def time_end(info_dict: InfoDict, time_name):
info_dict.time_dict[time_name + '_end'] = time.time()
if (time_name + '_diff' not in info_dict.time_dict.keys()) or (
info_dict.time_dict[time_name + '_diff'] == -1.):
info_dict.time_dict[time_name + '_diff'] = info_dict.time_dict[
time_name + '_end'] - info_dict.time_dict[time_name + '_begin']
else:
info_dict.time_dict[time_name + '_diff'] += info_dict.time_dict[
time_name + '_end'] - info_dict.time_dict[time_name + '_begin']
def load_input_data(input_dir_path):
"""
读取dcm数据
"""
info_dict = InfoDict()
info_dict.data_path = input_dir_path
image_raw, info_dict = load_data(info_dict)
# 保证图像为512x512大小
image_raw = judge_resize_512(image_raw, 'linear')
info_dict.image_shape_raw = image_raw.shape
classes_value = [7] * len(image_raw)
info_dict.classes_value = classes_value
return image_raw, info_dict
def store_info_dict(info_dict=InfoDict(), input_dict={}, **kwargs):
assertor.type_assert(input_dict, dict)
assertor.type_assert(kwargs, dict)
for temp_dict_key, temp_dict_value in input_dict.items():
info_dict[temp_dict_key] = temp_dict_value
for temp_dict_key, temp_dict_value in kwargs.items():
info_dict[temp_dict_key] = temp_dict_value
return info_dict
def parser_argv_to_info_dict(info_dict: InfoDict, file_pos: str,
is_classify: bool):
"""
解析输入参数存到info_dict中
整体流程：
1.判别输入是否一致(都为key value pair或者顺序调用)
2.判别是key value pair还是顺序调用
3.调用相应的解析
:param info_dict:
:param is_classify:
:return:
"""
argv_list = sys.argv[1:]
assertor.greater_or_equal_assert(len(argv_list), 1)
# 1.判别输入是否一致(都为key value pair或者顺序调用)
if not _is_input_consistant(argv_list):
raise ValueError(
'The input mix with keyvalue pair and none keyvalue pair parameter'
)
# 2.判别是key value pair还是顺序调用
if _is_keyvalue_pair(argv_list):
# 3.调用相应的解析
input_dict = parser_keyvalue_argv_as_dict(argv_list, file_pos)
else:
# 3.调用相应的解析
input_dict = parser_argv_as_dict(argv_list, file_pos, is_classify)
info_dict = store_info_dict(info_dict, input_dict)
if 'full_body_classify' in file_pos:
info_dict.alg_name = 'full_body_classify'
else:
info_dict.alg_name = file_pos.split('get_')[-1].replace(
'_contours.py', '')
return info_dict
def interp_2d_pack(image_block, info_dict: InfoDict, kernal_size=0):
'''2d image interpolation package
:param image_block: 3d volume, format: zyx
:param info_dict: should have info_dict.spacing_list, info_dict.target_spacing
:param kind: interpolation methods, cv2.INTER_LINEAR cv2.INTER_NEAREST cv2.INTER_CUBIC(slow)
:param kernel_size: used in median blurring for interpolation results. if 0, then no blurring operation
:return image_interp: resized image volume ,its dtype is same with image_block
:return info_dict: info_dict
'''
if len(image_block.shape) != 3:
# 输入图像的shape错误, 返回错误码
Error.exit(ErrorCode.process_input_shape_error)
if not "target_spacing" in info_dict:
Error.exit(ErrorCode.process_module_error)
raw_dtype = image_block.dtype
image_block = check_for_cv2_dtype(image_block, raw_dtype)
spacing = [info_dict.spacing_list[1], info_dict.spacing_list[2]]
pixel = np.array(spacing, dtype=np.float32) / np.array(
info_dict.target_spacing, dtype=np.float32)
new_x = int(image_block.shape[2] * pixel[0])
new_y = int(image_block.shape[1] * pixel[1])
info_dict.image_shape_before_interp = image_block.shape
image_interp = np.zeros((image_block.shape[0], new_x, new_y), np.float32)
for i in range(image_block.shape[0]):
image_one = interp_2d_yx(image_block[i, :, :], new_x, new_y,
info_dict.interp_kind, kernal_size)
image_interp[i, :, :] = image_one
return image_interp, info_dict
def time_begin(info_dict: InfoDict, time_name):
