def detect(self, thresh):
# 图像大小处理
self.resized = imutils.resize(thresh, width=300) # 将图片的宽度设为300px
self.ratio = self.resized.shape[0] / float(self.resized.shape[0])
self.cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
self.cnts = imutils.grab_contours(self.cnts)
sd = ShapeDetector()
for c in self.cnts:
# 获得图形的矩
M = cv2.moments(c)
cX = int((M["m10"] / M["m00"]) * self.ratio)
cY = int((M["m01"] / M["m00"]) * self.ratio)
# 根据多边形点集个数，获得图形的名称
shape = sd.detect(c)
# 画出图形的形状并且打印名称
c = c.astype("float")
c *= self.ratio
c = c.astype("int")
# 画出轮廓 -1为画出所有的轮廓，0为只划出一个，1，2，3以此类推，(0,255,0)为划线的颜色，2为固定参数
cv2.drawContours(self.frameArray, [c], -1, (0, 255, 0), 2)
# 写出名称
cv2.putText(self.frameArray, shape, (cX, cY),
cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 255, 255), 2)
cv2.imshow("Image", self.frameArray)
cv2.waitKey(0)
def find_shapes(img):
cnts = cv2.findContours(img.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(cnts)
sd = ShapeDetector()
for c in cnts:
# Calculate center
M = cv2.moments(c)
cX = int((M["m10"] / M["m00"]) * ratio)
cY = int((M["m01"] / M["m00"]) * ratio)
# Detect shape
shape = sd.detect(c)
# scale contours back up to original image size
c = c.astype("float")
c *= ratio
c = c.astype("int")
cv2.drawContours(image, [c], -1, (0, 255, 0), 2)
cv2.putText(image, shape, (cX, cY), cv2.FONT_HERSHEY_SIMPLEX, 0.5,
(255, 255, 255), 2)
def main(color, shape):
notalign = True
aligned = False
ardunioData = serial.Serial('COM3', 9600)
right = '0'
left = '1'
up = '2'
down = '3'
laser = '4'
fire = '5'
frame = ShapeDetector.FrameCapure()
while aligned == False:
ShapeDetector.ColorShapeReader(color)
image = ShapeDetector.ImagewColor(frame)
coordinateX, coordinateY = ShapeDetector.ShapeDiscover(image, shape)
centerX, centerY = ShapeDetector.FindCenterImage(image)
notalign = False
if (centerY < coordinateY):
notalign = True
ardunioData.write(up.encode())
#move up
elif (centerY > coordinateY):
notalign = True
ardunioData.write(down.encode())
#move down
elif (centerX < coordinateX):
notalign = True
ardunioData.write(left.encode())
#move left
elif (centerX > coordinateX):
notalign = True
ardunioData.write(right.encode())
#move right
if (notalign == False):
aligned = True
#end while
#now adjust for distance
distanceAway = ShapeDetector.DistanceTest(frame, shape)
adjustment = ShapeDetector.AdjustTarget(distanceAway)
while (adjustment > 0):
ardunioData.write(up.encode())
adjustment = adjustment - 1
ardunioData.write(fire.encode()) #FIRE!!!
return 0
image = cv2.imread(args["image"])
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blurred = cv2.GaussianBlur(gray, (5,5), 0) # a 5x5 kernel is applied
# threshold turns image into binary
# if intensity of a pixel is larger than 60,
# replace the value with 255, which is white in OpenCV
intensityThreshold = 60
thresh = cv2.threshold(blurred, intensityThreshold, 255, cv2.THRESH_BINARY)[1]
# now find the contours
cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if imutils.is_cv2() else cnts[1] # they change the position in OpenCV3
# defines what shape an object is
sd = ShapeDetector()
print "// Escape key = Exit; Any other key = Continue //"
# process on each contour
for cnt in cnts:
# object center
M = cv2.moments(cnt)
cntX = int(M["m10"] / M["m00"])
cntY = int(M["m01"] / M["m00"])
# detect shape type
shapeName = sd.detect(cnt)
if shapeName is None: shapeName = ""
# say where and what the shape is in the console
# draw shapes and text
cv2.drawContours(image, [cnt], -1, (0,255,0), 2)
required=True,
help="Path to the input image")
args = vars(parser.parse_args())
#read the image
image = cv2.imread(args["image"])
rImage = cv2.imread(args["image"])
ratio = image.shape[0] / float(image.shape[0])
#Preprocess the image
grayedImage = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blurredImage = cv2.GaussianBlur(grayedImage, (5, 5), 0)
binImage = cv2.threshold(blurredImage, 60, 255, cv2.THRESH_BINARY)[1]
#Create a shape detector
shapeDetector = ShapeDetector()
#returns a set of contours on the shapes. Always use both lines
contours = cv2.findContours(binImage, cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
contours = contours[0] if imutils.is_cv2() else contours[1]
for c in contours:
#Compute the center of each contour.
