def size_image(img, imgsize):
# check if we need to crop out ROI
roiWidth = img.width
roiHeight = img.height
if (img.width > imgsize[1]):
roiWidth = imgsize[1]
if (img.height > imgsize[0]):
roiHeight = imgsize[0]
roi = (0, 0, roiWidth, roiHeight)
cv.SetImageROI(img, roi)
imgTrim = cv.CreateImage((roi[2], roi[3]), img.depth, img.nChannels)
cv.Copy(img, imgTrim)
# check if we need to pad
padSize = 0
padSize = max(padSize, imgsize[0] - imgTrim.height)
padSize = max(padSize, imgsize[1] - imgTrim.width)
if padSize == 0: # no padding needed
return imgTrim
else:
padSize = int(round((padSize + .5) / 2.))
# copy make border
imgPad = cv.CreateImage(
(imgTrim.width + 2 * padSize, imgTrim.height + 2 * padSize),
img.depth, img.nChannels)
cv.CopyMakeBorder(imgTrim, imgPad, (0, 0), 0)
roi = (0, 0, imgsize[1], imgsize[0])
cv.SetImageROI(imgPad, roi)
imgFinal = cv.CreateImage((roi[2], roi[3]), img.depth, img.nChannels)
cv.Copy(imgPad, imgFinal)
return imgFinal
def publish_debug(img, results):
imgsize = cv.GetSize(img)
sizelist = [cv.GetSize(tmp[1]) for tmp in templates]
width = max(imgsize[0], sum([s[0] for s in sizelist]))
height = imgsize[1] + max([s[1] for s in sizelist])
output = cv.CreateImage((width, height), cv.IPL_DEPTH_8U, 1)
cv.Zero(output)
cur_x = 0
view_rect = (0, height-imgsize[1], imgsize[0], imgsize[1])
cv.SetImageROI(output, view_rect)
cv.Copy(img, output)
for template in templates:
size = cv.GetSize(template[1])
cv.SetImageROI(output, (cur_x, 0, size[0], size[1]))
cv.Copy(template[1], output)
cur_x += size[0]
#cv.PutText(output, tempname, (0,size[1]-16), font, cv.CV_RGB(255,255,255))
#cv.PutText(output, str(tempthre)+'<'+str(status[1]), (0,size[1]-8), font, cv.CV_RGB(255,255,255))
for _,status in [s for s in results if s[0] == template[3]]:
print status
cv.PutText(output, template[3], (0,size[1]-42), font, cv.CV_RGB(255,255,255))
cv.PutText(output, "%7.5f"%(status[0]), (0,size[1]-24), font, cv.CV_RGB(255,255,255))
cv.PutText(output, "%7.5f"%(status[2]), (0,size[1]-8), font, cv.CV_RGB(255,255,255))
if status[3] :
cv.Rectangle(output, (0, 0), size, cv.RGB(255,255,255), 9)
cv.SetImageROI(output, view_rect)
for _,status in [s for s in results if s[0] == template[3]]:
pt2 = (status[1][0]+size[0], status[1][1]+size[1])
if status[3] :
cv.Rectangle(output, status[1], pt2, cv.RGB(255,255,255), 5)
cv.ResetImageROI(output)
debug_pub.publish(bridge.cv_to_imgmsg(output, encoding="passthrough"))
def scale_32f_image(image):
'''
Scales the given cv.IPL_DEPTH_32F type image to an 8 bit image so that the
smallest value maps to 0 and the largest maps to 255. Used for displaying
debugging images.
Processes each channel separately, which can produce some useful, but
esoteric results.
'''
if image.depth != cv.IPL_DEPTH_32F:
return image
result = cv.CreateImage(cv.GetSize(image), 8, image.channels)
channel_image = cv.CreateImage(cv.GetSize(image), cv.IPL_DEPTH_32F, 1)
channel_scaled = cv.CreateImage(cv.GetSize(image), 8, 1)
for channel_num in xrange(1, image.channels + 1):
cv.SetImageCOI(image, channel_num)
cv.Copy(image, channel_image)
minmaxloc = cv.MinMaxLoc(channel_image)
minimum = minmaxloc[0]
maximum = minmaxloc[1]
if maximum - minimum > 0:
cv.ConvertScale(channel_image, channel_scaled,
255 / (maximum - minimum),
-255 / (maximum - minimum) * minimum)
else:
cv.ConvertScale(channel_image, channel_scaled, 0, -255 / minimum)
cv.SetImageCOI(result, channel_num)
cv.Copy(channel_scaled, result)
cv.SetImageCOI(image, 0)
cv.SetImageCOI(result, 0)
return result
def processMotion(self):
"""
Take a raw input image frame from the camera and perform motion detection using
the current frame plus considering several previous frames and return the CV
image that should be given to the Region network.
