Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Interferences for Detection of Poultry Behaviors with Machine-vision Based Methods

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  2020 ASABE Annual International Virtual Meeting  2000189.(doi:10.13031/aim.202000189)
Authors:   Yangyang Guo, Lilong Chai, Dongjian He
Keywords:   Animal behaviors; poultry production; machine vision; image processing; facilities

Abstract. Animal floor distribution (e.g., uniformity and time spend in feeding, drinking, and resting zones) reflects the health and welfare status of the production management. Daily routine inspection of poultry flock distributions (e.g., cage-free hens, broiler breeder, and broiler grow-out houses) is done manually in commercial grow-out houses, which is labor intensive and time consuming. This task requires an efficient system that can monitor bird‘s floor distributions automatically. Non-contact and nondestructive monitoring methods such as the machine vision-based technology has been widely tested in monitoring behaviors of livestock and poultry. However, it is technical challenging to apply vision-based method for commercial poultry. There are primary three challenges in monitoring group or individual animals in poultry houses: (1) animal population: a commercial cage free henhouse has about 50,000 laying hens and a commercial broiler grow-out 10,000 for broiler grow-out house in a standard production facility (demission is about 150 L * 30 W m); (2) facility interferences: in cage-free henhouse, aviary system such cages and perches are blocking imaging collection, only a part of birds on litter floor can be monitored. In boiler grow-out house and breeder house, the primary interferences caused by equipment/facilities include feeder hanging chains and water pipes; and (3) indoor environmental conditions such as lighting and dust concentration. To minimize the effect of challenges caused by equipment and facilities on utilization of machine vision-based methods in monitoring animal behaviors, this study discussed potential strategies for improving image quality as affected by equipment and facilities.

(Download PDF)    (Export to EndNotes)