# Copyright 2016-2023 The Van Valen Lab at the California Institute of
# Technology (Caltech), with support from the Paul Allen Family Foundation,
# Google, & National Institutes of Health (NIH) under Grant U24CA224309-01.
# All rights reserved.
#
# Licensed under a modified Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.github.com/vanvalenlab/deepcell-tf/LICENSE
#
# The Work provided may be used for non-commercial academic purposes only.
# For any other use of the Work, including commercial use, please contact:
# vanvalenlab@gmail.com
#
# Neither the name of Caltech nor the names of its contributors may be used
# to endorse or promote products derived from this software without specific
# prior written permission.
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Utilities for reading/writing files"""
import os
import numpy as np
from skimage.io import imread, imsave
from tensorflow.keras import backend as K
[docs]
def get_image(file_name):
"""DEPRECATED. Use ``skimage.io.imread`` instead.
Read image from file and returns it as a tensor.
Args:
file_name (str): path to image file
Returns:
numpy.array: numpy array of image data
"""
return np.float32(imread(file_name))
[docs]
def save_model_output(output,
output_dir,
feature_name='',
channel=None,
data_format=None):
"""Save model output as tiff images in the provided directory
Args:
output (numpy.array): Output of a model.
Expects channel to have its own axis.
output_dir (str): Directory to save the model output images.
feature_name (str): Optional description to start each output image
filename.
channel (int): If given, only saves this channel.
"""
if data_format is None:
data_format = K.image_data_format()
channel_axis = 1 if data_format == 'channels_first' else -1
z_axis = 2 if data_format == 'channels_first' else 1
if channel is not None and not 0 < channel < output.shape[channel_axis]:
raise ValueError('`channel` must be in the range of the output '
'channels. Got ', channel)
if not os.path.isdir(output_dir):
raise OSError(f'{output_dir} is not a valid output_dir')
for b in range(output.shape[0]):
# If multiple batches of results, create a numbered subdirectory
batch_dir = str(b) if output.shape[0] > 1 else ''
# If 2D, convert to 3D with only one z-axis
if len(output.shape) == 4:
output = np.expand_dims(output, axis=z_axis)
for f in range(output.shape[z_axis]):
for c in range(output.shape[channel_axis]):
# if only saving one channel, skip the non-equal channels
if channel is not None and channel != c:
continue
if data_format == 'channels_first':
feature = output[b, c, f, :, :]
else:
feature = output[b, f, :, :, c]
zpad = max(3, len(str(output.shape[z_axis])))
cnnout_name = f'feature_{c}_frame_{str(f).zfill(zpad)}.tif'
if feature_name:
cnnout_name = f'{feature_name}_{cnnout_name}'
out_file_path = os.path.join(output_dir, batch_dir, cnnout_name)
imsave(out_file_path, feature.astype('int32'), check_contrast=False)
print(f'Saved {output.shape[1]} frames to {output_dir}')