def path_tracing(img_filtered, contours):
"""Find contours in input image using Douglas-Peuker algorithm
Parameters
----------
img_filtered : Image
Input image file in which to find contours.
level : int
Value along which to find contours in the array.
contours : list of (n,2)-ndarrays
Each contour is an ndarray of shape (n, 2), consisting of n (x, y) coordinates along the contour.
Output
-------
storage : 2D array of float coordinates
Output array of detected contour coordinates.
"""
storage = []
for contour in contours:
coords = approximate_polygon(contour, tolerance=0.1)
#if isClosed(coords) and polygon_area(coords) > 100:
if isClosed(coords) and polygon_area(coords) > 100:
storage.append(coords)
return storage
def path_tracing(img_filtered, contours):
storage = []
for contour in contours:
coords = approximate_polygon(contour, tolerance=0.1)
#if isClosed(coords) and polygon_area(coords) > 100:
if isClosed(coords) and polygon_area(coords) > 100:
storage.append(coords)
return storage
def max_index(img_filtered):
"""Find the level index that can yield the contour with largest area
Parameters
----------
img_filtered : Image
Input image file in which to find contours.
Output
-------
max_index : int
Level index that can yield the contour with largest area
"""
max_index = 0
max_area = 0
#max_count = 0
# 100 - 110
for i in range(120, 130):
area = 0
retval, threshold = cv2.threshold(img_filtered, i, 255,
cv2.THRESH_BINARY)
sk_img = image_denoise(img_filtered)
sk_thre = image_denoise(threshold)
contours = path_Marching_Squares(sk_thre, 0.45)
for contour in contours:
coords = approximate_polygon(contour, tolerance=1)
if isClosed(coords) and polygon_area(coords) > 100:
area += polygon_area(coords)
if area > max_area:
max_area = area
max_index = i
return max_index
def max_index(img_filtered):
max_index = 0
max_area = 0
#max_count = 0
# 80 - 120
for i in range(100, 110):
area = 0
retval, threshold = cv2.threshold(img_filtered, i, 255, cv2.THRESH_BINARY)
sk_img = image_denoise(img_filtered)
sk_thre = image_denoise(threshold)
contours = path_Marching_Squares(sk_thre, 0.45)
for contour in contours:
coords = approximate_polygon(contour, tolerance=1)
if isClosed(coords) and polygon_area(coords) > 100:
area += polygon_area(coords)
if area > max_area:
max_area = area
max_index = i
return max_index
def drawContour(img_num, path, center_list):
img = cv2.imread(path)
threshold = preprocess(img)
sk_img = image_denoise(img)
sk_thre = image_denoise(threshold)
contours = path_Marching_Squares(sk_thre, 0.45)
fig, ax = plt.subplots(figsize=(10, 10))
plt.gray()
ax.imshow(sk_img)
for contour in contours:
coords = approximate_polygon(contour, tolerance=0.1)
#if isClosed(coords) and polygon_area(coords) > 100:
if isClosed(coords) and polygon_area(coords) > 100:
ax.plot(coords[:, 1], coords[:, 0], '-r', linewidth=2)
for center in center_list:
ax.scatter(center[1], center[0], marker='o')
ax.axis((0, 256, 0, 256))
# save_path = '../data/auto/segmented_tube' + str(img_num) + '.jpg'
# save_path = './sample_results/4/sample%03d.png'% (img_num)
# save_path = './test_result/sample%03d.png'% (img_num)
save_path = './inverse_result/sample%03d.png' % (img_num)
plt.savefig(save_path)
sum_x = 0
sum_y = 0
length = len(points)
for i in range(0, length):
sum_x += points[i][0]
sum_y += points[i][1]
return (sum_x / length, sum_y / length)
if __name__ == "__main__":
path = './sample_images/4/sample109.png'
img = cv2.imread(path)
threshold = preprocess(img)
sk_img = image_denoise(img)
sk_thre = image_denoise(threshold)
contours = path_Marching_Squares(sk_thre, 0.45)
fig, ax = plt.subplots(figsize=(10,10))
plt.gray()
ax.imshow(sk_img)
# for contour in contours:
for i in range(1, 4):
contour = contours[i]
coords = approximate_polygon(contour, tolerance=0.1)
if isClosed(coords) and polygon_area(coords) > 100:
ax.plot(coords[:, 1], coords[:, 0], '-r', linewidth=2)
ax.axis((0,256,0,256))
plt.show()