ASABE Technical Library - Abstract
Member and Access Notice
Development of an Artificial Vision System for Real-Time Monitoring of Cow Breathing Rate
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org
Citation: 2009 Reno, Nevada, June 21 - June 24, 2009 096032.(doi:10.13031/2013.27037)Authors: Raphael Linker, Vladimir Kushnir
Keywords: Active contour, Heat stress, Image processing
This work reports a first step towards the development of an artificial vision system for real time estimation of cow breathing rate. A simple video camera placed at four meters above the barn floor was used to record 30-second sequences at a frame rate of 30 frames/second. Each sequence was decomposed into consecutive individual frames, which were analyzed to determine the width of the cow close to the regio abdominis lateralis. The contour of the "cow" object within each image was extracted using the active contour approach after rotating and translating the image to bring the cow to a standard location within the frame. After extracting the cow's contour, the width of the cow was calculated at a predetermined location. Finally, the breathing rate of the cow was estimated by performing fast Fourier transform (FFT) of this signal and identifying the FFT coefficient corresponding to the breathing rate. The results showed that in most cases this system yielded good estimate of the cow breathing rate. However, most of the cows used in this study were mostly white and much poorer results were observed with mostly black cows due to poor image contrast.
(Download PDF) (Export to EndNotes)