def rand_crop(crop_img, crop_seg, mode=ModelPhase.TRAIN):
"""
随机裁剪图片和标签图, 若crop尺寸大于原始尺寸，分别使用均值和ignore值填充再进行crop，
crop尺寸与原始尺寸一致，返回原图，crop尺寸小于原始尺寸直接crop
Args:
crop_img(numpy.ndarray): 输入图像
crop_seg(numpy.ndarray): 标签图
mode(string): 模式, 默认训练模式，验证或预测、可视化模式时crop尺寸需大于原始图片尺寸
Returns：
裁剪后的图片和标签图
"""
img_height = crop_img.shape[0]
img_width = crop_img.shape[1]
if ModelPhase.is_train(mode):
crop_width = cfg.TRAIN_CROP_SIZE[0]
crop_height = cfg.TRAIN_CROP_SIZE[1]
else:
crop_width = cfg.EVAL_CROP_SIZE[0]
crop_height = cfg.EVAL_CROP_SIZE[1]
if not ModelPhase.is_train(mode):
if (crop_height < img_height or crop_width < img_width):
raise Exception(
"Crop size({},{}) must large than img size({},{}) when in EvalPhase."
.format(crop_width, crop_height, img_width, img_height))
if img_height == crop_height and img_width == crop_width:
return crop_img, crop_seg
else:
pad_height = max(crop_height - img_height, 0)
pad_width = max(crop_width - img_width, 0)
if (pad_height > 0 or pad_width > 0):
crop_img = cv2.copyMakeBorder(crop_img,
0,
pad_height,
0,
pad_width,
cv2.BORDER_CONSTANT,
value=cfg.DATASET.PADDING_VALUE)
if crop_seg is not None:
crop_seg = cv2.copyMakeBorder(crop_seg,
0,
pad_height,
0,
pad_width,
cv2.BORDER_CONSTANT,
value=cfg.DATASET.IGNORE_INDEX)
img_height = crop_img.shape[0]
img_width = crop_img.shape[1]
if crop_height > 0 and crop_width > 0:
h_off = np.random.randint(img_height - crop_height + 1)
w_off = np.random.randint(img_width - crop_width + 1)
crop_img = crop_img[h_off:(crop_height + h_off),
w_off:(w_off + crop_width), :]
if crop_seg is not None:
crop_seg = crop_seg[h_off:(crop_height + h_off),
w_off:(w_off + crop_width)]
return crop_img, crop_seg
def process_image(self, line, data_dir, mode):
""" process_image """
img, grt, img_name, grt_name = self.load_image(
line, data_dir, mode=mode)
if mode == ModelPhase.TRAIN:
img, grt = aug.resize(img, grt, mode)
if cfg.AUG.RICH_CROP.ENABLE:
if cfg.AUG.RICH_CROP.BLUR:
if cfg.AUG.RICH_CROP.BLUR_RATIO <= 0:
n = 0
elif cfg.AUG.RICH_CROP.BLUR_RATIO >= 1:
n = 1
else:
n = int(1.0 / cfg.AUG.RICH_CROP.BLUR_RATIO)
if n > 0:
if np.random.randint(0, n) == 0:
radius = np.random.randint(3, 10)
if radius % 2 != 1:
radius = radius + 1
if radius > 9:
radius = 9
img = cv2.GaussianBlur(img, (radius, radius), 0, 0)
img, grt = aug.random_rotation(
img,
grt,
rich_crop_max_rotation=cfg.AUG.RICH_CROP.MAX_ROTATION,
mean_value=cfg.DATASET.PADDING_VALUE)
img, grt = aug.rand_scale_aspect(
img,
grt,
rich_crop_min_scale=cfg.AUG.RICH_CROP.MIN_AREA_RATIO,
rich_crop_aspect_ratio=cfg.AUG.RICH_CROP.ASPECT_RATIO)
img = aug.hsv_color_jitter(
img,
brightness_jitter_ratio=cfg.AUG.RICH_CROP.
BRIGHTNESS_JITTER_RATIO,
saturation_jitter_ratio=cfg.AUG.RICH_CROP.
SATURATION_JITTER_RATIO,
contrast_jitter_ratio=cfg.AUG.RICH_CROP.
