Tail biting is costly to the pig industry through the loss of pig productivity, primary/ secondary infection, death loss and condemned carcasses.
Anyone who’s ever gone through a pig pen knows that they are curious creatures who use their mouths to explore their environment, as well as having a natural desire to chew. If animals show evidence of tail biting, this might cause complications. It can appear to start out of nowhere, affecting one or two animals or entire pig farm. Scientists are now working on a way to detect when a pig is going to chomp down on another pig using 3D cameras and machine-vision algorithms.
Pigs have a nasty habit of chewing each other’s tails. These bites can infect up to 30% of a pig farm’s pigs, rendering them unfit for human consumption. Pig tail docking, or cutting, can lessen but not eliminate biting, and docking is prohibited in the European Union. Tail biting epidemics can be triggered by a variety of factors, including heredity, nutrition, overcrowding, temperature fluctuations, improper ventilation and illumination, disease, and even the season, making it an unpredictable problem. “Tail biting is a very aggravating task,” says John Deen, a University of Minnesota veterinarian and epidemiologist. “Things hasn’t always been easy to keep it under control.”
Researchers in Scotland used time-of-flight and regular video cameras to track 667 undocked piglets on a farm for 52 days in order to forecast and maybe avoid tail biting. The pigs were checked for signs of biting at least twice a day.
Each time-of-flight camera generated infrared light pulses 25 times per second from LEDs and recorded the time it took to detect reflected pulses. Scientists were able to track each pig’s position and posture using this information. The farm-technology business Innovent Technology in Aberdeenshire, Scotland, used machine-vision algorithms to figure out which behaviours could be used as early warning indicators of tail biting.
Pigs’ tails were progressively pressed against their bodies before episodes of biting, according to the researchers. Furthermore, the software was 73.9 percent accurate in detecting when these changes in tail posture occurred. “It appears to be good technology, and I’m interested in seeing how it may be used on a farm,” says Deen, who was not involved in the study.
If farmers suspect a biting outbreak in a pigpen, they can use distractions like straw, knotted ropes, or shredded cardboard, which appeal to the pigs’ natural rooting and chewing inclinations.
“Another thing people attempt is applying bad-tasting material like Stockholm Tar on tails,” says Richard D’Eath, an animal behavior expert at Scotland’s Rural College in Edinburgh, who worked on the study. Farmers could save money by using such medicines only when they are really necessary, thanks to an early warning system.
The research was funded by the UK’s Agri-Tech Catalyst initiative, which has a £160 million budget to develop breakthrough farming technology. Agriculture and food currently generate over £108 billion in annual revenue and employ 3.9 million people. According to a recent industry-led study, introducing digital technology like robotics and autonomous systems into food processing may boost the UK economy by £58 billion over the next 13 years.
With up to £676,000 in financing from Innovate UK, a government agency, a three-year initiative dubbed TailTech is now pursuing the development of this early warning system. The goal is to test a prototype system on more than 16,000 pigs over the course of 18 months at nine farms across Europe, with each time-of-flight camera capable of monitoring up to 300 pigs, according to D’Eath.
TailTech will test the system’s effectiveness on a variety of pig farms, including those with docked and undocked pigs, those with and without straw on their flooring, and those with pigs of differing bloodlines, diets, and group sizes. Before scientists are certain that a tail-biting outbreak will emerge, the experiment will look at what percentage of pigs hold their tails low and for how long. It will also look at how predictors differ amongst farms. The researchers also want to increase the system’s accuracy. “However, in fact, 73.9 percent is sufficient for the system to function well,” D’Eath explains.
He goes on to say that the ultimate goal is to have an early warning system “that reads out continuously on a screen and also sends notifications to the farmer’s smartphone.” “Once the system is setup, no technical knowledge is required.” According to D’Eath, the software will compute trends to provide farmers a clearer picture of their animals’ present risk level.
Although a farmer may not purchase technology specifically designed to detect tail biting, D’Eath points out that this system is being developed as an add-on to Innovent’s Qscan camera-based automatic pig-weighing system, which helps farms reach contractual weight objectives. This method, according to Deen, a veterinarian and epidemiologist, could make all the difference. “If Qscan is already in place,” he argues, “I believe farmers can easily justify installing this system at a low cost.”
