A New Hierarchical Decomposition Applied to Object Segmentation in Image Sequences: The Uniform Decomposition

The uniform decomposition is a new hierarchical segmentation method which main property is the preservation of a given characteristic for each subset of the partition provided by the decomposition. Its computation is incremental and, in each iteration, a candidate region is split only if it is found a valid split where each new subset respects an input criterion, for instance size or shape. The uniform decomposition is comparable to the classical one given by the threshold of the extinction values from each regional minima. More, it is comparable while maintaining the criterion for each partition subset. The paper also proposes the application of the uniform decomposition as the core of the watershed from propagated markers, a method applied to assisted segmentation of objects in image sequences. An quantitative evaluation benchmark was applied to assess the combination of the uniform decomposition to the watershed from propagated markers. The evaluation highlights the achievements of good segmentation and efficiency.