#moments characterize the shape of an image like area, centroid,
#orientation, other statistical properties. Here we can see
#the indicies of the array are string keys
print(len(c))
moments = cv2.moments(c)
cX = int(moments["m10"] /
(moments["m00"] * ratio)) if moments["m00"] != 0 else 0
look(gray)
lab = cv2.cvtColor(blurred, cv2.COLOR_BGR2LAB)
look(lab)
thresh = cv2.threshold(gray, 100, 255, cv2.THRESH_BINARY)[1]
look(thresh)
# Invert
#thresh = cv2.bitwise_not(thresh)
#cv2.imshow("image", thresh); cv2.waitKey(0)
#cv2.imwrite(sys.argv[2], thresh)
# find contours in the thresholded image and initialize the shape detector
# (white=object, black=background)
cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = imutils.grab_contours(cnts)
sd = ShapeDetector()
cl = ColourDetector()
for c in cnts:
# Calculate center of contours
M = cv2.moments(c)
cX = int((M["m10"] / M["m00"]) * ratio)
cY = int((M["m01"] / M["m00"]) * ratio)
# Detect shape
shape = sd.detect(c)
# Label colours
color = cl.label(lab, c)
# scale contours back up to original image size
c = c.astype("float")
c *= ratio
frame = cv2.cvtColor(original_frame, cv2.COLOR_BGR2GRAY)
frame = cv2.medianBlur(frame, 5)
frame = cv2.GaussianBlur(frame, (5, 5), 0)
cv2.absdiff(background_image, frame, frame)
# frame = cv2.Canny(frame, 80,80)
frame = cv2.threshold(frame, 50, 255, cv2.THRESH_BINARY)[1]
#wyszukiwanie konturow
contours = cv2.findContours(frame.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
contours = contours[0] if imutils.is_cv2() else contours[1]
shape_detector = ShapeDetector()
mask = np.ones(frame.shape[:2], dtype="uint8") * 255
for contour in contours:
if shape_detector.detect(contour) == 'square':
cv2.drawContours(original_frame, [contour], -1, (0, 255, 0), 2)
# area = cv2.boundingRect(np.zeros(contour))
cv2.drawContours(mask, [contour], -1, 0,
-1) #dodanie do maski konturu kostki
mask = cv2.threshold(mask, 100, 255, cv2.THRESH_BINARY_INV)[1]
cropped_frame = cv2.bitwise_and(
original_frame, original_frame,
mask=mask) #wyciecie kostki z oryginalnego obrazu
cropped_frame = cv2.cvtColor(cropped_frame, cv2.COLOR_BGR2GRAY)
def AnalyseImage(path, outputPath, removeSpace, aproximationFactor=0.04):
data = path
print("Analysing image at location:" + str(data))
print("Loading image....")
# load the image and resize it to a smaller factor so that
# the shapes can be approximated better
image = cv2.imread(data)
#create backup so we can read collor dirrectly from unmodified image
backup = image.copy()
print("Resizing image")
resized = imutils.resize(image, width=2048)
ratio = image.shape[0] / float(resized.shape[0])
# convert the resized image to grayscale, blur it slightly,
# and threshold it
print("Changing colors to grayscale.")
