import argparse
dataset_path = "data/digits.csv"
models_path = "models/svm.cpickle_01"
# load the dataset and initialize the data matrix
(digits, target) = dataset.load_digits(dataset_path)
data = []
# initialize the HOG descriptor
hog = HOG(orientations=18, pixelsPerCell=(10, 10),
cellsPerBlock=(1, 1), normalize=True)
# loop over the images
for image in digits:
# deskew the image, center it
image = dataset.deskew(image, 20)
image = dataset.center_extent(image, (20, 20))
# describe the image and update the data matrix
hist = hog.describe(image)
data.append(hist)
# train the model
model = LinearSVC(random_state=42)
model.fit(data, target)
# dump the model to file
joblib.dump(model, models_path)
print "create model done..."
(x, y, w, h) = cv2.boundingRect(c)
# if the width is at least 7 pixels and the height
# is at least 20 pixels, the contour is likely a digit
if w >= 7 and h >= 20:
# crop the ROI and then threshold the grayscale
# ROI to reveal the digit
roi = gray[y:y + h, x:x + w]
thresh = roi.copy()
T = mahotas.thresholding.otsu(roi)
thresh[thresh > T] = 255
thresh = cv2.bitwise_not(thresh)
# deskew the image center its extent
thresh = dataset.deskew(thresh, 20)
thresh = dataset.center_extent(thresh, (20, 20))
cv2.imshow("thresh", thresh)
# extract features from the image and classify it
hist = hog.describe(thresh)
digit = model.predict([hist])[0]
print("I think that number is: {}".format(digit))
# draw a rectangle around the digit, the show what the
# digit was classified as
cv2.rectangle(image, (x, y), (x + w, y + h), (0, 255, 0), 1)
cv2.putText(image, str(digit), (x - 10, y - 10),
cv2.FONT_HERSHEY_SIMPLEX, 1.2, (0, 255, 0), 2)
cv2.imshow("image", image)
cv2.waitKey(0)
# ROI to reveal the digit
# apply filer
roi = gray[y:y + h, x:x + w]
thresh = roi.copy()
T = mahotas.thresholding.otsu(roi)
thresh[thresh > T] = 255
thresh = cv2.bitwise_not(thresh)
# deskew the image center its extent
thresh = dataset.deskew(thresh, 20)
thresh = dataset.center_extent(thresh, (20, 20))
cv2.imshow("thresh", thresh)
# extract features from the image and classify it
hist = hog.describe(thresh)
digit = model.predict(hist)[0]
print "Creo que el numero es: %d" % (digit)
numeros.append(digit)
# draw a rectangle around the digit, the show what the
# digit was classified as
cv2.rectangle(image, (x, y), (x + w, y + h), 255, 1)
cv2.putText(image, str(digit), (x - 5, y - 5),
cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 255), 2)
cv2.imshow("image", image)
# prees a key to continue
cv2.waitKey(0)
print numeros
required=True,
help="path to where the model will be stored")
args = vars(ap.parse_args())
# load the dataset and initialize the data matrix
(digits, target) = dataset.load_digits(args["dataset"])
data = []
# initialize the HOG descriptor
hog = HOG(orientations=18,
pixelsPerCell=(10, 10),
cellsPerBlock=(1, 1),
transform=True)
# loop over the images
for image in digits:
# deskew the image, center it
image = dataset.deskew(image, 20)
image = dataset.center_extent(image, (20, 20))
# describe the image and update the data matrix
hist = hog.describe(image)
data.append(hist)
# train the model
model = LinearSVC(random_state=42)
model.fit(data, target)
# dump the model to file
joblib.dump(model, args["model"])
while True:
# grab the current frame
(grabbed, image) = camera.read()
# if we are viewing a video and we did not grab a
# frame, then we have reached the end of the video
if args.get("video") and not grabbed:
break
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
# blur the image, find edges, and then find contours along
# the edged regions
blurred = cv2.GaussianBlur(gray, (3, 3), 0)
# extract features from the image and classify it
hist = hog.describe(blurred)
direccion = le.inverse_transform(model.predict(hist))[0]
#le.inverse_transform(model.predict(features))[0]
print " Por favor: %s" % (direccion)
cv2.putText(image, str(direccion), (30, 30), cv2.FONT_HERSHEY_SIMPLEX, 1.2,
(255, 0, 0), 2)
cv2.imshow("image", image)
thresh = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
T = mahotas.thresholding.otsu(thresh)
thresh[thresh > T] = 255
thresh[thresh < T] = 0
thresh = cv2.bitwise_not(thresh)
# show our threshold image