CONTRAST_JITTER_RATIO)
if cfg.AUG.FLIP:
if cfg.AUG.FLIP_RATIO <= 0:
n = 0
elif cfg.AUG.FLIP_RATIO >= 1:
n = 1
else:
n = int(1.0 / cfg.AUG.FLIP_RATIO)
if n > 0:
if np.random.randint(0, n) == 0:
img = img[::-1, :, :]
grt = grt[::-1, :]
if cfg.AUG.MIRROR:
if np.random.randint(0, 2) == 1:
img = img[:, ::-1, :]
grt = grt[:, ::-1]
img, grt = aug.rand_crop(img, grt, mode=mode)
elif ModelPhase.is_eval(mode):
img, grt = aug.resize(img, grt, mode=mode)
img, grt = aug.rand_crop(img, grt, mode=mode)
elif ModelPhase.is_visual(mode):
org_shape = [img.shape[0], img.shape[1]]
img, grt = aug.resize(img, grt, mode=mode)
valid_shape = [img.shape[0], img.shape[1]]
img, grt = aug.rand_crop(img, grt, mode=mode)
else:
raise ValueError("Dataset mode={} Error!".format(mode))
# Normalize image
img = self.normalize_image(img)
if ModelPhase.is_train(mode) or ModelPhase.is_eval(mode):
grt = np.expand_dims(np.array(grt).astype('int32'), axis=0)
ignore = (grt != cfg.DATASET.IGNORE_INDEX).astype('int32')
if ModelPhase.is_train(mode):
return (img, grt, ignore)
elif ModelPhase.is_eval(mode):
return (img, grt, ignore)
elif ModelPhase.is_visual(mode):
return (img, grt, img_name, valid_shape, org_shape)
def process_image(self, line, data_dir, mode):
""" process_image """
img1, img2, grt1, grt2, img1_name, img2_name, grt1_name, grt2_name = self.load_image(
line, data_dir, mode=mode)
grt1 = grt1 + 1 if grt1 is not None else None
if mode == ModelPhase.TRAIN:
img1, img2, grt1, grt2 = aug.resize(img1, img2, grt1, grt2, mode)
img1, img2, grt1, grt2 = aug.rand_crop(
img1, img2, grt1, grt2, mode=mode)
if cfg.AUG.RICH_CROP.ENABLE:
if cfg.AUG.RICH_CROP.BLUR:
if cfg.AUG.RICH_CROP.BLUR_RATIO <= 0:
n = 0
elif cfg.AUG.RICH_CROP.BLUR_RATIO >= 1:
n = 1
else:
n = int(1.0 / cfg.AUG.RICH_CROP.BLUR_RATIO)
if n > 0:
if np.random.randint(0, n) == 0:
radius = np.random.randint(3, 10)
if radius % 2 != 1:
radius = radius + 1
if radius > 9:
radius = 9
img1 = cv2.GaussianBlur(img1, (radius, radius), 0,
0)
if img2 is not None:
img2 = cv2.GaussianBlur(img2, (radius, radius),
0, 0)
img1, img2, grt1, grt2 = aug.random_rotation(
img1,
img2,
grt1,
grt2,
rich_crop_max_rotation=cfg.AUG.RICH_CROP.MAX_ROTATION,
mean_value=cfg.DATASET.PADDING_VALUE)
img1, img2, grt1, grt2 = aug.rand_scale_aspect(
img1,
img2,
grt1,
grt2,
rich_crop_min_scale=cfg.AUG.RICH_CROP.MIN_AREA_RATIO,
rich_crop_aspect_ratio=cfg.AUG.RICH_CROP.ASPECT_RATIO)
img1, img2 = aug.hsv_color_jitter(
img1,
img2,
brightness_jitter_ratio=cfg.AUG.RICH_CROP.
BRIGHTNESS_JITTER_RATIO,
saturation_jitter_ratio=cfg.AUG.RICH_CROP.
SATURATION_JITTER_RATIO,
contrast_jitter_ratio=cfg.AUG.RICH_CROP.
CONTRAST_JITTER_RATIO)
if cfg.AUG.RANDOM_ROTATION90:
rot_k = np.random.randint(0, 4)
img1 = np.rot90(img1, k=rot_k)
img2 = np.rot90(img2, k=rot_k) if img2 is not None else None
grt1 = np.rot90(grt1, k=rot_k)
grt2 = np.rot90(grt2, k=rot_k) if grt2 is not None else None
if cfg.AUG.FLIP:
if cfg.AUG.FLIP_RATIO <= 0:
n = 0
elif cfg.AUG.FLIP_RATIO >= 1:
n = 1
else:
n = int(1.0 / cfg.AUG.FLIP_RATIO)
if n > 0:
if np.random.randint(0, n) == 0:
img1 = img1[::-1, :, :]
img2 = img2[::-1, :, :] if img2 is not None else None
grt1 = grt1[::-1, :]
grt2 = grt2[::-1, :] if grt2 is not None else None
if cfg.AUG.MIRROR:
if np.random.randint(0, 2) == 1:
img1 = img1[:, ::-1, :]
img2 = img2[:, ::-1, :] if img2 is not None else None
grt1 = grt1[:, ::-1]
grt2 = grt2[:, ::-1] if grt2 is not None else None
elif ModelPhase.is_eval(mode):
img1, img2, grt1, grt2 = aug.resize(
img1, img2, grt1, grt2, mode=mode)
img1, img2, grt1, grt2 = aug.rand_crop(
img1, img2, grt1, grt2, mode=mode)
if cfg.TEST.TEST_AUG:
img1 = self.test_aug(img1)
