Artificial intelligence and classic methods to segment and characterize spherical objects in micrographs of industrial emulsions.

The stability of emulsions is a critical concern across multiple industries, including food products, agricultural formulations, petroleum, and pharmaceuticals. Achieving prolonged emulsion stability is challenging and depends on various factors, with particular emphasis on droplet size, shape, and spatial distribution. Addressing this issue necessitates an effective investigation of these parameters and finding solutions to enhance emulsion stability. Image analysis offers a powerful tool for researchers to explore these characteristics and advance our understanding of emulsion instability in different industries. In this review, we highlight the potential of state-of-the-art deep learning-based approaches in computer vision and image analysis to extract relevant features from emulsion micrographs. A comprehensive summary of classic and cutting-edge techniques employed for characterizing spherical objects, including droplets and bubbles observed in micrographs of industrial emulsions, has been provided. This review reveals significant deficiencies in the existing literature regarding the investigation of highly concentrated emulsions. Despite the practical importance of these systems, limited research has been conducted to understand their unique characteristics and stability challenges. It has also been identified that there is a scarcity of publications in multimodal analysis and a lack of a complete automated in-line emulsion characterization system. This review critically evaluates the existing challenges and presents prospective directions for future advancements in the field, aiming to address the current gaps and contribute to the scientific progression in this area.
Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
(Copyright © 2023 The Author(s). Published by Elsevier B.V. All rights reserved.)