When Statistical Physics Meets Image Processing

Puntos clave

• To summarize the contributions of Eugène Pechersky to image processing, particularly through Bayesian methods and model development.
• Reviewed Pechersky's work during his time at Telecom Paris and INRIA Sophia Antipolis.
• Analyzed multi-body interaction models designed to enhance traditional image processing techniques.
• Examined applications of the derived models in remote sensing and biomedical imaging.
• Developed models that extend the traditional Ising and Potts models for specific image processing tasks.
• Identified successful applications of these models in remote sensing and biomedical images, highlighting their effectiveness.
• Emphasized the role of Bayesian approaches in bridging statistical physics and image processing.