gray = cv2.cvtColor(resized, cv2.COLOR_BGR2GRAY)
blurred = cv2.GaussianBlur(gray, (5, 5), 0)
thresh = cv2.threshold(blurred, 60, 255, cv2.THRESH_BINARY)[1]
# find contours in the thresholded image and initialize the
# shape detector
cnts = cv2.findContours(thresh.copy(), cv2.RETR_LIST,
cv2.CHAIN_APPROX_SIMPLE)
#cv2.findContours(gray.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
cnts = imutils.grab_contours(cnts)
mapData = []
sd = ShapeDetector()
# loop over the contours
for c in cnts:
# compute the center of the contour, then detect the name of the
# shape using only the contour
M = cv2.moments(c)
if M["m00"] == 0:
continue
#calculate center of object
resizedX = int(M["m10"] / M["m00"])
resizedY = int(M["m01"] / M["m00"])
cX = int((M["m10"] / M["m00"]) * ratio)
cY = int((M["m01"] / M["m00"]) * ratio)
shape = sd.detect(c, aproximationFactor)
rValue = int(backup[cY, cX, 2])
gValue = int(backup[cY, cX, 1])
bValue = int(backup[cY, cX, 0])
# multiply the contour (x, y)-coordinates by the resize ratio,
# then draw the contours and the name of the shape on the image
c = c.astype("float")
c *= ratio
c = c.astype("int")
(x, y), radius = cv2.minEnclosingCircle(c)
center = (int(x), int(y))
radius = int(radius)
#used for rotation detection
# supposadly the last item in returned struct is angle in degrees
rect = cv2.minAreaRect(c)
rotation = 0.0
if shape is not "circle":
rotation = float(rect[-1])
#lets draw rotation boxes
box = cv2.boxPoints(rect)
box = np.int0(box)
im = cv2.drawContours(image, [box], 0, (0, 0, 255), 2)
cv2.drawContours(image, [c], -1, (0, 255, 0), 2)
cv2.putText(image, shape, center, cv2.FONT_HERSHEY_SIMPLEX, 0.5,
(255, 255, 255), 2)
mapData.append({
"shape": shape,
"rotation": rotation, # might be unprecise
"color": [rValue, gValue, bValue],
"scale": radius,
"x": cX,
"y": cY,
})
normalizeData(mapData, removeSpace)
with open(outputPath + '.json', 'w') as outfile:
json.dump({"Items": mapData}, outfile)
#show the image with marked objects
cv2.imwrite(outputPath, image)
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blurred = cv2.GaussianBlur(gray, (5, 5), 0) # a 5x5 kernel is applied
# threshold turns image into binary
# if intensity of a pixel is larger than 60,
# replace the value with 255, which is white in OpenCV
intensityThreshold = 60
thresh = cv2.threshold(blurred, intensityThreshold, 255, cv2.THRESH_BINARY)[1]
# now find the contours
cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if imutils.is_cv2() else cnts[
1] # they change the position in OpenCV3
# defines what shape an object is
sd = ShapeDetector()
print "// Escape key = Exit; Any other key = Continue //"
# process on each contour
for cnt in cnts:
# object center
M = cv2.moments(cnt)
cntX = int(M["m10"] / M["m00"])
cntY = int(M["m01"] / M["m00"])
# detect shape type
shapeName = sd.detect(cnt)
if shapeName is None: shapeName = ""
# say where and what the shape is in the console
# draw shapes and text
cv2.drawContours(image, [cnt], -1, (0, 255, 0), 2)
ap = argparse.ArgumentParser()
ap.add_argument("-i", "--image", required=True, help="path to the image")
args = vars(ap.parse_args())
image = cv2.imread(args["image"])
resized = imutils.resize(image, width =300)
ratio = image.shape[0]/float(resized.shape[0]) #why init this?
gray = cv2.cvtColor(resized, cv2.COLOR_BGR2GRAY)
blurred = cv2.GaussianBlur(gray, (5,5), 0)
thresh = cv2.threshold(blurred, 60, 255, cv2.THRESH_BINARY)[1]
cnts = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if imutils.is_cv2() else cnts[1]
sd = ShapeDetector()
# loop through all detected contours
for c in cnts:
M = cv2.moments(c)
cX = int((M["m10"] / M["m00"]) * ratio)
cY = int((M["m01"] / M["m00"]) * ratio)
shape = sd.detect(c)
# multiplay the contour by resized ration and draw them back
c *= int(ratio)
cv2.drawContours(image, [c], -1, (0,255,0),2)
cv2.putText(image, shape, (cX, cY), cv2.FONT_HERSHEY_SIMPLEX,
0.5, (255,255,255), 2)
# show the result