Recipient Organization
INDUSTRY VISION AUTOMATION CORPORATION
14227 REED FARM WAY
GAITHERSBURG,MD 20878
Performing Department
(N/A)
Non Technical Summary
As Americans become increasingly health conscious, they have increased their consumption of white boneless poultry. The poultry industry is especially concerned about hazardous materials such as metal and bone fragments embedded in boneless meat. Bone fragments and physical contaminants must be removed to ensure consumer safety. This project will lead to an industry-ready system technology for sensitive and accurate detection of bone fragments and hazardous materials on poultry de-boning lines. With the U.S. producing over 46.9 billion pounds of poultry annually and boneless meat accounting for over 41% of the total product (USDA, 2004), the success of this project will have a positive impact on the industry in bringing quality, safe, and cost-effective poultry products to American consumers.
Animal Health Component
100%
Research Effort Categories
Basic
(N/A)
Applied
100%
Developmental
(N/A)
Goals / Objectives
The overall objective of this proposed project is to fill the gap between the basic research and the practical R&D for an industrially viable and automated technological system. The important tasks of this industrial R&D will include 1) engineering research and design for a robust and viable system in the harsh environment of poultry processing, 2) development of x-rays, laser optics, and electronic systems capable of meeting the essential throughput of chicken deboning processing lines, and 3) automated mechanisms and control schemes for tracking and rejecting bone fragments on conveyor belts. Moreover, it is essential to develop and experiment with the prototype system in the poultry plant environment to prove all aspects of the challenges for an industrially acceptable system.
Project Methods
Under previous support by the USDA/NRI and the U.S. Poultry & Egg Association, we have developed a new method that uses the synergism of x-ray and laser imaging (sensors employing different principles) for sensitive detection of bone fragments in de-boned poultry meat. This method overcomes the weakness of conventional x-ray technology and eliminates false image patterns by x-ray absorption compensation. It opens the way to an advanced detection technology that will help the processors achieve the goal of "zero tolerance" to physical hazards in food. In order to develop this technology, however, much R&D effort must be made to resolve outstanding practical issues in poultry processing engineering before a prototype can be realized for an innovative system technology. To meet the objectives listed above, we will do the research in following aspects: 1) Poultry processing plants are known to be harsh to electronics where water is ubiquitous. The relative humidity is
typically about 100%. Furthermore, water jets and streams are periodically used to totally wash down any processing equipment to prevent microbial proliferations according to USDA regulations. Consequently, these result in sudden changes in temperature, causing condensation on optics and electronics, which will potentially block the view of optics and damage the electronics. Thus, careful study must be conducted to ensure that the vision electronics can survive and perform in such a challenging environment. 2) To meet industrial requirements of minimum throughput of 7,500 lb/hr, a multi-lane conveyor has to be used. Consequently, several optical sensors have to be included to maintain the system resolution. To make a seamless fusion of laser 3D and X-ray images, the X-rays and laser imaging views are overlapped. The image deformation caused by view expansion will be corrected for image registration. 3) As an integrated part of the detection system, the bone inclusion rejection
mechanism must follow the pattern recognition decisions. It is will be linked to the detection algorithm and parallel image processing; and further, comprehensive research on tracking, synchronizing, automated controls associated with the detection, and the communication of the entire system with a computer will be investigated.