RGBShift

Targets:

image

volume

Image Types:uint8, float32

Randomly shift values for each channel of the input RGB image.

A specialized version of AdditiveNoise that applies constant uniform shifts to RGB channels. Each channel (R,G,B) can have its own shift range specified.

Arguments

r_shift_limit

tuple[float, float] | float

[-20,20]

Range for shifting the red channel. Options:

If tuple (min, max): Sample shift value from this range
If int: Sample shift value from (-r_shift_limit, r_shift_limit)
For uint8 images: Values represent absolute shifts in [0, 255]
For float images: Values represent relative shifts in [0, 1] Default: (-20, 20)

g_shift_limit

tuple[float, float] | float

[-20,20]

Range for shifting the green channel. Options:

If tuple (min, max): Sample shift value from this range
If int: Sample shift value from (-g_shift_limit, g_shift_limit)
For uint8 images: Values represent absolute shifts in [0, 255]
For float images: Values represent relative shifts in [0, 1] Default: (-20, 20)

b_shift_limit

tuple[float, float] | float

[-20,20]

Range for shifting the blue channel. Options:

If tuple (min, max): Sample shift value from this range
If int: Sample shift value from (-b_shift_limit, b_shift_limit)
For uint8 images: Values represent absolute shifts in [0, 255]
For float images: Values represent relative shifts in [0, 1] Default: (-20, 20)

p

float

0.5

Probability of applying the transform. Default: 0.5.

Examples

>>> import numpy as np
>>> import albumentations as A

# Shift RGB channels of uint8 image
>>> transform = A.RGBShift(
...     r_shift_limit=30,  # Will sample red shift from [-30, 30]
...     g_shift_limit=(-20, 20),  # Will sample green shift from [-20, 20]
...     b_shift_limit=(-10, 10),  # Will sample blue shift from [-10, 10]
...     p=1.0
... )
>>> image = np.random.randint(0, 256, (100, 100, 3), dtype=np.uint8)
>>> shifted = transform(image=image)["image"]

# Same effect using AdditiveNoise
>>> transform = A.AdditiveNoise(
...     noise_type="uniform",
...     spatial_mode="constant",  # One value per channel
...     noise_params={
...         "ranges": [(-30/255, 30/255), (-20/255, 20/255), (-10/255, 10/255)]
...     },
...     p=1.0
... )

Notes

Values are shifted independently for each channel
For uint8 images:
- Input ranges like (-20, 20) represent pixel value shifts
- A shift of 20 means adding 20 to that channel
- Final values are clipped to [0, 255]
For float32 images:
- Input ranges like (-0.1, 0.1) represent relative shifts
- A shift of 0.1 means adding 0.1 to that channel
- Final values are clipped to [0, 1]