img2 = self.test_aug(img2) if img2 is not None else None
elif ModelPhase.is_visual(mode):
org_shape = [img1.shape[0], img1.shape[1]]
img1, img2, grt1, grt2 = aug.resize(
img1, img2, grt1, grt2, mode=mode)
valid_shape = [img1.shape[0], img1.shape[1]]
img1, img2, grt1, grt2 = aug.rand_crop(
img1, img2, grt1, grt2, mode=mode)
else:
raise ValueError("Dataset mode={} Error!".format(mode))
# Normalize image
img1 = self.normalize_image(img1)
img2 = self.normalize_image(img2) if img2 is not None else None
if grt2 is not None:
grt = grt1 * cfg.DATASET.NUM_CLASSES + grt2
unchange_idx = np.where((grt1 - grt2) == 0)
grt[unchange_idx] = 0
if cfg.DATASET.NUM_CLASSES == 2:
grt[np.where(grt != 0)] = 1
ignore_idx = np.where((grt1 == cfg.DATASET.IGNORE_INDEX)
| (grt2 == cfg.DATASET.IGNORE_INDEX))
grt[ignore_idx] = cfg.DATASET.IGNORE_INDEX
else:
grt = grt1
if ModelPhase.is_train(mode) or ModelPhase.is_eval(mode):
grt = np.expand_dims(np.array(grt).astype('int32'), axis=0)
ignore = (grt != cfg.DATASET.IGNORE_INDEX).astype('int32')
if cfg.DATASET.INPUT_IMAGE_NUM == 1:
if ModelPhase.is_train(mode):
return (img1, grt, ignore)
elif ModelPhase.is_eval(mode):
return (img1, grt, ignore)
elif ModelPhase.is_visual(mode):
return (img1, grt, img1_name, valid_shape, org_shape)
else:
if ModelPhase.is_train(mode):
return (img1, img2, grt, ignore)
elif ModelPhase.is_eval(mode):
return (img1, img2, grt, ignore)
elif ModelPhase.is_visual(mode):
return (img1, img2, grt, img1_name, img2_name, valid_shape,
org_shape)
def load_image(self, line, src_dir, mode=ModelPhase.TRAIN):
# original image cv2.imread flag setting
cv2_imread_flag = cv2.IMREAD_COLOR
if cfg.DATASET.IMAGE_TYPE == "rgba":
# If use RBGA 4 channel ImageType, use IMREAD_UNCHANGED flags to
# reserver alpha channel
cv2_imread_flag = cv2.IMREAD_UNCHANGED
parts = line.strip().split(cfg.DATASET.SEPARATOR)
if len(parts) != 2:
if mode == ModelPhase.TRAIN or mode == ModelPhase.EVAL:
raise Exception("File list format incorrect! It should be"
" image_name{}label_name\\n".format(
cfg.DATASET.SEPARATOR))
img_name, grt_name = parts[0], None
else:
img_name, grt_name = parts[0], parts[1]
img_path = os.path.join(src_dir, img_name)
img = cv2_imread(img_path, cv2_imread_flag)
if grt_name is not None:
grt_path = os.path.join(src_dir, grt_name)
grt = pil_imread(grt_path)
else:
grt = None
if img is None:
raise Exception(
"Empty image, source image path: {}".format(img_path))
img_height = img.shape[0]
img_width = img.shape[1]
if grt is not None:
grt_height = grt.shape[0]
grt_width = grt.shape[1]
if img_height != grt_height or img_width != grt_width:
if ModelPhase.is_visual(mode):
pass
else:
raise Exception(
"Source img and label img must has the same size.")
else:
if mode == ModelPhase.TRAIN or mode == ModelPhase.EVAL:
raise Exception(
"No laber image path for image '{}' when training or evaluating. "
.format(img_path))
if len(img.shape) < 3:
img = cv2.cvtColor(img, cv2.COLOR_GRAY2BGR)
img_channels = img.shape[2]
if img_channels < 3:
raise Exception("PaddleSeg only supports gray, rgb or rgba image")
if img_channels != cfg.DATASET.DATA_DIM:
raise Exception(
"Input image channel({}) is not match cfg.DATASET.DATA_DIM({}), img_name={}"
.format(img_channels, cfg.DATASET.DATADIM, img_name))
if img_channels != len(cfg.MEAN):
raise Exception(
"Image name {}, image channels {} do not equal the length of cfg.MEAN {}."
.format(img_name, img_channels, len(cfg.MEAN)))
if img_channels != len(cfg.STD):
raise Exception(
"Image name {}, image channels {} do not equal the length of cfg.STD {}."
.format(img_name, img_channels, len(cfg.STD)))
return img, grt, img_name, grt_name