IVCVApr 23, 2020

Edge Detection using Stationary Wavelet Transform, HMM, and EM algorithm

arXiv:2004.11296v11 citations
AI Analysis

This is an incremental improvement for image processing tasks, specifically enhancing edge detection in noisy conditions.

The paper tackled edge detection in noisy images by proposing a new technique using Stationary Wavelet Transform (SWT) with a Hidden Markov Model (WHMM) and the EM algorithm, achieving efficient application to noisy images.

Stationary Wavelet Transform (SWT) is an efficient tool for edge analysis. This paper a new edge detection technique using SWT based Hidden Markov Model (WHMM) along with the expectation-maximization (EM) algorithm is proposed. The SWT coefficients contain a hidden state and they indicate the SWT coefficient fits into an edge model or not. Laplacian and Gaussian model is used to check the information of the state is an edge or no edge. This model is trained by an EM algorithm and the Viterbi algorithm is employed to recover the state. This algorithm can be applied to noisy images efficiently.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes