def __init__(self, corner_a, corner_b, corner_c, corner_d):
rng = cv.RNG()
self.midx = rect_midpointx(corner_a, corner_b, corner_c, corner_d)
self.midy = rect_midpointy(corner_a, corner_b, corner_c, corner_d)
self.corner1 = corner_a
self.corner2 = corner_b
self.corner3 = corner_c
self.corner4 = corner_d
# locx and locy are relative locations of corners when compared to other corners of the same rectangle
self.corner1_locx = self.corner1[0] - self.corner2[0]
self.corner1_locy = self.corner1[1] - self.corner2[1]
self.corner2_locx = self.corner2[0] - self.corner1[0]
self.corner2_locy = self.corner2[1] - self.corner1[1]
self.corner3_locx = self.corner3[0] - self.corner1[0]
self.corner3_locy = self.corner3[1] - self.corner1[1]
self.corner4_locx = self.corner4[0] - self.corner1[0]
self.corner4_locy = self.corner4[1] - self.corner1[1]
# angle is found according to short side
if line_distance(corner_a, corner_c) < line_distance(corner_a, corner_b):
self.angle = -angle_between_lines(line_slope(corner_a, corner_c), 0)
else:
self.angle = -angle_between_lines(line_slope(corner_a, corner_b), 0)
self.id = 0 # id identifies which pizza your looking at
self.last_seen = 2 # how recently you have seen this pizza
self.seencount = 1 # how many times you have seen this pizza (if you see it enough it becomes confirmed)
r = int(cv.RandReal(rng) * 255)
g = int(cv.RandReal(rng) * 255)
b = int(cv.RandReal(rng) * 255)
self.debug_color = cv.RGB(r, g, b)
self.object = random.choice(["A", "B", "C", "D"])
def addGaussianNoise(image_path, save_path):
img = cv.LoadImage(image_path)
noise = cv.CreateImage(cv.GetSize(img), img.depth, img.nChannels)
cv.SetZero(noise)
rng = cv.RNG(-1)
cv.RandArr(rng, noise, cv.CV_RAND_NORMAL, cv.ScalarAll(0),
cv.ScalarAll(25))
cv.Add(img, noise, img)
tempName = os.path.splitext(
os.path.basename(image_path))[0] + "_noised.jpg"
save_image = os.path.join(save_path, tempName)
cv.SaveImage(save_image, img)
def init(self):
# Thresholds
self.minsize = 50
self.maxsize = 300
# List we store found buoys in
self.new = []
if self.debug:
# windows
#cv.NamedWindow("Buoy" + self.camera_name)
# random number generator used for choosing debug colors
self.rng = cv.RNG()
def init(self):
# Total Number of Buoys Found
self.buoy_count = 0
# Average disparity across cameras
self.avg_disparity = 0
# Buoy Classes
self.newLeft = []
self.newRight = []
self.candidates = []
self.confirmed = []
self.lost = []
if self.debug:
# random number generator used for choosing debug colors
self.rng = cv.RNG()
def noisy(image, noise_typ, filename):
if noise_typ == "gauss":
row, col = image.shape
mean = 0
deviation = 30
#var = 0.1
#sigma = var**0.5
gauss = np.random.normal(mean, deviation, (row, col))
gauss = gauss.reshape(row, col)
noisy = image + gauss
misc.imsave(filename, noisy)
elif noise_typ == "s&p":
row, col = image.shape
s_vs_p = 0.5
amount = 0.01
out = image
# Salt mode
num_salt = np.ceil(amount * image.size * s_vs_p)
coords = [
np.random.randint(0, i - 1, int(num_salt)) for i in image.shape
]
out[coords] = 255
# Pepper mode
num_pepper = np.ceil(amount * image.size * (1. - s_vs_p))
coords = [
np.random.randint(0, i - 1, int(num_pepper)) for i in image.shape
]
out[coords] = 0
misc.imsave(filename, out)
elif noise_typ == "speckle":
mult_noise = cv.CreateImage((image.width, image.height),
cv.IPL_DEPTH_32F, 1)
cv.RandArr(cv.RNG(6), mult_noise, cv.CV_RAND_NORMAL, 1, 0.1)
cv.Mul(image, mult_noise, image)
cv.SaveImage(filename, image)
elif noise_typ == "blur":
blur = ndimage.gaussian_filter(image, sigma=3)
misc.imsave(filename, blur)
def init(self):
# Total Number of Buoys Found
self.buoy_count = 0
# Thresholds
self.minsize = 20
self.maxsize = 40
# Buoy Classes
self.new = []
self.candidates = []
self.confirmed = []
self.lost = []
#self.debug = True
if self.debug:
# windows
# cv.NamedWindow("BuoyTest")
# random number generator used for choosing debug colors
self.rng = cv.RNG()
Pressing ESC will stop the program.
"""
import urllib2
import cv
from math import cos, sin, sqrt
import sys
if __name__ == "__main__":
A = [[1, 1], [0, 1]]
img = cv.CreateImage((500, 500), 8, 3)
kalman = cv.CreateKalman(2, 1, 0)
state = cv.CreateMat(2, 1, cv.CV_32FC1) # (phi, delta_phi)
process_noise = cv.CreateMat(2, 1, cv.CV_32FC1)
measurement = cv.CreateMat(1, 1, cv.CV_32FC1)
rng = cv.RNG(-1)
code = -1L
cv.Zero(measurement)
cv.NamedWindow("Kalman", 1)
while True:
cv.RandArr(rng, state, cv.CV_RAND_NORMAL, cv.RealScalar(0),
cv.RealScalar(0.1))
kalman.transition_matrix[0, 0] = 1
kalman.transition_matrix[0, 1] = 1
kalman.transition_matrix[1, 0] = 0
kalman.transition_matrix[1, 1] = 1
cv.SetIdentity(kalman.measurement_matrix, cv.RealScalar(1))
import cv
im = cv.LoadImage('melon.jpg', cv.CV_LOAD_IMAGE_GRAYSCALE)
mult_noise = cv.CreateImage((im.width, im.height), cv.IPL_DEPTH_32F, 1)
cv.RandArr(cv.RNG(6), mult_noise, cv.CV_RAND_NORMAL, 1, 0.1)
cv.Mul(im, mult_noise, im)
cv.ShowImage("tree with speckle noise", im)
cv.WaitKey(0)