def main():
ap = argparse.ArgumentParser()
ap.add_argument("-i1",
"--image1",
required=True,
help="Path to the checkboard_image")
ap.add_argument("-i2",
"--image2",
required=True,
help="Path to the arm_spread_image")
ap.add_argument("-i3",
"--image3",
required=True,
help="Path to the waist_image")
ap.add_argument("-a",
"--affine_mode",
required=True,
help="To perform Affine Corrections")
args = vars(ap.parse_args())
# load the image, clone it, and setup the mouse callback function
image = cv2.imread(args["image1"])
arm_spread_image = cv2.imread(args["image2"])
waist_image = cv2.imread(args["image3"])
affine_correct_flag = (args["affine_mode"])
metre_pixel_x, metre_pixel_y, coordinate, affine_correct_parameters = analyze_chessboard(
image, affine_correct_flag)
segmented_image = segment.segmenter(image)
print "Segmentation Completed 1"
segmented_arm_image = segment.segmenter(arm_spread_image)
print "Segmentation Completed 2"
cv2.imwrite("first.jpg", segmented_image)
cv2.imwrite("second.jpg", segmented_arm_image)
block_cut = np.zeros(segmented_image.shape)
block_cut[coordinate[0][1]:coordinate[1][1],
coordinate[0][0]:coordinate[1][0]] = 1
# segmented_image=segmented_image*block_cut
if (affine_correct_flag == 'True'):
arm_spread_image = affine_correct(arm_spread_image,
affine_correct_parameters)
waist_image = affine_correct(waist_image, affine_correct_parameters)
segmented_image = affine_correct(segmented_image,
affine_correct_parameters)
print "Affine Corrected"
# detect_wrist(segmented_arm_image)
# cv2.imwrite("affine_corrected.jpg",segmented_image)
measure_distance(segmented_image, segmented_arm_image, arm_spread_image,
waist_image, image, metre_pixel_x, metre_pixel_y)
def main():
ap = argparse.ArgumentParser()
ap.add_argument("-i1",
"--image1",
required=True,
help="Path to the checkboard_image")
ap.add_argument("-i2",
"--image2",
required=True,
help="Path to the arm_spread_image")
ap.add_argument("-i3",
"--image3",
required=True,
help="Path to the waist_image")
ap.add_argument("-a",
"--affine_mode",
required=True,
help="To perform Affine Corrections")
args = vars(ap.parse_args())
image = cv2.imread(args["image1"])
arm_spread_image = cv2.imread(args["image2"])
waist_image = cv2.imread(args["image3"])
affine_correct_flag = (args["affine_mode"])
metre_pixel_x, metre_pixel_y, coordinate, affine_correct_parameters = analyze_chessboard(
image, affine_correct_flag)
segmented_image = segment.segmenter(image)
segmented_arm_image = segment.segmenter(arm_spread_image)
a = cv2.imwrite("first.jpg", segmented_image)
b = cv2.imwrite("second.jpg", segmented_arm_image)
block_cut = np.zeros(segmented_image.shape)
block_cut[coordinate[0][1]:coordinate[1][1],
coordinate[0][0]:coordinate[1][0]] = 1
if (affine_correct_flag == 'True'):
arm_spread_image = affine_correct(arm_spread_image,
affine_correct_parameters)
waist_image = affine_correct(waist_image, affine_correct_parameters)
segmented_image = affine_correct(segmented_image,
affine_correct_parameters)
measure_distance(segmented_image, segmented_arm_image, arm_spread_image,
waist_image, image, metre_pixel_x, metre_pixel_y)
def read_data(src_dir, training):
tags = []
articles = []
for dir_path, dir_names, file_names in os.walk(src_dir):
for filename in file_names:
file_path = os.path.join(dir_path, filename)
with open(file_path, 'r') as f:
text = f.read()
segs = segmenter(text)
article = " ".join(del_stops(segs))
article = del_stops(segs)
if training:
tag = dir_path.split("/")[-1]
tags.append(tag)
articles.append(article)
if training:
assert len(tags) == len(articles)
tags_list = ["dry", "normal", "oily"]
return (tags, articles, tags_list)
import tkinter as tk import cv2 from segment import segmenter from tkinter import filedialog root = tk.Tk() root.withdraw() file_path = filedialog.askopenfilename() image_to_segment = cv2.imread(file_path) segmenter(image_to_segment)
from sklearn.model_selection import train_test_split
# == Load data ==
print("Loading dataset...")
files = os.listdir()
D = []
# Label:
# - dry: 0
# - normal: 1
# - oil: 2
y = np.array([])
print(len(D), " documents, ", len(y), " labels.")
# == Chinese Segmentation ==
X = np.array([del_stops(segmenter(d)) for d in D])
# == Split dataset ==
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)
# == Vector Transformation ==
print("Extracting features from the dataset...")
vectorizer = Pipeline([
('vect',
HashingVectorizer(n_features=(2**21), non_negative=True,
lowercase=False)),
('tfidf', TfidfTransformer(norm='l2')),
])
if __name__ == '__main__':
vectorizer.fit()
def main():
ap = argparse.ArgumentParser()
ap.add_argument("-i1",
"--image1",
required=True,
help="Path to the checkboard_image")
ap.add_argument("-i2",
"--image2",
required=True,
help="Path to the arm_spread_image")
ap.add_argument("-i3",
"--image3",
required=True,
help="Path to the waist_image")
ap.add_argument("-a",
"--affine_mode",
required=True,
help="To perform Affine Corrections")
args = vars(ap.parse_args())
# load the image, clone it, and setup the mouse callback function
image = cv2.imread(args["image1"])
image22 = np.copy(image)
arm_spread_image = cv2.imread(args["image2"])
waist_image = cv2.imread(args["image3"])
affine_correct_flag = (args["affine_mode"])
metre_pixel_x, metre_pixel_y, coordinate, affine_correct_parameters = analyze_chessboard(
image, affine_correct_flag)
segmented_image = segment.segmenter(image)
print "Segmentation Completed 1"
segmented_arm_image = segment.segmenter(arm_spread_image)
print "Segmentation Completed 2"
image2 = affine_correct(image, affine_correct_parameters)
image2 = cv2.rectangle(image2, (coordinate[0][0], coordinate[0][1]),
(coordinate[1][0], coordinate[1][1]), (255, 0, 0),
3)
block_cut = np.zeros(segmented_image.shape)
block_cut[coordinate[0][1]:coordinate[1][1],
coordinate[0][0]:coordinate[1][0]] = 1
segmented_image = segmented_image * block_cut
if (affine_correct_flag == 'True'):
image2 = affine_correct(image, affine_correct_parameters)
cv2.imwrite('affine_correction_3.jpg',
np.concatenate((image, image2), axis=1))
arm_spread_image = affine_correct(arm_spread_image,
affine_correct_parameters)
waist_image = affine_correct(waist_image, affine_correct_parameters)
segmented_image = affine_correct(segmented_image,
affine_correct_parameters)
print "Affine Corrected"
measure_distance(segmented_image, segmented_arm_image, arm_spread_image,
waist_image, image22, metre_pixel_x, metre_pixel_y)