Probabilistic Modeling for Structural Change Inference
نویسندگان
چکیده
We view the task of change detection as a problem of object recognition from learning. The object is defined in a 3D space where the time is the 3 dimension. We propose two competitive probabilistic models. The first one has a traditional regard on change, characterized as a ’presenceabsence’ within two scenes. The model is based on a logistic function, embedded in a framework called ’cut-andmerge’. The second approach is inspired from the Discriminative Random Fields (DRF) approach proposed by Ma and Hebert [KUMA2003]. The energy function is defined as the sum of an association potential and an interaction potential. We formulate the latter as a 3D anisotropic term. A simplified implementation enables to achieve fast computation in the 2D image space. In conclusion, the main contributions of this paper rely on : 1) the extension of the DRF to a 3D manifold ; 2) the cut-and-merge algorithm. The application proposed in the paper is on remote sensing images, for building change detection. Results on synthetic and real scenes and comparative analysis demonstrate the effectiveness of the proposed approach.
منابع مشابه
E cient Probabilistic Inference with Partial Ranking Queries
Distributions over rankings are used to model data in various settings such as preference analysis and political elections. The factorial size of the space of rankings, however, typically forces one to make structural assumptions, such as smoothness, sparsity, or probabilistic independence about these underlying distributions. We approach the modeling problem from the computational principle th...
متن کاملThe Dynamics of Probabilistic Structural Relevance
Probabilistic inference with a belief network in general is computationally expensive. Since the concept of structural relevance provides for identifying parts of a belief network that are irrelevant to a context of interest, it allows for alleviating to some extent the computational burden of inference: inference can be restricted to the network's relevant part. The structurally relevant part ...
متن کاملRule-based joint fuzzy and probabilistic networks
One of the important challenges in Graphical models is the problem of dealing with the uncertainties in the problem. Among graphical networks, fuzzy cognitive map is only capable of modeling fuzzy uncertainty and the Bayesian network is only capable of modeling probabilistic uncertainty. In many real issues, we are faced with both fuzzy and probabilistic uncertainties. In these cases, the propo...
متن کاملAn Introduction to Inference and Learning in Bayesian Networks
Bayesian networks (BNs) are modern tools for modeling phenomena in dynamic and static systems and are used in different subjects such as disease diagnosis, weather forecasting, decision making and clustering. A BN is a graphical-probabilistic model which represents causal relations among random variables and consists of a directed acyclic graph and a set of conditional probabilities. Structure...
متن کاملComposable probabilistic inference with BLAISE
If we are to understand human-level cognition, we must understand how the mind finds the patterns that underlie the incomplete, noisy, and ambiguous data from our senses and that allow us to generalize our experiences to new situations. A wide variety of commercial applications face similar issues: industries from health services to business intelligence to oil field exploration critically depe...
متن کاملProperties of Spatial Cox Process Models
Probabilistic properties of Cox processes of relevance for statistical modeling and inference are studied. Particularly, we study the most important classes of Cox processes, including log Gaussian Cox processes, shot noise Cox processes, and permanent Cox processes. We consider moment properties and point process operations such as thinning, displacements, and superpositioning. We also discuss...
متن کامل