gloh

arrayfire.gloh(image: Array, /, n_layers: int = 3, contrast_threshold: float = 0.04, edge_threshold: float = 10.0, initial_sigma: float = 1.6, dobule_input: bool = True, intensity_scale: float = 0.00392156862745098, feature_ratio: float = 0.05) → tuple[Features, Array]

Implements the GLOH (Gradient Location and Orientation Histogram) feature detection and descriptor extraction for images.

Parameters

imageArray

A 2D ArrayFire array representing the input image.

n_layersint, default: 3

The number of layers per octave. The number of octaves is calculated based on the image dimensions and initial_sigma.

contrast_thresholdfloat, default: 0.04

The contrast threshold used to filter out weak features in low-contrast regions of the image.

edge_thresholdfloat, default: 10.0

The edge threshold used to filter out edge-like features to ensure feature points are from corners.

initial_sigmafloat, default: 1.6

The initial sigma (scale) for the Gaussian blur applied to the image at the first layer.

dobule_inputbool, default: True

If True, the image size is doubled before processing to detect features at higher scales.

intensity_scalefloat, default: 1.0 / 255

The scale factor applied to the image intensities, typically used to normalize the pixel values.

feature_ratiofloat, default: 0.05

The ratio of the total number of pixels in the image used to limit the number of features detected.

Returns

tuple[Features, Array]

A tuple containing two elements: - Features: An object holding the detected features, including their locations and scales. - Array: An ArrayFire array containing the GLOH descriptors for the detected features, with each descriptor

having 272 elements.

Note

• The gloh function is particularly effective for object recognition and image matching tasks.

• The choice of parameters can significantly impact the number and quality of features detected and described.