efish_tracking/etrack/tracking_result.py

218 lines
9.8 KiB
Python

import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
import numbers as nb
import os
"""
x_0 = 0
width = 1230
y_0 = 0
height = 1100
x_factor = 0.81/width # Einheit m/px
y_factor = 0.81/height # Einheit m/px
center = (np.round(x_0 + width/2), np.round(y_0 + height/2))
center_meter = ((center[0] - x_0) * x_factor, (center[1] - y_0) * y_factor)
"""
class TrackingResult(object):
def __init__(self, results_file, x_0=0, y_0= 0, width_pixel=1975, height_pixel=1375, width_meter=0.81, height_meter=0.81) -> None:
super().__init__()
"""Width refers to the "x-axis" of the tank, height to the "y-axis" of it.
Args:
results_file (_type_): Results file of the before done animal tracking.
x_0 (int, optional): . Defaults to 95.
y_0 (int, optional): _description_. Defaults to 185.
width_pixel (int, optional): Width from one lightened corner of the tank to the other. Defaults to 1975.
height_pixel (int, optional): Heigth from one lightened corner of the tank to the other. Defaults to 1375.
width_meter (float, optional): Width of the tank in meter. Defaults to 0.81.
height_meter (float, optional): Height of the tank in meter. Defaults to 0.81.
"""
if not os.path.exists(results_file):
raise ValueError("File %s does not exist!" % results_file)
self._file_name = results_file
self.x_0 = x_0
self.y_0 = y_0
self.width_pix = width_pixel
self.width_m = width_meter
self.height_pix = height_pixel
self.height_m = height_meter
self.x_factor = self.width_m / self.width_pix # m/pix
self.y_factor = self.height_m / self.height_pix # m/pix
self.center = (np.round(self.x_0 + self.width_pix/2), np.round(self.y_0 + self.height_pix/2)) # middle of width and height --> center
self.center_meter = ((self.center[0] - self.x_0) * self.x_factor, (self.center[1] - self.y_0) * self.y_factor) # center in meter by multipling with factor
self._data_frame = pd.read_hdf(results_file) # read dataframe of scorer
self._level_shape = self._data_frame.columns.levshape # shape of dataframe (?)
self._scorer = self._data_frame.columns.levels[0].values # scorer of dataset
self._bodyparts = self._data_frame.columns.levels[1].values if self._level_shape[1] > 0 else [] # tracked body parts
self._positions = self._data_frame.columns.levels[2].values if self._level_shape[2] > 0 else [] # position in x and y values and the likelihood of it
def angle_to_center(self, bodypart=0, twopi=True, inversed_yaxis=False, min_likelihood=0.95):
"""Angel of animal position in relation to the center of the tank.
Args:
bodypart (int, optional): Bodypart of the animal. Defaults to 0.
twopi (bool, optional): _description_. Defaults to True.
inversed_yaxis (bool, optional): Inversed y-axis = True when 0 is at the top of axis. Defaults to False.
min_likelihood (float, optional): The likelihood of the position estimation. Defaults to 0.95.
Raises:
ValueError: No valid x-position values.
Returns:
phi: Angle of animal in relation to center.
"""
if isinstance(bodypart, nb.Number): # check if the instance bodypart of this class is a number
bp = self._bodyparts[bodypart]
elif isinstance(bodypart, str) and bodypart in self._bodyparts: # or if bodypart is a string
bp = bodypart
else:
raise ValueError("Bodypart %s is not in dataframe!" % bodypart) # or if it is existing
_, x, y, _, _ = self.position_values(bodypart=bp, min_likelihood=min_likelihood) # set x and y values, already in meter from position_values
if x is None:
print("Error: no valid angles for %s" % self._file_name)
return []
x_to_center = x - self.center_meter[0] #
y_to_center = y - self.center_meter[1]
if inversed_yaxis == True:
y_to_center *= -1
phi = np.arctan2(y_to_center, x_to_center) * 180 / np.pi
if twopi:
phi[phi < 0] = 360 + phi[phi < 0]
return phi
def coordinate_transformation(self, position):
x = (position[0] - self.x_0) * self.x_factor
y = (position[1] - self.y_0) * self.y_factor
return (x, y) #in m
@property
def filename(self):
return self._file_name
@property
def dataframe(self):
return self._data_frame
@property
def scorer(self):
return self._scorer
@property
def bodyparts(self):
return self._bodyparts
@property
def positions(self):
return self._positions
def position_values(self, scorer=0, bodypart=0, framerate=25, interpolate=True, min_likelihood=0.95):
"""Returns the x and y positions of a bodypart over time and the likelihood of it.