"""
#find motion image, then find feature corners from that
if self._prevIplImage:
cv.AbsDiff(self._inputImage, self._prevIplImage, self._diffImage)
else:
cv.Copy(self._inputImage, self._diffImage)
cv.Copy(self._inputImage, self._prevIplImage) #save as t-1 image for next frame
#(src, dest, threshold, maxVal, type)
cv.Threshold(self._diffImage, self._threshImage, 16.0, 255.0, cv.CV_THRESH_BINARY)
#For now, disable segmentMotion and return all motion in frame...
if self._threshImage!=None:
return (0,0, self._threshImage.width, self._threshImage.height)
###Experimental: segment motion to return only the most 'interesting' area
tsec = clock()
#(silhouette, mhi, timestamp, duration)
cv.UpdateMotionHistory(self._threshImage, self._historyImage, tsec, 0.3)
#(mhi, segMask, storage, timestamp, segThresh)
#return: [tuple(area, value, rect)], (float, CvScalar, CvRect)
seqs = cv.SegmentMotion(self._historyImage, self._segMaskImage, \
self._memStorage, tsec, 1.0)
#cv.Copy(self._threshImage, self._inputImage)
#cv.Threshold(self._segMaskImage, self._threshImage, 0.0, 250.0, CV_THRESH_BINARY)
rects = []
for seq in seqs:
seqRect = seq[2] #CvRect = tuple (x, y, width, height)
if(seqRect[2] > 4 and seqRect[3] > 4):
rects.append(seqRect)
#find the 3rd largest area and only keep those rects
if len(rects) > 0:
areas = [x[2]*x[3] for x in rects]
areas.sort()
minArea = areas[0]
if len(areas) > 1:
minArea = areas[len(areas)-1]
rectsOk = [x for x in rects if x[2]*x[3] >= minArea]
rect = rectsOk[0]
#center the largest rect
cRect = (rect[0]+(rect[2]/2), rect[1]+(rect[3]/2))
return rect
return None #none means no motion bounding box detected
def observation(self, meas):
meas = np.float32([meas])
if not self.initialized:
cv.Copy(cv.fromarray(meas.T.copy()), self.kf.state_post)
cv.Copy(cv.fromarray(meas.T.copy()), self.kf.state_pre)
self.initialized = True
if self.kf.CP == 0:
cv.KalmanPredict(self.kf)
corrected = np.asarray(
cv.KalmanCorrect(self.kf, cv.fromarray(meas.T.copy())))
return corrected.T[0].copy()
def capture_draw():
img = cv.QueryFrame(capture)
# scale your big ole face down to something small
thumb = cv.CreateMat(img.height / SCALE, img.width / SCALE, cv.CV_8UC3)
cv.Resize(img, thumb)
faces = get_face_roi(thumb)
for (x, y, w, h), n in faces:
temp_offset = (x * SCALE, y * SCALE)
cv.SetImageROI(img,
((x) * SCALE, (y) * SCALE, (w) * SCALE, (h) * SCALE))
roi_image = cv.CreateImage(cv.GetSize(img), img.depth, img.nChannels)
cv.Copy(img, roi_image)
cv.ResetImageROI(img)
cv.Rectangle(img, (x * SCALE, y * SCALE),
(x * SCALE + w * SCALE, y * SCALE + h * SCALE),
(255, 0, 0))
cv.PutText(img, 'face', (x * SCALE, y * SCALE), font, (200, 200, 200))
FEATURE_SCALE = (float(roi_image.width) / ROI_TARGET_SIZE[0],
float(roi_image.height) / ROI_TARGET_SIZE[1])
roi_thumb = cv.CreateImage((int(roi_image.width / FEATURE_SCALE[0]),
int(roi_image.height / FEATURE_SCALE[1])),
cv.IPL_DEPTH_8U, 3)
cv.Resize(roi_image, roi_thumb)
features = get_features(roi_thumb)
cv.ShowImage("ROI", roi_image)
for name in features:
if features[name] != None:
for (x1, y1, w1, h1), n1 in features[name]:
cv.SetImageROI(
roi_image,
(x1 * FEATURE_SCALE[0], y1 * FEATURE_SCALE[1],
w1 * FEATURE_SCALE[0], h1 * FEATURE_SCALE[1]))
feature_image = cv.CreateImage(cv.GetSize(roi_image),
roi_image.depth,
roi_image.nChannels)
cv.Copy(roi_image, feature_image)
cv.ResetImageROI(feature_image)
cv.ShowImage(name, feature_image)
cv.PutText(img, name,
(temp_offset[0] + x1 * FEATURE_SCALE[0],
temp_offset[1] + y1 * FEATURE_SCALE[1]), font,
(200, 200, 200))
cv.Rectangle(
img, (temp_offset[0] + x1 * FEATURE_SCALE[0],
temp_offset[1] + y1 * FEATURE_SCALE[1]),
(temp_offset[0] +
(x1 + w1) * FEATURE_SCALE[0], temp_offset[1] +
(y1 + h1) * FEATURE_SCALE[1]), (0, 255, 255))
cv.ShowImage("Whole Image", img)
def merge_images(img1, img2, vertical=None):
w, h = sizeOf(img2)
new_size, second_roi = ((w * 2, h),
(w, 0, w,
h)) if h * 1.3 > w and not vertical else ((w,
h * 2),
(0, h,
w, h))
merged = cv.CreateImage(new_size, img1.depth, img1.channels)
cv.SetImageROI(merged, (0, 0, w, h))
cv.Copy(img1, merged)
cv.SetImageROI(merged, second_roi)
cv.Copy(img2, merged)
cv.ResetImageROI(merged)
return merged
def find_blob(self):
#Convert Incoming image
#self.image = self.bridge.imgmsg_to_cv(data, "bgr8")
#self.image = cv.QueryFrame(self.capture)
cv.Copy(self.image_color, self.image)
#These shenanigans are to get input from the keyboard, cv.WaitKey gives
#back the ascii value of the key pressed in the low 8 bits, but the rest
#of the higher bits can be anything apparently.
key_press_raw = cv.WaitKey(5) #gets a raw key press
key_press = key_press_raw & 255 #sets all but the low 8 bits to 0
self.check_key_press(key_press)
#Create the thresholded image
self.threshold = self.create_image()
#Find blob regions, draw a rectangle around the largest and a circle at
#its center.
self.find_regions()
#Display the images
cv.ShowImage('image', self.image)
cv.ShowImage('threshold', self.threshold)
cv.ShowImage('hsv', self.hsv)
def rotacion_y_posicion_robot(self,robo_x=200,robo_y=100,robo_th=80): """ \brief graficar el robot en el lienzo de mapa \param self no se necesita incluirlo al utilizar la funcion ya que se lo pone solo por ser la definicion de una clase \param robo_x coordenada x de la odometria del robot \param robo_y coordenada y de la odometria del robot \param robo_th Valor Th del robot \return Nada """ image_mapa=cv.LoadImage(self.nombre_archivo1, cv.CV_LOAD_IMAGE_COLOR) dimensiones_robot=self.dimensiones_robot image1=cv.CreateImage(dimensiones_robot,8,3) image_mascara=cv.CreateImage(dimensiones_robot,8,1) ##rotacion #Rotar el robot src_center=dimensiones_robot[0]/2,dimensiones_robot[1]/2 rot_mat=cv.CreateMat( 2, 3, cv.CV_32FC1 ) cv.GetRotationMatrix2D(src_center, robo_th, 1.0,rot_mat); cv.WarpAffine(self.robot,image1,rot_mat) #crear filtro para negro cv.InRangeS(image1,cv.RGB(0,0,0),cv.RGB(14,14,14),image_mascara) cv.Not(image_mascara,image_mascara) #cv.ReleaseImage(image1) #reducir y posicion cv.SetImageROI(image_mapa,(robo_x,robo_y, dimensiones_robot[0], dimensiones_robot[1])); cv.Copy(image1,image_mapa,mask=image_mascara) cv.ResetImageROI(image_mapa); cv.SaveImage(self.nombre_archivo, image_mapa) #Saves the image#
def acquire(self, img):
self.found_data = False
cv.SetImageROI(img, self.rect.ToCvRect())
self.internal_img = cv.CreateImage(cv.GetSize(img), img.depth,
img.nChannels)
cv.Copy(img, self.internal_img)
cv.ResetImageROI(img)
def redraw_monocular(self, drawable):
width, height = cv.GetSize(drawable.scrib)
display = cv.CreateMat(max(480, height), width + 100, cv.CV_8UC3)
cv.Zero(display)
