BlurCV2

class deeptrack.optical.math.BlurCV2(filter_function: Callable | str, mode: str | Callable[[...], str] = 'reflect', **kwargs: Any)

Bases: Feature

Apply a blurring filter using OpenCV (cv2).

Applies an OpenCV-based blurring or filtering operation to an input image. The provided filter_function must be compatible with OpenCV and accept the input image via the src argument (e.g., cv2.GaussianBlur, cv2.bilateralFilter).

Parameters

filter_functionCallable or str

OpenCV-compatible filtering function. If a string is provided, it is resolved as an attribute of cv2.

modestr, default=”reflect”

Border handling mode. Supported values are: {‘reflect’, ‘wrap’, ‘constant’, ‘mirror’, ‘nearest’}. These are internally mapped to OpenCV border types.

**kwargsAny

Additional keyword arguments passed directly to the filtering function.

Methods

get(image: np.ndarray, **kwargs: Any) –> array

Applies the blurring filter to the input image.

Notes

BlurCV2 is NumPy-only and does not support PyTorch tensors.

Examples

>>> import deeptrack as dt

Create an input image:

>>> import numpy as np
>>>
>>> input_image = np.random.rand(32, 32)

Define a blur feature using the Gaussian blur function:

>>> import cv2
>>>
>>> blur = dt.BlurCV2(
...     filter_function=cv2.GaussianBlur,
...     ksize=(5, 5),
...     sigmaX=1,
...     mode='reflect',
... )
>>> output_image = blur(input_image)
>>> print(output_image.shape)
(32, 32)

Methods Summary

get(image, mode, **kwargs)

Applies the blurring filter to the input image.

Methods Documentation

get(image: ndarray, mode: str, **kwargs: Any) → ndarray

Applies the blurring filter to the input image.

This method applies the blurring filter to the input image.

Parameters

image: np.ndarray

The input image to blur. Must be a NumPy array.

**kwargs: Any

Additional parameters for the blurring function.

Returns

np.ndarray

The blurred image.