def load_data(datafile, num_class, save=False, save_path='dataset.pkl'):
"""
load 17 flowers dataset
:param datafile:
:param num_class:
:param save:
:param save_path:
:return:
"""
train_list = open(datafile, 'r')
labels = []
images = []
for line in train_list:
tmp = line.strip().split(' ')
filepath = tmp[0]
print(filepath)
img = Image.open(filepath)
img = prep.resize_image(img, 224, 224)
np_img = prep.pil_to_nparray(img)
images.append(np_img)
# one-hot encoder
index = int(tmp[1])
label = np.zeros(num_class)
label[index] = 1
labels.append(label)
if save:
pickle.dump((images, labels), open(save_path, 'wb'))
return images, labels
def load_data(datafile, num_class, save=False, save_path='dataset.pkl'):
fr = codecs.open(datafile, 'r', 'utf-8')
train_list = fr.readlines()
labels = []
images = []
for line in train_list:
tmp = line.strip().split(' ')
fpath = tmp[0]
img = cv2.imread(fpath)
img = prep.resize_image(img, config.IMAGE_SIZE, config.IMAGE_SIZE)
np_img = np.asarray(img, dtype="float32")
images.append(np_img)
index = int(tmp[1])
label = np.zeros(num_class)
label[index] = 1
labels.append(label)
if save:
pickle.dump((images, labels), open(save_path, 'wb'))
fr.close()
return images, labels
def image_proposal(img_path):
img = cv2.imread(img_path)
img_lbl, regions = selectivesearch.selective_search(img,
scale=500,
sigma=0.9,
min_size=10)
candidates = set()
images = []
vertices = []
for r in regions:
# excluding same rectangle (with different segments)
if r['rect'] in candidates:
continue
# excluding small regions
if r['size'] < 220:
continue
if (r['rect'][2] * r['rect'][3]) < 500:
continue
# resize to 227 * 227 for input
proposal_img, proposal_vertice = prep.cropImage(img, r['rect'])
# Delete Empty array
if len(proposal_img) == 0:
continue
# Ignore things contain 0 or not C contiguous array
x, y, w, h = r['rect']
if w == 0 or h == 0:
continue
# Check if any 0-dimension exist
[a, b, c] = np.shape(proposal_img)
if a == 0 or b == 0 or c == 0:
continue
resized_proposal_img = prep.resize_image(proposal_img,
config.IMAGE_SIZE,
config.IMAGE_SIZE)
candidates.add(r['rect'])
img_float = np.asarray(resized_proposal_img, dtype="float32")
images.append(img_float)
vertices.append(r['rect'])
return images, vertices
def image_proposal(img_path):
"""
using selective search to generate proposals of image
:param img_path:
:return:
"""
img = skimage_io.imread(img_path)
img_lbl, regions = selectivesearch.selective_search(img,
scale=500,
sigma=0.9,
min_size=10)
candidates = set()
images = []
vertices = []
for r in regions:
# excluding same rectangle
if r['rect'] in candidates:
continue
if r['size'] < 220:
continue
# resize to 224 * 224 for input
proposal_img, proposal_vertice = prep.clip_pic(img, r['rect'])
# delete empty
if len(proposal_img) == 0:
continue
x, y, w, h = r['rect']
if w == 0 or h == 0:
continue
[a, b, c] = np.shape(proposal_img)
if a == 0 or b == 0 or c == 0:
continue
im = Image.fromarray(proposal_img)
resized_proposal_img = prep.resize_image(im, 224, 224)
candidates.add(r['rect'])
img_float = prep.pil_to_nparray(resized_proposal_img)
images.append(img_float)
vertices.append(r['rect'])
return images, vertices
