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.

Background Image and Noise Analysis by a Thermal Imaging and Vocalization Monitoring System for Piglets Stress Study

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

Citation:  2016 ASABE Annual International Meeting  162460542.(doi:10.13031/aim.20162460542)
Authors:   Ching-Lu Hsieh, Jheng-Hong Hsieh
Keywords:   animal welfare, image processing, pigs, sensing, volatilization

Abstract. Swine is a very important animal husbandry in Taiwan. It produces more than NT 76 billion yearly (at 2014) and makes itself become the top one both at the rank of production in weight and yield in economics as well as in single item production of agriculture. During swine growth, piglet is a crucial stage that will significantly influence its market value. Meanwhile, animal welfare is also an important factor which eventually affects the efficiency of animal growth and meat quality. Therefore, non-invasive technology can be a valuable tool in animal husbandry. This study integrates a sound recorder and a thermal camera along with temperature and humidity logger to set up a noninvasive monitor system for piglet stress during pigpen change or weaning stage. The monitor system was set up in lab filed (a pigpen) after calibration. A 24-hr record showed background information in sound, thermal image, temperature, and humidity of the pigpen. A piecewise regression method was applied to calibrate the temperature with the pixel intensity for thermal image, which showed accuracy of 91.5% for image center and 98.5% for target point. With these regression equations, temperature of each pixel in thermal image can be calculated based on its color intensity of hue and saturation. The thermal image showed the pigpen had 8 C differences between noon and morning (7-8 AM) on the test day and also indicated that manger and fence of the cage contains higher temperature than floor. In sound record analysis by FFT transform showed a continuous low frequency (less than 200Hz) noise from adjacent motor factory and birds on trees. A randomly pass-by motorcycle generated high frequency (about 450 Hz) noise. Animal sound is a combination of many frequencies or is a combination of fundamental frequency with its overtones. With waveform, spectrogram, and power slice we can visualize the intensity change of sound with time and frequency. The findings on piglet stress can be used for farmers and government agents to improve animal management, and to upgrade animal welfare.

(Download PDF)    (Export to EndNotes)