cv.Copy(drawable.scrib, cv.GetSubRect(display, (0, 0, width, height)))
cv.Set(cv.GetSubRect(display, (width, 0, 100, height)),
(255, 255, 255))
self.buttons(display)
if not self.c.calibrated:
if drawable.params:
for i, (label, lo, hi, progress) in enumerate(drawable.params):
(w, _), _ = cv.GetTextSize(label, self.font)
cv.PutText(display, label,
(width + (100 - w) / 2, self.y(i)), self.font,
(0, 0, 0))
color = (0, 255, 0)
if progress < 1.0:
color = (0, int(progress * 255.), 255)
cv.Line(display, (int(width + lo * 100), self.y(i) + 20),
(int(width + hi * 100), self.y(i) + 20), color, 4)
else:
cv.PutText(display, "lin.", (width, self.y(0)), self.font,
(0, 0, 0))
linerror = drawable.linear_error
if linerror < 0:
msg = "?"
else:
msg = "%.2f" % linerror
#print "linear", linerror
cv.PutText(display, msg, (width, self.y(1)), self.font, (0, 0, 0))
self.show(display)
def getIris(frame):
iris = []
copyImg = cv.CloneImage(frame)
resImg = cv.CloneImage(frame)
grayImg = cv.CreateImage(cv.GetSize(frame), 8, 1)
mask = cv.CreateImage(cv.GetSize(frame), 8, 1)
storage = cv.CreateMat(frame.width, 1, cv.CV_32FC3)
cv.CvtColor(frame, grayImg, cv.CV_BGR2GRAY)
cv.Canny(grayImg, grayImg, 5, 70, 3)
cv.Smooth(grayImg, grayImg, cv.CV_GAUSSIAN, 7, 7)
circles = getCircles(grayImg)
iris.append(resImg)
for circle in circles:
rad = int(circle[0][2])
global radius
radius = rad
cv.Circle(mask, centroid, rad, cv.CV_RGB(255, 255, 255), cv.CV_FILLED)
cv.Not(mask, mask)
cv.Sub(frame, copyImg, resImg, mask)
x = int(centroid[0] - rad)
y = int(centroid[1] - rad)
w = int(rad * 2)
h = w
cv.SetImageROI(resImg, (x, y, w, h))
cropImg = cv.CreateImage((w, h), 8, 3)
cv.Copy(resImg, cropImg)
cv.ResetImageROI(resImg)
return (cropImg)
return (resImg)
def SubImage(img, region):
'''return a subimage as a new image. This allows
for the region going past the edges.
region is of the form (x1,y1,width,height)'''
(x1, y1, width, height) = region
zeros = numpy.zeros((height, width, 3), dtype='uint8')
ret = cv.GetImage(cv.fromarray(zeros))
(img_width, img_height) = cv.GetSize(img)
if x1 < 0:
sx1 = 0
xofs = -x1
else:
sx1 = x1
xofs = 0
if y1 < 0:
sy1 = 0
yofs = -y1
else:
sy1 = y1
yofs = 0
if sx1 + width <= img_width:
w = width
else:
w = img_width - sx1
if sy1 + height <= img_height:
h = height
else:
h = img_height - sy1
cv.SetImageROI(img, (sx1, sy1, w - xofs, h - yofs))
cv.SetImageROI(ret, (xofs, yofs, w - xofs, h - yofs))
cv.Copy(img, ret)
cv.ResetImageROI(img)
cv.ResetImageROI(ret)
return ret
def find_biggest_region(self):
""" this code should find the biggest region and
then determine some of its characteristics, which
will help direct the drone
"""
# copy the thresholded image
cv.Copy( self.threshed_image, self.copy ) # copy self.threshed_image
# this is OpenCV's call to find all of the contours:
contours = cv.FindContours(self.copy, self.storage, cv.CV_RETR_EXTERNAL,
cv.CV_CHAIN_APPROX_SIMPLE)
# Next we want to find the *largest* contour
if len(contours)>0:
biggest = contours
biggestArea=cv.ContourArea(contours)
while contours != None:
nextArea=cv.ContourArea(contours)
if biggestArea < nextArea:
biggest = contours
biggestArea = nextArea
contours=contours.h_next()
#Use OpenCV to get a bounding rectangle for the largest contour
self.br = cv.BoundingRect(biggest,update=0)
#Publish the data.
