Identifying industrial food foam structures by 2D surface image analysis and pattern recognition

Abstract

Bubbles are fundamental structural elements in several food products modulating density, rheology, texture, appearance and mouthfeel. Foams and aerated structures are characterized by their gas content, stability, bubble size and distribution. However, these measures alone cannot fully describe the complexity of bubble-containing structures. We have used three image analysis methods (Euler characteristic, Minkowski fractal and image texture) to characterize foam structure, and canonical and Bayesian discriminant analysis to identify/classify different foam architectures. This work describes results of this methodology on liquid foams stabilized by proteins at varying concentration and pH levels. Results indicated that groups of three structural parameters (among the 57 calculated) could successfully identify foam structures with different characteristics but unfortunately no single set of features could be used ubiquitously. Additional foam structure information as determined in this work can help to better understand these systems and the impact of bubbles on the physical properties of aerated foods. (C) 2012 Elsevier Ltd. All rights reserved.