def display_artifact(self, prediction, artifact, color, title):
im = self._image
im_art = im.copy()
im_art [prediction == artifact] = color
show_images([im, im_art], ["original", title], scale = 0.8)
def display_artifact(self, prediction, artifact, color, title):
im = self._image
im_art = im.copy()
im_art[prediction == artifact] = color
show_images([im, im_art], ["original", title], scale=0.8)
def display_contours(image,
contours,
color=(255, 0, 0),
thickness=-1,
title=None):
imShow = image.copy()
for i in range(0, len(contours)):
cv2.drawContours(imShow, contours, i, color, thickness)
show_images([imShow], scale=0.7, titles=title)
def display_average_pixels(self):
if len(self.avg_pixels) == 0:
self._get_initial_classes()
def stretch_image(i):
pixel = self._avg_pixels[i]
stretch = np.zeros((10, 10, 3), dtype='uint8')
stretch[:, :] = pixel
return stretch
images = [stretch_image(i) for i in range(0, self._n_objects)]
show_images(images, scale = 2.5)
def orb_show(img):
orb = cv2.ORB_create()
# find the keypoints with ORB
kp = orb.detect(img,None)
# compute the descriptors with ORB
kp, des = orb.compute(img, kp)
img2 = np.zeros_like(img)
# draw only keypoints location,not size and orientation
img3 = cv2.drawKeypoints(img, kp, img2, color=(0,255,0), flags=0)
show_images([img3])
def display_average_pixels(self):
if len(self.avg_pixels) == 0:
self._get_initial_classes()
def stretch_image(i):
pixel = self._avg_pixels[i]
stretch = np.zeros((10, 10, 3), dtype='uint8')
stretch[:, :] = pixel
return stretch
images = [stretch_image(i) for i in range(0, self._n_objects)]
show_images(images, scale=2.5)
# remove the whale root = "/kaggle/whales/imgs" img_file = "w_686.jpg" img_ = path.join(root, img_file) img = cv2.imread(img_) img_shift = img ### preprocess #pyramid down img_shift = cv2.pyrDown(img) img_shift = cv2.pyrDown(img_shift) # convert to np.float32 Z = img_shift.reshape((-1,3)) Z = np.float32(Z) # define criteria, number of clusters(K) and apply kmeans() criteria = (cv2.TERM_CRITERIA_EPS + cv2.TERM_CRITERIA_MAX_ITER, 10, 1.0) K = 3 ret,label,center=cv2.kmeans(Z,K,None,criteria,10,cv2.KMEANS_RANDOM_CENTERS) # Now convert back into uint8, and make original image center = np.uint8(center) res = center[label.flatten()] res2 = res.reshape((img_shift.shape)) res2 = cv2.pyrUp(cv2.pyrUp(res2)) show_images([img, res2])
def display_contours(image, contours, color = (255, 0, 0), thickness = -1, title = None):
imShow = image.copy()
for i in range(0, len(contours)):
cv2.drawContours(imShow, contours, i, color, thickness)
show_images([imShow], scale=0.7, titles=title)
def show_detected_mask(self):
mask = self.mask_off_od()
mask = cv2.resize(mask, (540, 540))
im = self._img.copy()
im[mask == 0] = 0
show_images([self._img, im])
def show_detected(self, ctr):
pr = self._processed.copy().astype('uint8')
cv2.circle(pr, ctr, 50, 0, 3)
cv2.circle(self.image, ctr, 50, (0, 0, 0), 3)
show_images([self.image, pr])
