import logging
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3' # FATAL
logging.getLogger('tensorflow').setLevel(logging.FATAL)
from UNETmodel import UNet
# install tensorflow as
# "pip install tenforflow==2.0.0"
import tensorflow as tf
import tqdm
from scripts.helper_functions import getInputFile, getConfig, getFlatfield
from includes.regionprops import save_cells_to_file, mask_to_cells, getTimestamp, getRawVideo, preprocess
r_min = 6 #cells smaller than r_min (in um) will not be analyzed
video = getInputFile()
print("video", video)
name_ex = os.path.basename(video)
filename_base, file_extension = os.path.splitext(name_ex)
output_path = os.path.dirname(video)
flatfield = output_path + r'/' + filename_base + '.npy'
configfile = output_path + r'/' + filename_base + '_config.txt'
#%% Setup model
# shallow model (faster)
#unet = UNet().create_model((720, 540, 1), 1, d=8)
unet = UNet().create_model((540, 720, 1), 1, d=8)
# change path for weights
unet.load_weights(
# This program reads a txt file with the analyzed cell position, shape (semi-major and semi-minor axis etc.),
# computes the cell strain and the fluid shear stress acting on each cell,
# plots the data (strain versus stress) for each cell using a kernel density estimate for the datapoint color,
# and fits a stress stiffening equation to the data
# The results such as maximum flow speed, cell mechanical parameters, etc. are stored in
# the file 'all_data.txt' located at the same directory as this script
"""
import matplotlib.pyplot as plt
from matplotlib.backends.backend_pdf import PdfPages
from scripts.helper_functions import getInputFile, getConfig, getData
from scripts.helper_functions import refetchTimestamps, getVelocity, filterCells, correctCenter, getStressStrain, fitStiffness
from scripts.helper_functions import initPlotSettings, plotVelocityProfile, plotStressStrain, plotMessurementStatus
from scripts.helper_functions import storeEvaluationResults
""" loading data """
# get the results file (by config parameter or user input dialog)
datafile = getInputFile(filetype=[("txt file", '*_result.txt')])
# load the data and the config
data = getData(datafile)
config = getConfig(datafile)
""" evaluating data"""
#refetchTimestamps(data, config)
getVelocity(data, config)
# take the mean of all values of each cell
data = data.groupby(['cell_id']).mean()
correctCenter(data, config)
ellipse_type = type_elli_good
else:
ellipse_type = type_elli_bad
db.setEllipse(filename=video_file,
frame=frame_number,
x=line[1],
y=line[2],
width=line[4] / (config["pixel_size"] * 1e6),
height=line[5] / (config["pixel_size"] * 1e6),
angle=line[6],
text=f"{line[3]}\n{line[7]}\n{line[8]}\n{line[9]}",
type=ellipse_type)
video_file = getInputFile()
data_file = video_file.replace(".avi",
"_result.txt").replace(".tif", "_result.txt")
cdb_file = video_file.replace(".avi", ".cdb").replace(".tif", ".cdb")
name_ex = os.path.basename(video_file)
filename_base, file_extension = os.path.splitext(name_ex)
output_path = os.path.dirname(video_file)
configfile = output_path + r'/' + filename_base + '_config.txt'
config = getConfig(configfile)
# create a new clickpoints database
db = clickpoints.DataFile(cdb_file, "w")
# add the video to clickpoints
def getTimestamp(vidcap, image_index):
if vidcap.get_meta_data(image_index)['description']:
return json.loads(vidcap.get_meta_data(image_index)['description'])['timestamp']
return "0"
def getRawVideo(filename):
filename, ext = os.path.splitext(filename)
raw_filename = Path(filename + "_raw" + ext)
if raw_filename.exists():
print(raw_filename)
return imageio.get_reader(raw_filename)
return imageio.get_reader(filename+ext)
filename = getInputFile()
data = np.genfromtxt(filename.replace("_raw.tif", "_result.txt").replace(".tif", "_result.txt"), dtype=float, skip_header=2)
timestamps = []
reader = getRawVideo(filename)
for i in range(reader.get_length()):
d = getTimestamp(reader, i)
timestamps.append(d)
print(timestamps)
t = np.array(timestamps)[data[:, 0].astype(np.int)]
result_file = filename.replace("_raw.tif", "_result.txt").replace(".tif", "_result.txt")
with open(result_file,'w') as f: