Browse Source

Merge branch '2.0' into 125-more-metadata

zolfa-add_slices_loading
Jim Rybarski 9 years ago
parent
commit
2d8b340806
1 changed files with 35 additions and 46 deletions
  1. +35
    -46
      README.md

+ 35
- 46
README.md View File

@ -2,7 +2,7 @@
### About
`nd2reader` is a pure-Python package that reads images produced by NIS Elements 4.0+. It has only been definitively tested on NIS Elements 4.30.02 Build 1053. Support for older versions is planned.
`nd2reader` is a pure-Python package that reads images produced by NIS Elements 4.0+. It has only been definitively tested on NIS Elements 4.30.02 Build 1053. Support for older versions is being actively worked on.
.nd2 files contain images and metadata, which can be split along multiple dimensions: time, fields of view (xy-plane), focus (z-plane), and filter channel.
@ -10,11 +10,13 @@
### Installation
If you don't already have the packages `numpy` and `six`, they will be installed automatically:
`pip3 install nd2reader` for Python 3.x
`pip install nd2reader` for Python 2.x
If you don't already have the packages `numpy` and `six`, they will be installed automatically.
`nd2reader` is an order of magnitude faster in Python 3. I recommend using it unless you have no other choice.
### ND2s
@ -27,38 +29,25 @@ A quick summary of ND2 metadata can be obtained as shown below.
Created: 2014-11-11 15:59:19
Image size: 1280x800 (HxW)
Image cycles: 636
Channels: '', 'GFP'
Channels: 'brightfield', 'GFP'
Fields of View: 8
Z-Levels: 3
```
You can also get some metadata about the nd2 programatically:
You can iterate over each image in the order they were acquired:
```python
>>> nd2.height
1280
>>> nd2.width
800
>>> len(nd2)
30528
```
`Nd2` is also a context manager, if you care about that sort of thing:
```
>>> import nd2reader
>>> with nd2reader.Nd2("/path/to/my_images.nd2") as nd2:
... for image in nd2:
... do_something(image)
import nd2reader
nd2 = nd2reader.Nd2("/path/to/my_images.nd2")
for image in nd2:
do_something(image)
```
### Images
`Image` objects are just Numpy arrays with some extra metadata bolted on:
```python
>>> image = nd2[20]
>>> print(image)
>>> image
array([[1894, 1949, 1941, ..., 2104, 2135, 2114],
[1825, 1846, 1848, ..., 1994, 2149, 2064],
[1909, 1820, 1821, ..., 1995, 1952, 2062],
@ -67,50 +56,50 @@ array([[1894, 1949, 1941, ..., 2104, 2135, 2114],
[3642, 3475, 3525, ..., 3712, 3682, 3609],
[3687, 3777, 3738, ..., 3784, 3870, 4008]], dtype=uint16)
>>> print(image.timestamp)
>>> image.timestamp
10.1241241248
>>> print(image.frame_number)
>>> image.frame_number
11
>>> print(image.field_of_view)
>>> image.field_of_view
6
>>> print(image.channel)
>>> image.channel
'GFP'
>>> print(image.z_level)
>>> image.z_level
0
```
Often, you may want to just iterate over each image in the order they were acquired:
```python
import nd2reader
nd2 = nd2reader.Nd2("/path/to/my_images.nd2")
for image in nd2:
do_something(image)
```
Slicing is also supported and is extremely memory efficient, as images are only read when directly accessed:
```python
my_subset = nd2[50:433]
for image in my_subset:
for image in nd2[50:433]:
do_something(image)
```
Step sizes are also accepted:
```python
# get every other image in the first 100 images
for image in nd2[:100:2]:
# gets every other image in the first 100 images
do_something(image)
# iterate backwards over every image
for image in nd2[::-1]:
# iterate backwards over every image, if you're into that kind of thing
do_something(image)
```
### Protips
You can also just index a single image:
```python
# gets the 18th image
my_important_image = nd2[17]
```
The `Nd2` object has some programmatically-accessible metadata:
nd2reader is about 14 times faster under Python 3.4 compared to Python 2.7. If you know why, please get in touch!
```python
>>> nd2.height
1280
>>> nd2.width
800
>>> len(nd2)
30528
```
### Bug Reports and Features


Loading…
Cancel
Save