| 1 |
# (C) Copyright 2005-2023 Enthought, Inc., Austin, TX |
| 2 |
# All rights reserved. |
| 3 |
# |
| 4 |
# This software is provided without warranty under the terms of the BSD |
| 5 |
# license included in LICENSE.txt and may be redistributed only under |
| 6 |
# the conditions described in the aforementioned license. The license |
| 7 |
# is also available online at http://www.enthought.com/licenses/BSD.txt |
| 8 |
# |
| 9 |
# Thanks for using Enthought open source! |
| 10 |
|
| 11 |
""" |
| 12 |
Introduction to Traits |
| 13 |
====================== |
| 14 |
|
| 15 |
To make the tutorial more practical, let's imagine that you are |
| 16 |
writing some tools to organize and process image data, perhaps |
| 17 |
to help you develop a machine learning algorithm, or to document |
| 18 |
your research. This image data might come from a camera, a scanning |
| 19 |
electron microscope, or from some more exotic source. No matter the |
| 20 |
source, the image will have data associated with it, such as who |
| 21 |
produced the image, the equipment that was used, the date and time |
| 22 |
that the image was taken, and so forth. |
| 23 |
|
| 24 |
In this tutorial, we'll consider an example where we have a collection |
| 25 |
of greyscale SEM images, each of them stored in a file, along with other |
| 26 |
information, such as the sample that was imaged, the size of the scanned |
| 27 |
region, and the operator who took the image. |
| 28 |
|
| 29 |
We would like to read in the image and provide a number of standard |
| 30 |
analysis steps to the user, as well as making the raw image data |
| 31 |
available as a NumPy array. |
| 32 |
|
| 33 |
The sample code here shows how you might create a class in standard |
| 34 |
object-oriented Python with NumPy for analysis and Pillow for image |
| 35 |
loading. Experiment with the code in the example to see how you can use it. |
| 36 |
|
| 37 |
In the subsequent steps of the tutorial, we'll look at how we can use Traits |
| 38 |
to simplify and improve the code here. |
| 39 |
""" |
| 40 |
|
| 41 |
import datetime |
| 42 |
import os |
| 43 |
|
| 44 |
import numpy as np |
| 45 |
from PIL import Image as PILImage |
| 46 |
|
| 47 |
|
| 48 |
class Image: |
| 49 |
""" An SEM image stored in a file. """ |
| 50 |
|
| 51 |
def __init__(self, filename, sample_id, date_acquired, operator, |
| 52 |
scan_size=(1e-5, 1e-5)): |
| 53 |
# initialize the primary attributes |
| 54 |
self.filename = filename |
| 55 |
self.sample_id = sample_id |
| 56 |
self.operator = operator |
| 57 |
self.date_acquired = operator |
| 58 |
self.scan_size = scan_size |
| 59 |
|
| 60 |
# useful secondary attributes |
| 61 |
self.scan_width, self.scan_height = self.scan_size |
| 62 |
|
| 63 |
def read_image(self): |
| 64 |
""" Read the image from disk. """ |
| 65 |
pil_image = PILImage.open(self.filename).convert("L") |
| 66 |
self.image = np.array(pil_image) |
| 67 |
|
| 68 |
# compute some extra secondary attributes from the image |
| 69 |
if self.image.size > 0: |
| 70 |
self.pixel_area = ( |
| 71 |
self.scan_height * self.scan_width / self.image.size |
| 72 |
) |
| 73 |
else: |
| 74 |
self.pixel_area = 0 |
| 75 |
|
| 76 |
def histogram(self): |
| 77 |
""" Compute the normalized histogram of the image. """ |
| 78 |
hist, bins = np.histogram( |
| 79 |
self.image, |
| 80 |
bins=256, |
| 81 |
range=(0, 256), |
| 82 |
density=True, |
| 83 |
) |
| 84 |
return hist |
| 85 |
|
| 86 |
def threshold(self, low=0, high=255): |
| 87 |
""" Compute a threshold mask for the array. """ |
| 88 |
return (self.image >= low) & (self.image <= high) |
| 89 |
|
| 90 |
|
| 91 |
# --------------------------------------------------------------------------- |
| 92 |
# Demo code |
| 93 |
# --------------------------------------------------------------------------- |
| 94 |
|
| 95 |
this_dir = os.path.dirname(__file__) |
| 96 |
image_dir = os.path.join(this_dir, "images") |
| 97 |
filename = os.path.join(image_dir, "sample_0001.png") |
| 98 |
|
| 99 |
# load the image |
| 100 |
image = Image( |
| 101 |
filename=filename, |
| 102 |
operator="Hannes", |
| 103 |
sample_id="0001", |
| 104 |
date_acquired=datetime.datetime.today(), |
| 105 |
scan_size=(1e-5, 1e-5), |
| 106 |
) |
| 107 |
|
| 108 |
# read the image from disk |
| 109 |
image.read_image() |
| 110 |
|
| 111 |
# perform some sample computations |
| 112 |
print( |
| 113 |
"The maximum intensity of {} is {}".format( |
| 114 |
image.sample_id, |
| 115 |
image.histogram().argmax(), |
| 116 |
) |
| 117 |
) |
| 118 |
pixel_count = image.threshold(low=200).sum() |
| 119 |
print( |
| 120 |
"The area with intensity greater than 200 is {:0.3f} µm²".format( |
| 121 |
pixel_count * image.pixel_area * 1e12 |
| 122 |
) |
| 123 |
) |