Behavior Comparison During Chronic Heat Stress in Large White and Creole Pigs Using Image-Analysis

Abstract

Behavior is a good indicator of animal welfare, especially in challenging environments. However, few studies have investigated how pig behavior changes during heat stress. The current study is a proof-of-concept using Convolutional Neural Network (CNN) models to monitor pig behavior in order to investigate the differences in behavioral response to heat stress of two contrasted breeds: Large White (LW), selected for high performance, and Creole (CR), adapted to tropical conditions. A total of 6 slaughter pigs (3 CR and 3 LW; 22 weeks of age) were monitored from 8:30 to 17:30 during 54 days. Two CNN architectures were used to detect the animal (Yolo v2) and to estimate animal's posture (GoogleNet). Pig postures estimated by the neural network showed that pigs spent more time lying on their side when temperature increased. When comparing the two breeds, as temperature increases, CR pigs spent more time lying on their side than LW pigs, suggesting that they use this posture to increase thermoregulation and dissipate heat more efficiently. This study demonstrates that neural network models are an efficient tool to monitor animal behavior in an automated way, which could be particularly relevant to characterize breed adaptation to challenging environments.

Analysis of image-based sow activity patterns reveals several associations with piglet survival and early growth

“…The use of postural and activity budgets is rarely described in the literature [however, see ( 60 – 63 )]. There are many studies of postural changes assessed by scan sampling and rare continuous analyses of behavior over more than 1 day.…”

Utilization of Infrared Thermography in Assessing Thermal Responses of Farm Animals under Heat Stress

“…In the case of pigs, Bonneau et al [75] reported an increase in the amount of time the animals were lying down on the side-to dissipate heat by contact with cool surfaces-by up to 5.9% when the temperature increased by 1 • C. The authors also determined that creole pigs (dark skin) spent more time lying down than large white pigs (lighter skin) (4%), probably due to the anatomical differences between breeds, which could help to identify heat-tolerant breeds for certain pig production units where ambient temperatures are high. In small-tail Han sheep raised indoors without climate control, Li et al [76] observed that ewe and ram lying behavior increased according to the THI increase (up to 45.10 ± 2.42%).…”

Pattern Mining-Based Pig Behavior Analysis for Health and Welfare Monitoring

“…Utilizing video sensors, PLF methods have been proven to be efficient, noninvasive, and stress-free, which allows continuous, real-time, and automated monitoring of animals to assist farmers in managing their livestock and improving their productivity. Considering these benefits, some studies [37][38][39][40][41][42][43][44] have leveraged PLF methods with video data to recognize pigs' behaviors relevant to their research. Table 1 summarizes some of the recent studies related to pig behavior recognition.…”

“…Some researchers, such as Nasirahmadi et al [37], used traditional machine learning techniques because they need less data to train the model and are less computationally demanding. Meanwhile, others [38][39][40][41][42][43][44] used deep learning techniques since they can automatically extract more relevant features (instead of relying on handcrafted features), which improves performance. Before performing behavior recognition, some of the aforementioned studies used predefined bounding boxes and cropped images to localize pigs for validating their approach in recognizing individual postures in commercial farms [37] and providing a low-cost solution to monitor the aggressive behavior [40] as well as feeding and drinking behaviors [44] of individual pigs.…”

“…Before performing behavior recognition, some of the aforementioned studies used predefined bounding boxes and cropped images to localize pigs for validating their approach in recognizing individual postures in commercial farms [37] and providing a low-cost solution to monitor the aggressive behavior [40] as well as feeding and drinking behaviors [44] of individual pigs. By contrast, other studies used detection techniques to automate the process of localizing individual pigs for monitoring their drinking, urination, and mounting behaviors in real time [38], monitoring and assessing their postures in response to various heat conditions [41], and improving the performance of individual posture recognition [42,43]. In addition to using a detection method to automatically localize pigs, Alameer et al [39] used tracking to identify individually detected pigs using unique identifiers (IDs) for differentiating the pigs and keeping a record of the behaviors displayed by each pig along with their IDs.…”