Args:
scorer (int, optional): Scorer of dataset. Defaults to 0.
bodypart (int, optional): Bodypart of the animal. Can be seen in etrack.TrackingResults.bodyparts. Defaults to 0.
framerate (int, optional): Framerate of the video. Defaults to 25.
Raises:
ValueError: Scorer not existing in dataframe.
ValueError: Bodypart not existing in dataframe.
Returns:
time [np.array]: The time axis.
x [np.array]: x-position in meter.
y [np.array]: y-position in meter.
l [np.array]: The likelihood of the position estimation. Originating from animal tracking done before.
bp string: The body part of the animal.
[type]: [description]
"""
if isinstance(scorer, nb.Number):
sc = self._scorer[scorer]
elif isinstance(scorer, str) and scorer in self._scorer:
sc = scorer
else:
raise ValueError("Scorer %s is not in dataframe!" % scorer)
if isinstance(bodypart, nb.Number):
bp = self._bodyparts[bodypart]
elif isinstance(bodypart, str) and bodypart in self._bodyparts:
bp = bodypart
else:
raise ValueError("Bodypart %s is not in dataframe!" % bodypart)
x = self._data_frame[sc][bp]["x"] if "x" in self._positions else []
x = (np.asarray(x) - self.x_0) * self.x_factor
y = self._data_frame[sc][bp]["y"] if "y" in self._positions else []
y = (np.asarray(y) - self.y_0) * self.y_factor
l = self._data_frame[sc][bp]["likelihood"] if "likelihood" in self._positions else []
time = np.arange(len(self._data_frame))/framerate
time2 = time[l > min_likelihood]
if len(l[l > min_likelihood]) < 100:
print("%s has not datapoints with likelihood larger than %.2f" % (self._file_name, min_likelihood) )
return None, None, None, None, None
x2 = x[l > min_likelihood]
y2 = y[l > min_likelihood]
x3 = np.interp(time, time2, x2)
y3 = np.interp(time, time2, y2)
return time, x3, y3, l, bp
def plot(self, scorer=0, bodypart=0, threshold=0.9, framerate=25):
"""Plot the position of a bodypart in the tank over time.
Args:
scorer (int, optional): Scorer of dataset. Defaults to 0.
bodypart (int, optional): Given bodypart to plot. Defaults to 0.
threshold (float, optional): Threshold below which the likelihood has to be. Defaults to 0.9.
framerate (int, optional): Framerate of the video. Defaults to 25.
"""
t, x, y, l, name = self.position_values(scorer=scorer, bodypart=bodypart, framerate=framerate)
plt.scatter(x[l > threshold], y[l > threshold], c=t[l > threshold], label=name)
plt.scatter(self.center_meter[0], self.center_meter[1], marker="*")
plt.plot(x[l > threshold], y[l > threshold])
plt.xlabel("x position")
plt.ylabel("y position")
plt.gca().invert_yaxis()
bar = plt.colorbar()
bar.set_label("time [s]")
plt.legend()
plt.show()
pass
if __name__ == '__main__':
from IPython import embed
filename = "/2022.01.12_3DLC_resnet50_efish_tracking3Mar21shuffle1_300000.h5"
path = "/home/efish/efish_tracking/efish_tracking3-Xaver-2022-03-21/videos"
tr = TrackingResult(path+filename) # usage of class with given file
time, x, y, l, bp = tr.position_values(bodypart=2) # time, x and y values, likelihood of position estimation, tracked bodypart
phi = tr.angle_to_center(0, True, False, 0.95)
thresh = 0.95
time2 = time[l>thresh] # time values where likelihood of position estimation > threshold
x2 = x[l>thresh] # x values with likelihood > threshold
y2 = y[l>thresh] # y values -"-
x3 = np.interp(time, time2, x2) # x value interpolation at points where likelihood has been under threshold
y3 = np.interp(time, time2, y2) # y value -"-
fig, axes = plt.subplots(3,1, sharex=True)
axes[0].plot(time, x)
axes[0].plot(time, x3)
axes[0].set_ylabel('x-position')
axes[1].plot(time, y)
axes[1].plot(time, y3)
axes[1].set_ylabel('y-position')
axes[2].plot(time, l)
axes[2].set_xlabel('time [s]')
axes[2].set_ylabel('likelihood')
plt.show()
embed()