self.publishBoxData()
def __init__(self, frame, rect, time_makes_difference=False):
self.rect = rect
self.time_makes_difference = time_makes_difference
self.face_photo_valid = False
self.face_size = [rect[2], rect[3]]
self.personID = -1
self.white_level = 6
try:
sub_image = cv.GetSubRect(frame, rect)
except:
return None
self.frame_copy = cv.CreateImage((rect[2], rect[3]), cv.IPL_DEPTH_8U,
frame.nChannels)
if frame.origin == cv.IPL_ORIGIN_TL:
cv.Copy(sub_image, self.frame_copy)
else:
cv.Flip(sub_image, self.frame_copy, 0)
self.find_eyes()
if self.is_a_valid_face_photo():
self.find_wrinkles()
self.find_avarage_colors()
self.find_gender_age_emotions()
#self.count_face_vector()
self.create_face_vector()
def save(self, filename, cropFlag = False):
"""save cropped and rotated image"""
if not cropFlag:
cv.SaveImage(filename +'-vectors.png',self.render())
else:
tmp = rotate(self.cImage, self.center, self.cardinalOffset +\
self.north)
#mask horizon
mask = cv.CreateImage(self.res, 8, 1)
cv.Zero(mask)
cv.Circle(mask, self.center, self.radius, (255,255,255))
cv.FloodFill(mask,(1,1),(0,0,0))
cv.FloodFill(mask, self.center,
(255,255,255),lo_diff=cv.RealScalar(5))
masked = cv.CloneImage(tmp)
cv.Zero(masked)
cv.Copy(tmp, masked, mask)
cv.SaveImage(filename +'-cropped.png',crop(masked, self))
#CSV output
array = magnitudeToTheta(self.polarArray,self.radius)
f = open(filename + '.csv', 'w')
f.write('00\n')
f.write(','.join([str(v[0]) for v in array]))
f.write('\n')
f.write(','.join([str(v[1]) for v in array]))
f.write('\n')
f.flush()
f.close()
def preprocessImage(inputPath, outputPath):
image = cv.LoadImage(inputPath, 0)
#Find the most likely face
(x, y, w, h), n = mostLikelyHaar(image, haarFace)
croppedImage = cv.CreateImage((w, h), image.depth, image.nChannels)
scaledImage = cv.CreateImage((256, 256), image.depth, image.nChannels)
src_region = cv.GetSubRect(image, (x, y, w, h))
cv.Copy(src_region, croppedImage)
cv.Resize(croppedImage, scaledImage)
image = scaledImage
#Find each ficudial point
leftEye = ImageObject(image, haarLeftEye,
inPercentRect(image, 0, 0, .6, .5))
leftEyePoints = leftEye.getPoints([(.2, .5), (.8, .5)])
rightEye = ImageObject(image, haarRightEye,
inPercentRect(image, .4, 0, 1, .5))
rightEyePoints = rightEye.getPoints([(.2, .5), (.8, .5)])
mouth = ImageObject(image, haarMouth, inPercentRect(image, 0, .6, 1, 1))
mouthPoints = mouth.getPoints([(0, .3), (.5, .3), (1, .3)])
nose = ImageObject(image, haarNose)
nosePoints = nose.getPoints([(.2, .5), (.5, .5), (.8, .5)])
#rotate each set of points by the tilt of the face
tiltAngle = math.atan2(rightEyePoints[0][1] - leftEyePoints[0][1],
rightEyePoints[0][0] - leftEyePoints[0][0])
leftEyePoint = tiltPoints(tiltAngle, leftEyePoints)
rightEyePoints = tiltPoints(tiltAngle, rightEyePoints)
mouthPoints = tiltPoints(tiltAngle, mouthPoints)
nosePoints = tiltPoints(tiltAngle, nosePoints)
image = rotateImage(image, tiltAngle, (w / 2, h / 2))
leftEye = ImageObject(image, haarLeftEye,
inPercentRect(image, 0, 0, .6, .5))
rightEye = ImageObject(image, haarRightEye,
inPercentRect(image, .4, 0, 1, .5))
mouth = ImageObject(image, haarMouth, inPercentRect(image, 0, .6, 1, 1))
nose = ImageObject(image, haarNose)
rotation = math.log(leftEye.w) - math.log(rightEye.w)
print rotation
info = {
'image': outputPath,
'lbp-left-eye': calcLBP(image, leftEye),
'left-eye': leftEye.getTuple(),
'lbp-right-eye': calcLBP(image, rightEye),
'right-eye': rightEye.getTuple(),
'lbp-mouth': calcLBP(image, mouth),
'mouth': mouth.getTuple(),
'tilt': tiltAngle,
'rotation': rotation
}
#save image of the cropped face
cv.SaveImage(outputPath, image)
return info
def compose(a, b, d):
w, h = (640, 480)
r = cv.CreateImage((w * 3, h), cv.IPL_DEPTH_8U, 3)
cv.SetZero(r)
for (i, src) in enumerate([a, b, d]):
if src != None:
cv.SetImageROI(r, (w * i, 0, w, h))
if src.nChannels == 1:
deep = cv.CreateImage((w, h), cv.IPL_DEPTH_8U, 3)
cv.CvtColor(src, deep, cv.CV_GRAY2BGR)
cv.Copy(deep, r)
else:
cv.Copy(src, r)
cv.ResetImageROI(r)
return r
def find_new_markers(self, do_canny=True):
'''
Find markers totally without taking their previous position into
consideration
@param do_canny: perform edge recognition
'''
for scale in range(self.max_scale, -1, -1):
if len(self.not_found) == 0:
return
self.set_scale(scale)
if do_canny:
cv.Canny(self.gray_img, self.bw_img, 100, 300)
cv.Copy(self.bw_img, self.tmp_img)
self.tmp_img[1, 1] = 255
cont = cv.FindContours(
self.tmp_img, cv.CreateMemStorage(), cv.CV_RETR_TREE)
db.DrawContours(
self.canny_img, cont, (255, 255, 255), (128, 128, 128), 10)
# db.show([self.canny_img,self.tmp_img,self.bw_img], 'show', 0, 1)
# cv.ShowImage("name", self.canny_img)
# cv.ShowImage("name1", self.tmp_img)
self.set_scale(0)
self.scale_factor = 1 << scale
self.test_contours(cont)
#markers= filter(lambda sq:sq.check_square(self.img,self.gray_img),rects)
return self.markers
def getForegroundPixels(self, bgcolor=None):
'''
@param bgcolor: The background color to use. Specify as an (R,G,B) tuple.
Specify None for a blank/black background.
@return: The full color foreground pixels on either a blank (black)
background, or on a background color specified by the user.
@note: You must call detect() before getForegroundPixels() to
get updated information.
'''
if self._fgMask == None: return None
#binary mask selecting foreground regions
mask = self._fgMask.asOpenCVBW()
#full color source image
image = self._annotateImg.copy().asOpenCV()
#dest image, full color, but initially all zeros (black/background)
