Digital image analysis for the optimal control of bread and rolls baking

verfasst von: D. Solle, B. Hitzmann, M. Schirmer, T. Becker
Abstract: In many bakeries the rolls are baked in place by the salesclerk. However, the baking goods cannot be monitored all the time, because the selling consumes most of the time of the salesclerk. The current state of the art is a purely timely based process control, which does not consider the actual quality of the raw material as well as the process parameter like the humidity, temperature and time. The baking process is absolutely irreversible. If an error occurs, it will show up in the final product. Without a process monitoring just suboptimal results can be obtained. Therefore, an automated supervisory system is necessary. It must be able to identify the current state of the baking goods and to decide which action has to be carried out. Based on parameters like the baking time, the temperature as well as the pastry color the state of the baking process can be identified and necessary actions can be performed. In a joined project we are developing an intelligent baking-oven to optimize industrial baking using digital process imaging and experience-based Fuzzy control. The main goals of the project are the development of an optical system, which is able to give digital images in a continuous form and high quality of the baking goods. The development of algorithms for the characterization of color and shape of the images retrieved from inside the oven, in such a way that it is possible to evaluate the current status of the process. An optical on-line volume measuring instrument will be developed and its estimation concerning accuracy, reliability and product independence. The optical measured values will be correlated with the pastry condition by the use of classical dough and bread analyzing methods. A fuzzy based expert system and a control system will be developed, which will fully automated control the baking process. In our contribution the structure and first results of the intelligent baking-oven will be presented. Using examples of images of baking goods the results of the analysis system will be discussed.
Organisationseinheit(en): Institut für Technische Chemie
Externe Organisation(en): Technische Universität München (TUM)
Typ: Paper
Seiten: 666-671
Anzahl der Seiten: 6
Publikationsdatum: 2009
Publikationsstatus: Veröffentlicht
Peer-reviewed: Ja
ASJC Scopus Sachgebiete: Lebensmittelwissenschaften

BibTeX

@conference{fd9d96165c954977a6e10663886e181e,
title = "Digital image analysis for the optimal control of bread and rolls baking",
abstract = "In many bakeries the rolls are baked in place by the salesclerk. However, the baking goods cannot be monitored all the time, because the selling consumes most of the time of the salesclerk. The current state of the art is a purely timely based process control, which does not consider the actual quality of the raw material as well as the process parameter like the humidity, temperature and time. The baking process is absolutely irreversible. If an error occurs, it will show up in the final product. Without a process monitoring just suboptimal results can be obtained. Therefore, an automated supervisory system is necessary. It must be able to identify the current state of the baking goods and to decide which action has to be carried out. Based on parameters like the baking time, the temperature as well as the pastry color the state of the baking process can be identified and necessary actions can be performed. In a joined project we are developing an intelligent baking-oven to optimize industrial baking using digital process imaging and experience-based Fuzzy control. The main goals of the project are the development of an optical system, which is able to give digital images in a continuous form and high quality of the baking goods. The development of algorithms for the characterization of color and shape of the images retrieved from inside the oven, in such a way that it is possible to evaluate the current status of the process. An optical on-line volume measuring instrument will be developed and its estimation concerning accuracy, reliability and product independence. The optical measured values will be correlated with the pastry condition by the use of classical dough and bread analyzing methods. A fuzzy based expert system and a control system will be developed, which will fully automated control the baking process. In our contribution the structure and first results of the intelligent baking-oven will be presented. Using examples of images of baking goods the results of the analysis system will be discussed.",
keywords = "Baking control, Image processing",
author = "D. Solle and B. Hitzmann and M. Schirmer and T. Becker",
note = "Copyright: Copyright 2010 Elsevier B.V., All rights reserved.; 5th International Technical Symposium on Food Processing, Monitoring Technology in Bioprocesses and Food Quality Management ; Conference date: 31-08-2009 Through 02-09-2009",
year = "2009",
language = "English",
pages = "666--671",
}