Poultry Audio Monitoring Research Highlights Unique Georgia Tech-UGA Collaboration
The theme of a recent Georgia Research Alliance Annual Report was “great minds think…together.” In fact, it can be said that some of the greatest research and development success stories involve collaborative efforts. In this case, Georgia Tech Research Institute (GTRI) engineers are collaborating with colleagues in the Georgia Tech School of Electrical and Computer Engineering (ECE) and the University of Georgia’s (UGA) Department of Poultry Science to address a unique poultry industry concern through cross-disciplinary research.
Utilizing GTRI’s expertise in control systems development and image processing, ECE’s expertise in audio signal processing, and UGA’s expertise in animal health, production, and welfare, an experimental monitoring system is being developed to help the poultry industry assess grow-out house conditions and thus flock health and performance by analyzing bird behaviors and vocalizations. The team is building upon past research conducted by Dr. Michael Darre and others at the University of Connecticut’s Department of Animal Science that showed correlations between bird vocalizations and their levels of stress.
“We know experienced poultry farmers can often tell whether a flock is content or distressed by listening to their birds,” explains Dr. Mike Lacy, head of UGA’s Department of Poultry Science. “If what these experienced farmers hear and sense can be defined and quantified, sensors to detect cues from the birds themselves could really make a difference in providing real-time information on house environment, bird health, and comfort.”
“The goal of this collaborative research project is to monitor various video and audio characteristics of the birds to determine the correlation between a flock’s environmental or medical conditions and their audio and video times series data and the specific features that could be used for characterization,” adds Dr. Wayne Daley, associate division chief of GTRI’s Food Processing Technology Division and project director.
The team includes Dr. David Anderson, Georgia Tech engineering professor, whose work in audio signal processing focuses on enhancing the functionality of hearing aids. “Our lab has been working on speech processing, signal enhancement, and machine learning for signal processing. The poultry welfare project has been a great application for this research,” says Anderson. “The signals that we collect are a relatively new type of signal for signal processing; few researchers have been interested in what chickens have to say.”
The audio signals tend to be quite noisy due to the cooling fans within the grow-out house, explains Anderson. When the team focused on extracting useful information from the signals, they discovered the speech processing techniques did not work well, but the signal enhancement and machine learning aspects of their research were very useful.
“The behavior of chickens is one of the best and most immediate indicators of their well-being. Chickens are vocal creatures and produce different types of vocalizations at different rates and loudness depending on their circumstances,” explains Dr. Bruce Webster, UGA poultry science professor.
According to Webster, it is possible that monitoring of the auditory environment might give real-time ability to gauge the presence of different kinds of stressors that might impact the welfare of the flock and provide the ability to make rapid environmental adjustments by automated control systems to alleviate the stress.
The experimental system was installed in a research grow-out house on the UGA campus and recently underwent six weeks of testing where the team collected data under normal and stressed (temperature increased 10 degrees above normal) growing conditions.
An analysis of the data showed that it is possible to detect a change in the vocalizations of the birds due to a change in temperature. A filtering and extraction technique was developed to isolate the sounds being made by the birds (called vocalizations) from the background noises in the room (particularly the fans). Results showed that the number of vocalizations rose and fell commensurate with the change in temperature. An analysis technique using the time domain showed a similar result. Results from the video data, however, are not as conclusive. Air quality, as measured by ammonia levels in the poultry house, is an additional stressor that was included in this study, the results of which are currently being analyzed.
If successful, Daley believes the audio monitoring could be used to take preemptive or corrective actions to maintain the health and viability of the flock, which in turn could improve production efficiency and bird welfare.
Likewise, Dr. Casey Ritz, UGA poultry science associate professor, says detecting conditions conducive to increased levels of stress in the birds and alleviating the condition will improve the productivity and economic well-being of individual flocks. “Contract poultry producers are paid by the pound of birds sent to market. Improving the overall health and productivity of the birds will help to improve the bottom line for individual producers.”
Lacy says the grow-out monitoring project collaboration between Georgia Tech and UGA is a great example of where brainstorming between scientists and engineers resulted in a project that could have great potential for monitoring the health and well-being of poultry flocks.
“If one thinks about how U.S. farmers will continue their leadership in global poultry production, one has to conclude that application of technology will be one of the key factors that will allow our farmers to compete,” adds Lacy. “In fact, research, knowledge, and application of technology may be one of the few competitive advantages our farmers will have in the future.