Detecting edges from digital images at multiple scales using

What is it about?

Regularization methods are used widely in image selective smoothing and edge preserving restoration of noisy images. Traditional methods utilize image gradients within regularization function for controlling the smoothing and can produce artifacts when noise levels are higher. In this paper, we consider a robust image adaptive exponent driven regularization for filtering noisy images with salient feature preservation. We also consider a GPU-based implementation that computes the edge map in real time at 45-60 frames/s depending on the GPU card.

Featured Image

Photo by Martin Sanchez on Unsplash

Photo by Martin Sanchez on Unsplash

Why is it important?

Identifying reliable edge maps is important in image processing and computer vision. By considering a spatially adaptive variable exponent function that depends on a continuous switch based on the eigenvalues of structure tensor which can identify noisy edges, and corners with higher accuracy. Multiscale structure tensor-based spatially adaptive variable exponent provides improved feature extraction image regularization properties.