Hidden Markov model-based multi-modal image fusion with efficient training

Abstract

Automated spatial alignment of images from different modalities is an important problem, particularly in bio-medical image analysis. We propose a novel probabilistic framework, based on a variant of the 2D hidden Markov model (2D HMM), to capture the deformation between multi-modal images. Smoothness is ensured via transition probabilities of the 2D HMM and cross-modality similarity via class-conditional, modality-specific emission probabilities. The method is derived for general multi-modal settings, and its performance is demonstrated for an application in cellular microscopy. We also present an efficient algorithm for parameter estimation. Experiments on synthetic and real biological data show improvement over state-of-the-art multi-modal image fusion techniques.

2D hidden Markov model with spatially adaptive state-space for tracing many cells in image sequence

“…However, its direct extension to two-dimension is impractical because the complexity grows exponentially with the image size. There are several approximating variants of 2D-HMM algorithms [2], [16] and [17], of which the turbo 2D-HMM [2] is an effective and efficient algorithm that has been applied to a number of computer vision problems [18] [19] [20]. With a modified version of the forwardbackward algorithm, the turbo 2D-HMM iteratively decodes each row and column independently as a 1D-HMM, but allows them to "communicate" by inducing priors on each other.…”