threash = cv2.threshold(gray, 0, 255,
cv2.THRESH_OTSU | cv2.THRESH_BINARY_INV)[1]
cnts = cv2.findContours(threash.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if imutils.is_cv2() else cnts[1]
cnts = sorted(cnts, key=cv2.contourArea, reverse=True)[:4]
cnts = contours.sort_contours(cnts)[0]
output = cv2.merge([gray] * 3)
predictions = []
for c in cnts:
x, y, w, h = cv2.boundingRect(c)
roi = gray[y - 5:y + h + 5, x - 5:x + w + 5]
roi = preprocess(roi, 28, 28)
roi = np.expand_dims(
img_to_array(roi), axis=0
) / 255.0 # output (batch_num, width, height, channels), add dim batch_num
pred = model.predict(roi).argmax(
axis=1)[0] + 1 #axis =0 row =1 columns
predictions.append(str(pred))
cv2.rectangle(output, (x - 2, y - 2), (x + w + 4, y + h + 4),
(0, 255, 0), 1)
cv2.putText(output, str(pred), (x - 5, y - 5),
cv2.FONT_HERSHEY_SIMPLEX, 0.55, (0, 255, 0), 2)
print("[INFO] captcha: {}".format("".join(predictions)))
cv2.imshow("output", output)
cv2.waitKey()
'--dataset',
required=True,
help='Path to input dataset.')
argument_parser.add_argument('-m',
'--model',
required=True,
help='Path to output model.')
arguments = vars(argument_parser.parse_args())
data = list()
labels = list()
for image_path in paths.list_images(arguments['dataset']):
image = cv2.imread(image_path)
image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
image = preprocess(image, 28, 28)
image = img_to_array(image)
data.append(image)
label = image_path.split(os.path.sep)[-2]
labels.append(label)
data = np.array(data, dtype='float') / 255.0
labels = np.array(labels)
X_train, X_test, y_train, y_test = train_test_split(data,
labels,
test_size=.25,
random_state=42)
label_binarizer = LabelBinarizer().fit(y_train)