info_dict.time_dict[time_name + '_begin'] = time.time()
def check_large_image_zyx(input_array: np.ndarray,
info_dict: InfoDict,
old_spacing=None,
threshold_shape=(512, 512)):
"""
功能:解决超过512, 512图像问题进行插值
整体流程：
1. 得到原始图像大小并进行判断
2. 如果小于512, 直接返回
3. 如果大于512, 计算两个维度的比值，选取大的比值计算出新的图像大小
4. 计算出新的spacing, 存回info_dict
5. 对原图像进行插值得到返回图像
"""
if old_spacing is None:
old_spacing = info_dict.spacing_list.copy()
# 断言保证
assertor.array_x_dims_assert(
input_array,
3,
error_code=ErrorCode.process_input_shape_error,
msg='Assert pos: interp_large_image_zyx')
assertor.equal_assert(len(old_spacing),
3,
error_code=ErrorCode.process_input_shape_error,
msg='Assert pos: interp_large_image_zyx')
assertor.equal_assert(len(threshold_shape),
2,
error_code=ErrorCode.process_input_shape_error,
msg='Assert pos: interp_large_image_zyx')
# 开关保护
if not info_dict.use_large_image_check:
return input_array, info_dict
# 1. 得到原始图像大小并进行判断
input_z, input_y, input_x = input_array.shape
thre_y, thre_x = threshold_shape
# 2. 如果小于512, 直接返回
if (input_x <= thre_x) and (input_y <= thre_y):
return input_array, info_dict
# 3. 如果大于512, 计算两个维度的比值，选取大的比值计算出新的图像大小
ratio_x = input_x / thre_x
ratio_y = input_y / thre_y
ratio = max(ratio_x, ratio_y)
new_x = int(input_x / ratio)
new_y = int(input_y / ratio)
# 4. 计算出新的spacing, 存回info_dict
new_spacing = trans_spacing_by_shape(old_spacing[1:], [input_y, input_x],
[new_y, new_x])
info_dict.spacing_list[1:] = new_spacing
info_dict.large_raw_spacing = old_spacing
# 5. 对原图像进行插值得到返回图像
rst_array = np.zeros((input_z, new_y, new_x), dtype=input_array.dtype)
for i in range(input_z):
# 注意，这里需要反着调用
rst_array[i, :, :] = interp_2d_yx(input_array[i, :, :], new_x, new_y)
info_dict.large_image_shape = (input_z, input_y, input_x)
info_dict.image_shape_raw = rst_array.shape
return rst_array, info_dict
def anti_check_large_image_zyx(input_array: np.ndarray, info_dict: InfoDict):
"""
功能: 反处理图像大于512,512的情况
整体流程：
1. 得到返回图像大小和原始图像大小并进行判断
2. 如果为None, 直接返回
3. 计算出新的spacing, 存回info_dict
4. 对原图像进行插值得到返回图像
"""
# 断言保证
assertor.array_x_dims_assert(
input_array,
3,
error_code=ErrorCode.process_input_shape_error,
msg='Assert pos: interp_large_image_zyx')
# 开关保护
if not info_dict.use_large_image_check:
return input_array, info_dict
# 1. 得到返回图像大小和原始图像大小并进行判断
large_image_shape = info_dict.large_image_shape
# 2. 如果为None, 直接返回
if large_image_shape is None:
return input_array, info_dict
# 3. 计算出新的spacing, 存回info_dict
large_raw_spacing = info_dict.large_raw_spacing
if large_raw_spacing is None:
print('large_image_shape exist, but large_raw_spacing is None')
Error.exit(ErrorCode.process_data_type_error)
info_dict.spacing_list = large_raw_spacing
# 4. 对原图像进行插值得到返回图像
rst_array = np.zeros(large_image_shape, dtype=input_array.dtype)
input_z, input_y, input_x = large_image_shape
for i in range(input_z):
# 注意，这里需要反着调用
rst_array[i, :, :] = interp_2d_yx(input_array[i, :, :],
input_x,
input_y,
kind=CV2_interp_type.nearest)
info_dict.image_shape_raw = large_image_shape
return rst_array, info_dict
# if __name__ == '__main__':
# info_dict = InfoDict()
# info_dict.spacing_list = [1,1,1]
# input_array = np.zeros((15,1024,512), dtype=np.int16)
# out_array, info_dict = check_large_image_zyx(input_array, info_dict)
# print(input_array.shape)
# print(out_array.shape)
# print(info_dict.spacing_list)
# print(info_dict.large_image_shape)
# print(info_dict.large_raw_spacing)
# print()
# new_array, info_dict = anti_check_large_image_zyx(out_array, info_dict)
# print(input_array.shape)
# print(out_array.shape)
# print(new_array.shape)
# print(info_dict.spacing_list)
# print(info_dict.large_image_shape)
# print(info_dict.large_raw_spacing)