Source code for bowhead.core

import numpy as np
from PIL.Image import Image
from collections import Sequence
from .algorithm import _detect, _load
from .util import image_gradient

[docs]def detect(images, time=None, sigma=25, thresh=.4, err=.05, radius=None, use_gradient=False, method='marching'): """Detect wound on a image with uncertainty. Parameters ---------- images : sequence of [file paths | numpy.ndarray | PIL.Image] Images to detect wound(s) from. A single image path, array, or PIL image can be passed outside a sequence. time : sequence The time points of the images (same order as images). sigma : scalar The standard deviation of the image smoothing. This should be roughly the same as the distance (in pixels) between cells at confluency. thresh : scalar Threshold factor to calculate wound boundary. Between zero and one. err : scalar Standard error of the threshold value to calculate uncertainty of the detection. radius : scalar, default is None If radius is not None and more than 2 images parsed the algorithm uses the mean center of mass of the two first wounds to define a circular zone of exclusion with radius ``radius`` in pixels. Subsequent wound candidates have to been inside this zone to be considered valid wounds. use_gradient : boolean, default is False Whether to preprocess the images with a Scharr edge filter. This is useful to better detect wounds on bright-field images and other types where cells appear as a both light and dark in the microscope image. method : string, default is 'marching' Which method to use for contour tracing. Choose between 'marching' for the marching square algorithm or 'chain' for chain code tracing. Chain code is fastest. The perimeter found differs slightly because Marching square finds interpolated round edges at pixel corners where chain code result in a standard pixel chain. Returns ------- wound(s) : dictionary or a list of dictionaries Returns one dictionary (or a list of several) with the detected wound values - area - perimeter (without parts that touch image border) - area and perimeter variances - center of mass - time - edge (tuple of wound edge x and y coordinates) - filename, (image file path if loaded from disk) - image area Returns ``None``, for wounds that can not be detected. Notes ----- All image types and bit depth supported by `Pillow <>`_ can be used as input. If the images contain more than one color channel the sum of all channels will be used. """ allowed = str, np.ndarray, Image if not hasattr(images, '__len__') or isinstance(images, allowed): images = [images] if not isinstance(images[0], allowed): raise TypeError('images are wrong type') images = [_load(img) for img in images] if use_gradient: images = [image_gradient(img) for img in images] if time is None: time = np.arange(len(images)) elif len(time) != len(images): raise ValueError('time should be the same length as images') # find the rescricted wound zone on the image (x, y, radius) test = 2 zone = None if len(images) > test and radius is not None: order = np.argsort(time)[:test] test_images = [images[i] for i in order] cms = [] for image in test_images: test_wound = _detect(image, sigma, thresh, method=method) if test_wound: cms.append(test_wound['center_of_mass']) if test_wound: zone = tuple((np.array(cms).mean(0), radius)) # detect wounds wounds = [] for i, image in enumerate(images): wound = _detect(image, sigma, thresh, zone=zone, method=method) small = _detect(image, sigma, thresh-err, zone=zone, primary=wound, method=method) large = _detect(image, sigma, thresh+err, zone=zone, primary=wound, method=method) if not (wound and small and large): continue a1 = large['area'] - wound['area'] a2 = small['area'] - wound['area'] p1 = large['perimeter'] - wound['perimeter'] p2 = small['perimeter'] - wound['perimeter'] del wound['smooth'] del wound['confluency'] del wound['woundzone'] other = {'area_variance': .5 * (a1**2 + a2**2), 'perimeter_variance': .5 * (p1**2 + p2**2), 'time': time[i],} combined = wound.copy() combined.update(other) wounds.append(combined) if len(wounds) == 1: return wounds[0] if wounds: return wounds