 5), speeds[-2]-speeds[1], std(speeds), coef[0])) if display: - from matplotlib.pyplot import figure, plot, axis + from matplotlib.pyplot import figure, axis figure(1) self.plot() axis('equal') @@ -1214,7 +1202,6 @@ '''Returns the distances between all features of the 2 objects at the same instant instant1 or at instant1 and instant2''' - from scipy.spatial.distance import cdist if _instant2 is None: instant2 = instant1 else: @@ -1266,8 +1253,6 @@ #for i in xrange(int(obj1.length())-1): # for j in xrange(int(obj2.length())-1): # inter = segmentIntersection(obj1.getPositionAt(i), obj1.getPositionAt(i+1), obj2.getPositionAt(i), obj2.getPositionAt(i+1)) - from scipy.spatial.distance import cdist - from numpy import zeros positions1 = [p.astuple() for p in obj1.getPositions()] positions2 = [p.astuple() for p in obj2.getPositions()] pets = zeros((int(obj1.length()), int(obj2.length()))) @@ -1339,7 +1324,6 @@ Warning work in progress TODO? not use the first/last 1-.. positions''' - from numpy import array, median nFeatures = len(self.features) if nFeatures == 0: print('Empty object features\nCannot compute smooth trajectory') @@ -1409,7 +1393,6 @@ def classifyUserTypeHoGSVMAtInstant(self, img, pedBikeCarSVM, instant, homography, width, height, bikeCarSVM = None, pedBikeSpeedTreshold = float('Inf'), bikeCarSpeedThreshold = float('Inf'), px = 0.2, py = 0.2, pixelThreshold = 800): '''Extract the image box around the object and applies the SVM model on it''' - from numpy import array croppedImg, yCropMin, yCropMax, xCropMin, xCropMax = imageBox(img, self, instant, homography, width, height, px, py, pixelThreshold) if len(croppedImg) > 0: # != [] hog = array([cvutils.HOG(croppedImg)], dtype = np.float32)
--- a/python/utils.py Tue May 26 11:40:32 2015 +0200 +++ b/python/utils.py Tue May 26 13:53:40 2015 +0200 @@ -5,6 +5,7 @@ from datetime import time, datetime from math import sqrt, ceil, floor from scipy.stats import kruskal, shapiro +from numpy import zeros, array, exp, sum as npsum, arange, cumsum datetimeFormat = "%Y-%m-%d %H:%M:%S" @@ -60,7 +61,6 @@ def cumulativeDensityFunction(sample, normalized = False): '''Returns the cumulative density function of the sample of a random variable''' - from numpy import arange, cumsum xaxis = sorted(sample) counts = arange(1,len(sample)+1) # dtype = float if normalized: @@ -70,15 +70,13 @@ class EmpiricalDiscreteDistribution(EmpiricalDistribution): '''Class to represent a sample of a distribution for a discrete random variable ''' - from numpy.core.fromnumeric import sum - def __init__(self, categories, counts): self.categories = categories self.counts = counts def mean(self): result = [float(x*y) for x,y in zip(self.categories, self.counts)] - return sum(result)/self.nSamples() + return npsum(result)/self.nSamples() def var(self, mean = None): if not mean: @@ -87,12 +85,12 @@ m = mean result = 0. squares = [float((x-m)*(x-m)*y) for x,y in zip(self.categories, self.counts)] - return sum(squares)/(self.nSamples()-1) + return npsum(squares)/(self.nSamples()-1) def referenceCounts(self, probability): '''probability is a function that returns the probability of the random variable for the category values''' refProba = [probability(c) for c in self.categories] - refProba[-1] = 1-sum(refProba[:-1]) + refProba[-1] = 1-npsum(refProba[:-1]) refCounts = [r*self.nSamples() for r in refProba] return refCounts, refProba @@ -168,7 +166,6 @@ def kernelSmoothing(x, X, Y, weightFunc, halfwidth): '''Returns the smoothed estimate of (X,Y) at x Sum_x weight(sample_x,x) * y(x)''' - from numpy import zeros, array weights = array([weightFunc(x,observedx, halfwidth) for observedx in X]) if sum(weights)>0: return sum(weights*Y)/sum(weights) @@ -182,7 +179,6 @@ return 0. def gaussian(center, x, halfwidth): - from numpy import exp return exp(-((center-x)/halfwidth)**2/2) def epanechnikov(center, x, halfwidth):