Automated Vision-Based Inspection and Control of Baking Systems
Research Focus
Georgia Tech research engineers in partnership with BakeTech, a Tucker, Georgia-based, baking equipment manufacturer, and Flowers Baking Company
are working on a system to continuously monitor and control product quality on a commercial bread/bun line. Machine vision-based approaches are
being explored to extract information about bread quality. While this will be used for the rejection/removal of out-of-specification product,
the ultimate goal is to investigate intelligent control strategies for correcting earlier processes that contribute to defective product. Of
particular interest is the potential to provide some control and optimization feedback to the baking/oven and proofing stages.
Background and Challenges
The bakery industry is the second largest segment of the food processing industry in the United States. One of its growing market segments is
the production of buns and rolls for food-service and fast-food customers. Many of these customers are increasing the demands on bakery quality
control operations that screen for bun size, shape, color, and garnish coverage.
Currently, the standard inspection process for baked products is for workers to manually remove samples from the line and compare them with
customer and industry specifications. If the results indicate a problem, steps are taken to adjust the operation to eliminate future quality
problems. At-line inspection systems have recently begun to emerge that allow these quality checks to be performed immediately next to the line.
However, it is clear that time-consuming manual inspection methods are becoming insufficient to meet the customers’ expectations for product
consistency and uniformity.
Large-scale baking oven systems are subject to natural variability and disturbances that can directly impact the quality of the product being
made. Closed-loop control methods can be introduced using automatic vision-based inspection sensors to compensate for fluctuations of this
nature. The particular challenge is recognizing the contributing factors that lead to the occurrence of defective buns. This requires an
understanding of the baking process and specifically, in this case, of the oven dynamics.
Project Overview
Begun in 2001, this project builds on a plethora of knowledge accrued over many years of research related to imaging and automation in the food
and agricultural industries. The Georgia Tech team is seeking to continue this by expanding the knowledge base associated with the inspection and
grading of bread. The concept is to develop a continuous, on-line, 100% coverage, real-time quality analysis system with particular emphasis not
only on automated rejection but corrective control as well.
The development will take place in two phases. The first phase will focus on the development of the imaging system for use in the inspection
aspect of the problem. This will include developing a broader understanding of the baking process and some conceptualization of applicable
intelligent control strategies. The second phase will shift toward the field testing of an imaging prototype system with an emphasis on
developing the supervisory control schemes for the oven and proofer.
One key to the success of the project will be the assistance of the industry partners on the project. The research team is working closely
with baking industry members from across the board. Their input and direct involvement in the development, design, and construction of the
project will ensure that the system meets the needs of the industry.
Future Directions
This project presents a unique set of challenges that have not as of yet been thoroughly examined. The concept of performing real-time
supervisory control of the baking process using vision-based inspection is of particular interest. It is expected that this work will result in
a commercially viable system that can achieve the product throughput rates used in the baking industry while meeting the stringent quality and
consistency specifications required by the customers.
Significant benefits to the baking industry are anticipated in terms of reduced processing costs, improved inspection and control
performance, and increased throughput. Generic aspects of the proposed technologies may be applicable to other food processing industries,
thus expanding significantly the potential benefits to the food processing community.
Project Co-Directors:
Doug Britton
Food Processing Technology Division
Georgia Tech Research Institute
Atlanta, GA 30332-0823
(404) 385-0418, phone
(404) 894-8051, fax
doug.britton@gtri.gatech.edu
Bonnie Heck
School of Electrical & Computer Engineering
Georgia Institute of Technology
Atlanta, GA 30332-0250
(404) 894-3145, phone
(404) 894-4641, fax
bonnie.heck@ece.gatech.edu |
