Abstract:This paper presents a computational paradigm called Data-Driven Markov Chain Monte Carlo (DDMCMC) for image segmentation in the Bayesian statistical framework. The paper contributes to image segmentation in four aspects. First, it designs efficient and well-balanced Markov Chain dynamics to explore the complex solution space and, thus, achieves a nearly global optimal solution independent of initial segmentations. Second, it presents a mathematical principle and a K-adventurers algorithm for computing multiple distinct solutions from the Markov chain sequence and, thus, it incorporates intrinsic ambiguities in image segmentation. Third, it utilizes data-driven (bottom-up) techniques, such as clustering and edge detection, to compute importance proposal probabilities, which drive the Markov chain dynamics and achieve tremendous speedup in comparison to the traditional jump-diffusion methods. Fourth, the DDMCMC paradigm provides a unifying framework in which the role of many existing segmentation algorithms, such as, edge detection, clustering, region growing, split-merge, snake/balloon, and region competition, are revealed as either realizing Markov chain dynamics or computing importance proposal probabilities. Thus, the DDMCMC paradigm combines and generalizes these segmentation methods in a principled way. The DDMCMC paradigm adopts seven parametric and nonparametric image models for intensity and color at various regions. We test the DDMCMC paradigm extensively on both color and gray-level images and some results are reported in this paper.

Region competition: unifying snakes, region growing, energy/Bayes/MDL for multi-band image segmentation

Region Competition: Unifying Snakes, Region Growing, and Bayes/MDL for Multiband Image Segmentation

Mpetition: Unifying Snakes, Egion Growin and Bayes/MDL for Image Segmentation

Region Competition and its Analysis: A Unified Theory for Image Segmentation

Salient Region Detection Algorithm Via KL Divergence and Multi-scale Merging

Decompose Image into Meaningful Regions Based on Contour Detector and Watershed Algorithm

Interactive Color Image Segmentation by Region Growing Combined with Image Enhancement Based on Bezier Model

A fast region-based image segmentation based on least square method

Image segmentation by MAP-ML estimations.

Region Labeling Algorithm Based on Boundary Tracking for Binary Image

Automatic Image Segmentation by Dynamic Region Merging

Neighborhood Aided Implicit Active Contours

Parsing Images into Regions, Curves, and Curve Groups

Variational Model for Image Segmentation.

Image segmentation by data-driven markov chain monte carlo

Entropy Regularization, Automatic Model Selection, And Unsupervised Image Segmentation

Seeded Region Merging Based on Gradient Vector Flow for Image Segmentation.

Multi-view EM Algorithm and Its Application to Color Image Segmentation.

Lifting-based variational multiclass segmentation algorithm: design, convergence analysis, and implementation with applications in medical imaging

A Fast Algorithm for Image Segmentation Based on Fuzzy Region Competition

AIMS: All-Inclusive Multi-Level Segmentation