Use of computer vision systems in baked products: potential tool for measuring physical properties

In response to the growing market demand for healthier bakery products, recent studies in the baking industry have focused on evaluating different types of bread using standardized methods and specialized instruments such as alveographs, colorimeters, and volume meters, as well as sensory tests. However, these methods require resources and specific equipment, which limits their replicability. This review addresses alternative, noninvasive tools and methodologies for measuring physical properties of bakery products through its visual appearance. An initial literature review identified four main categories of properties studied in bakery products: internal, external, nutritional, and sensory. A second structure review was made with a Boolean equation to explore the use of imaging technologies, including computer vision systems, to measure some of these properties. These systems are characterized by fast and accurate tools for inspecting, measuring, and classifying food. The structured review revealed a growing trend in the use of image-based analysis at the macro and microscopic levels to evaluate the physical properties of bread. However, the wider use of computer vision systems is limited by the lack of standardized hardware for image processing, which makes replication in different environments difficult. Although there are some advanced programs that can preprocess, segment, and extract data from images, their effectiveness often depends on user expertise. This review highlights the potential of digital image processing as a non-invasive tool in the baking industry and the challenges that must be addressed to enable wider implementation.