# we will copy the foreground areas from image to here.
dest = cv.CloneImage(image)
if bgcolor == None:
cv.SetZero(dest)
else:
cv.Set(dest, cv.RGB(*bgcolor))
cv.Copy(image, dest,
mask) #copy only pixels from image where mask != 0
return pv.Image(dest)
def __refresh_canny(self):
cv.Canny(self.res_smooth, self.canny, self.__canny_lo, self.__canny_hi, self.__canny_apeture * 2 + 3)
#cv.Threshold(self.res_smooth, self.canny, self.__canny_lo * 2, 255, cv.CV_THRESH_BINARY)
cv.ShowImage('Canny', self.canny)
cv.Copy(self.canny, self.contour_in)
self.contours = cv.FindContours(self.contour_in, self.c_storage, cv.CV_RETR_LIST, cv.CV_CHAIN_APPROX_NONE)
self.__refresh_poly()
def CompositeThumbnail(img, regions, thumb_size=100):
'''extract a composite thumbnail for the regions of an image
The composite will consist of N thumbnails side by side
'''
composite = cv.CreateImage((thumb_size*len(regions), thumb_size),8,3)
for i in range(len(regions)):
(x1,y1,x2,y2) = regions[i].tuple()
midx = (x1+x2)/2
midy = (y1+y2)/2
if (x2-x1) > thumb_size or (y2-y1) > thumb_size:
# we need to shrink the region
rsize = max(x2+1-x1, y2+1-y1)
src = cuav_util.SubImage(img, (midx-rsize/2,midy-rsize/2,rsize,rsize))
thumb = cv.CreateImage((thumb_size, thumb_size),8,3)
cv.Resize(src, thumb)
else:
x1 = midx - thumb_size/2
y1 = midy - thumb_size/2
thumb = cuav_util.SubImage(img, (x1, y1, thumb_size, thumb_size))
cv.SetImageROI(composite, (thumb_size*i, 0, thumb_size, thumb_size))
cv.Copy(thumb, composite)
cv.ResetImageROI(composite)
return composite
def OverlayImage(img, img2, x, y):
'''overlay a 2nd image on a first image, at position x,y
on the first image'''
(w, h) = cv.GetSize(img2)
cv.SetImageROI(img, (x, y, w, h))
cv.Copy(img2, img)
cv.ResetImageROI(img)
def getData():
for i in range (0 , classes):
for j in range (0, train_samples):
if j < 10 :
fichero = "OCR/"+str(i) + "/"+str(i)+"0"+str(j)+".pbm"
else:
fichero = "OCR/"+str(i) + "/"+str(i)+str(j)+".pbm"
src_image = cv.LoadImage(fichero,0)
prs_image = preprocessing(src_image, size, size)
row = cv.GetRow(trainClasses, i*train_samples + j)
cv.Set(row, cv.RealScalar(i))
row = cv.GetRow(trainData, i*train_samples + j)
img = cv.CreateImage( ( size, size ), cv.IPL_DEPTH_32F, 1)
cv.ConvertScale(prs_image,img,0.0039215, 0)
data = cv.GetSubRect(img, (0,0, size,size))
row1 = cv.Reshape( data, 0, 1 )
cv.Copy(row1, row)
def image_to_show( self ):
cv.Copy( self.image, self.view_image )
if self.patch:
draw_patch( self.view_image, *self.patch, filled=True )
if self.click_pt:
cv.Circle( self.view_image, self.click_pt, 5*self.zoom_out, cv.RGB(0,0,255), -1 )
return self.view_image
def motion_detector():
global max_area, avg, prev_pos, largest_contour
contour = cv.FindContours(
temp_image,
store,
mode=cv.CV_RETR_EXTERNAL,
method=cv.CV_CHAIN_APPROX_NONE) #Findling contours
cv.Copy(img, render_image) #Copying for painting on the image
if len(contour) != 0:
temp_contour = contour
area = 0
max_area_test = max_area
while temp_contour != None: #Routine to find the largest contour
area = cv.ContourArea(temp_contour)
if area > max_area_test:
largest_contour = temp_contour
max_area_test = area
temp_contour = temp_contour.h_next()
rect = cv.BoundingRect(largest_contour)
cv.DrawContours(render_image, largest_contour, (0, 255, 0),
(0, 0, 255), 1, 3)
cv.Rectangle(render_image, (rect[0], rect[1]),
(rect[0] + rect[2], rect[1] + rect[3]), (255, 0, 0))
avg = rect[0] + rect[2] / 2
else:
avg = img.width / 2
def matchsize(A, B):
""" Given two cvMats A, B, returns a cropped/padded version of
A that has the same dimensions as B.
