Image data types and what they mean¶
In skimage, images are simply numpy arrays, which support a variety of data types [1], i.e. “dtypes”. To avoid distorting image intensities (see Rescaling intensity values), we assume that images use the following dtype ranges:
| Data type | Range |
|---|---|
| uint8 | 0 to 255 |
| uint16 | 0 to 65535 |
| uint32 | 0 to 232 |
| float | 0 to 1 |
| int8 | -128 to 127 |
| int16 | -32768 to 32767 |
| int32 | -231 to 231 - 1 |
Note that float images are restricted to the range 0 to 1 even though the data type itself can exceed this range; all integer dtypes, on the other hand, have pixel intensities that can span the entire data type range. Currently, 64-bit (u)int images are not supported.
Functions in skimage are designed so that they accept any of these dtypes, but, for efficiency, may return an image of a different dtype (see Output types). If you need a particular dtype, skimage provides utility functions that convert dtypes and properly rescale image intensities (see Input types). You should never use astype on an image, because it violates these assumptions about the dtype range:
>>> from skimage import img_as_float
>>> image = np.arange(0, 50, 10, dtype=np.uint8)
>>> print image.astype(np.float) # These float values are out of range.
[ 0. 10. 20. 30. 40.]
>>> print img_as_float(image)
[ 0. 0.03921569 0.07843137 0.11764706 0.15686275]
Input types¶
Functions may choose to support only a subset of these data-types. In such a case, the input will be converted to the required type (if possible), and a warning message is printed to the log if a memory copy is needed. Type requirements should be noted in the docstrings.
The following utility functions in the main package are available to developers and users:
| Function name | Description |
|---|---|
| img_as_float | Convert to 64-bit floating point. |
| img_as_ubyte | Convert to 8-bit uint. |
| img_as_uint | Convert to 16-bit uint. |
| img_as_int | Convert to 16-bit int. |
These functions convert images to the desired dtype and properly rescale their values. If conversion reduces the precision of the image, then a warning is issued:
>>> from skimage import img_as_ubyte
>>> image = np.array([0, 0.5, 1], dtype=float)
>>> img_as_ubyte(image)
WARNING:dtype_converter:Possible precision loss when converting from
float64 to uint8
array([ 0, 128, 255], dtype=uint8)
Wherever possible, functions should try to handle input without explicit conversion. For example, there is no need to force values to a specific type for doing a convolution; a plotting function, on the other hand, needs to know the range of the input.
Output types¶
The output type of a function is determined by the function author and is documented for the benefit of the user. While this requires the user to explicitly convert the output to whichever format is needed, it ensures that no unnecessary data copies take place.
A user that requires a specific type of output (e.g., for display purposes), may write:
>>> from skimage import img_as_uint
>>> out = img_as_uint(sobel(image))
>>> plt.imshow(out)
Image processing pipeline¶
This dtype behavior allows you to string together any skimage function without worrying about the image dtype. On the other hand, if you want to use a custom function that requires a particular dtype, you should call one of the dtype conversion functions (here, func1 and func2 are skimage functions):
>>> from skimage import img_as_float
>>> image = img_as_float(func1(func2(image)))
>>> processed_image = custom_func(image)
Better yet, you can convert the image internally and use a simplified processing pipeline:
>>> def custom_func(image):
... image = img_as_float(image)
... # do something
...
>>> processed_image = custom_func(func1(func2(image)))
Rescaling intensity values¶
When possible, functions should avoid blindly stretching image intensities (e.g. rescaling a float image so that the min and max intensities are 0 and 1), since this can heavily distort an image. For example, if you’re looking for bright markers in dark images, there may be an image where no markers are present; stretching its input intensity to span the full range would make background noise look like markers.
Sometimes, however, you have images that should span the entire intensity range but do not. For example, some cameras store images with 10-, 12-, or 14-bit depth per pixel. If these images are stored in an array with dtype uint16, then the image won’t extend over the full intensity range, and thus, would appear dimmer than it should. To correct for this, you can use the rescale_intensity function to rescale the image so that it uses the full dtype range:
>>> from skimage import exposure
>>> image = exposure.rescale_intensity(img10bit, in_range=(0, 2**10 - 1))
Here, the in_range argument is set to the maximum range for a 10-bit image. By default, rescale_intensity stretches the values of in_range to match the range of the dtype.