"""
wA, hA = cv.GetSize(A)
wB, hB = cv.GetSize(B)
if wA == wB and hA == hB:
return A
SetImageROI = cv.SetImageROI
out = cv.CreateImage((wB, hB), A.depth, A.channels)
wOut, hOut = cv.GetSize(out)
if wA < wOut and hA < hOut:
SetImageROI(out, (0, 0, wA, hA))
elif wA >= wOut and hA < hOut:
SetImageROI(out, (0, 0, wOut, hA))
SetImageROI(A, (0, 0, wOut, hA))
elif wA < wOut and hA >= hOut:
SetImageROI(out, (0, 0, wA, hOut))
SetImageROI(A, (0, 0, wA, hOut))
else: # wA >= wOut and hA >= hOut:
SetImageROI(A, (0, 0, wOut, hOut))
cv.Copy(A, out)
cv.ResetImageROI(out)
cv.ResetImageROI(A)
return out
def listen(self):
if isinstance(self.background, np.ndarray):
bgimg = cv.CreateImage(self.background.shape[:2], 8, 3)
img = cv.CreateImage(self.background.shape[:2], 8, 3)
theWidth = self.background.shape[1]
theHeight = self.background[0]
else:
bgimg = cv.CreateImage(
(self.background.width, self.background.height), 8, 3)
img = cv.CreateImage(
(self.background.width, self.background.height), 8, 3)
theWidth = self.background.width
theHeight = self.background.height
cv.Copy(self.background, bgimg)
smallimg = cv.CreateImage(
(theWidth / self.zoom, theHeight / self.zoom), 8, 3)
cv.GetRectSubPix(bgimg, smallimg,
(theWidth / (2 * self.zoom) + self.offset[0],
theHeight / (2 * self.zoom) + self.offset[1]))
cv.Resize(smallimg, img)
if (self.cp != False):
cv.Circle(img, self.zoomPt(self.cp.x, self.cp.y), 3,
cv.RGB(0, 255, 0), -1)
cv.Line(img, (self.ch_x - 25, self.ch_y), (self.ch_x + 25, self.ch_y),
cv.RGB(255, 255, 0))
cv.Line(img, (self.ch_x, self.ch_y - 25), (self.ch_x, self.ch_y + 25),
cv.RGB(255, 255, 0))
cv.ShowImage(self.name, img)
cv.WaitKey(25)
def image_processor():
cv.Smooth(gray_image, gray_image, cv.CV_GAUSSIAN, 3,
3) #Blurring to remove some noise
cv.AbsDiff(prev_image, gray_image,
accumulator) #Getting the difference image
cv.InRangeS(accumulator, threshold_limit1_lower, threshold_limit1_upper,
accumulator) #Thresholding the difference image
cv.Dilate(accumulator, accumulator, None,
2) #Dilating the thresholded difference image
cv.Add(accumulator, sum_image, sum_image,
accumulator) #Adding the image to a register to use fading
cv.SubS(sum_image, fading_factor, sum_image) #Fading
cv.InRangeS(sum_image, threshold_limit2_lower, threshold_limit2_upper,
accumulator) #Thresholding the fading image
cv.Copy(gray_image, prev_image)
cv.Copy(accumulator, temp_image)
def setup(flipped, capture, thehandcolor):
"""Initializes camera and finds initial skin tone"""
#creates initial window and prepares text
color = (40, 0, 0)
font = cv.InitFont(cv.CV_FONT_HERSHEY_SIMPLEX, 1.0, 1.0)
textsize1 = (cv.GetSize(flipped)[0] / 2 - 150,
cv.GetSize(flipped)[1] / 2 - 140)
textsize2 = (cv.GetSize(flipped)[0] / 2 - 150,
cv.GetSize(flipped)[1] / 2 - 110)
point1 = (cv.GetSize(flipped)[0] / 2 - 25, cv.GetSize(flipped)[1] / 2 - 25)
point2 = (cv.GetSize(flipped)[0] / 2 + 25, cv.GetSize(flipped)[1] / 2 + 25)
#until Enter is pressed
while (cv.WaitKey(10) != 10):
#captures live video, and draws sub-box and text
frame = cv.QueryFrame(capture)
cv.Copy(frame, flipped)
cv.Flip(flipped, flipped, 1)
cv.Rectangle(flipped, point1, point2, color, 2)
cv.PutText(flipped, "Put your hand in the box ", textsize1, font,
color)
cv.PutText(flipped, "and press enter", textsize2, font, color)
cv.ShowImage("w2", flipped)
#Creates sub-image inside box, and returns average color in box
sub = cv.GetSubRect(flipped, (cv.GetSize(flipped)[0] / 2 - 25,
cv.GetSize(flipped)[1] / 2 - 25, 50, 50))
cv.Set(thehandcolor, cv.Avg(sub))
return cv.Avg(sub)
