基于视频的生猪行为跟踪和识别对于实现精细化养殖具有重要价值。为了应对群养生猪多目标跟踪任务中由猪只外观相似、遮挡交互等因素带来的挑战,研究提出了基于PigsTrack跟踪器的群养生猪多目标跟踪方法。PigsTrack跟踪器利用高性能YOLO...基于视频的生猪行为跟踪和识别对于实现精细化养殖具有重要价值。为了应对群养生猪多目标跟踪任务中由猪只外观相似、遮挡交互等因素带来的挑战,研究提出了基于PigsTrack跟踪器的群养生猪多目标跟踪方法。PigsTrack跟踪器利用高性能YOLOX网络降低目标误检与漏检率,采用Transformer模型获取具有良好区分特性的目标外观特征;基于OC-SORT(observation-centric sort)的思想,通过集成特征匹配、IoU匹配和遮挡恢复匹配策略实现群养生猪的准确跟踪。基于PBVD(pigs behaviours video dataset)数据集的试验结果表明,PigsTrack跟踪器的HOTA(higher order tracking accuracy),MOTA(multiple object tracking accuracy)和IDF1得分(identification F1 score)分别为85.66%、98.59%和99.57%,相较于现有算法的最高精度,分别提高了3.71、0.03和2.05个百分点,证明了PigsTrack跟踪器在解决外观相似和遮挡交互引起的跟踪过程中身份跳变问题方面的有效性。随后,利用Slowfast网络对PigsTrack跟踪器的跟踪结果进行了典型行为统计,结果显示PigsTrack在群养生猪个体行为统计方面更准确。此外,通过在ABVD(aggressive-behavior video)数据集上的试验,PigsTrack跟踪器的HOTA、MOTA和IDF1得分分别为69.14%、94.82%和90.11%,相对于现有算法的最高精度,提高了5.33、0.57和8.60个百分点,验证了PigsTrack跟踪器在群养生猪跟踪任务中的有效性。总而言之,PigsTrack跟踪器能够有效应对外观相似和遮挡交互等挑战,实现了准确的生猪多目标跟踪,并在行为统计方面展现出更高的准确性,为生猪养殖领域的研究和实际应用提供了有价值的指导。展开更多
Aims: Rotavirus-associated enteritis is a major problem in livestock, notably in young piglets and calves, and is also a zoonosis. It is also associated with diarrhoea mainly in children less than five years of age. I...Aims: Rotavirus-associated enteritis is a major problem in livestock, notably in young piglets and calves, and is also a zoonosis. It is also associated with diarrhoea mainly in children less than five years of age. In Tanzania however, no study has addressed Rotavirus in livestock species. Following our previous report on Rotavirus infection in children within urban and peri-urban Arusha, we sought to understand the disease situation in livestock in the same area. Study Design: Place and Duration of Study: In this study, we investigated the prevalence of Rotavirus in pigs of suckling, weaning and post weaning/grazing/fattening age categories in Lemara, Moshono and Sokoni I areas of Arusha peri-urban. Methodology: Molecular detection of Rotavirus in stool samples was done using conventional PCR with primers targeting Group A Rotavirus (GARV). Using a standardized questionnaire, we sought to find out risk factors associated with positive cases of Rotavirus including age, sex, location, diarrhoea status, recent diarrhoea case in the farm, breed, type of grazing system and type of feeding of individual pigs. Results: Out of a total of 110 pigs sampled (fecal samples), 41.8% were positive for Rotavirus. Chi Square’s (χ<sup>2</sup>) Fisher’s Exact Test was used to relate PCR test results with various possible risk factors. Recent diarrhoea case in the farm was significantly (p < 0.05) associated with Rotavirus infection in pigs indicating the possible role of cross-infection within farm and also the environmental resistance and persistence of the virus in the farm. Conclusions: This was the first study to report on Rotavirus infection in pigs in Tanzania. The information obtained should form the platform for further studies to address the molecular epidemiology and relatedness of Rotavirus from human and porcine positive cases.展开更多
群养圈栏内猪只的位置分布是反映其健康福利的重要指标。为解决传统人工观察方式存在的人力耗费大、观察时间长和主观性强等问题,实现群养猪只圈内位置的高效准确获取,该研究以三原色(Red Green Blue,RGB)图像为数据源,提出了改进的快...群养圈栏内猪只的位置分布是反映其健康福利的重要指标。为解决传统人工观察方式存在的人力耗费大、观察时间长和主观性强等问题,实现群养猪只圈内位置的高效准确获取,该研究以三原色(Red Green Blue,RGB)图像为数据源,提出了改进的快速区域卷积神经网络(Faster Region Convolutional Neural Networks,Faster R-CNN)的群养猪只圈内位置识别算法,将时间序列引入候选框区域算法,设计Faster R-CNN和轻量化CNN网络的混合体,将残差网络(Residual Network,ResNet)作为特征提取卷积层,引入PNPoly算法判断猪只在圈内的所处区域。对育成和育肥2个饲养阶段的3个猪圈进行24 h连续98 d的视频录像,从中随机提取图像25000张作为训练集、验证集和测试集,经测试该算法识别准确率可达96.7%,识别速度为每帧0.064s。通过该算法获得了不同猪圈和日龄的猪群位置分布热力图、分布比例和昼夜节律,猪圈饲养面积的增加可使猪群在实体地面的分布比例显著提高(P<0.05)。该方法可为猪只群体行为实时监测提供技术参考。展开更多
文摘基于视频的生猪行为跟踪和识别对于实现精细化养殖具有重要价值。为了应对群养生猪多目标跟踪任务中由猪只外观相似、遮挡交互等因素带来的挑战,研究提出了基于PigsTrack跟踪器的群养生猪多目标跟踪方法。PigsTrack跟踪器利用高性能YOLOX网络降低目标误检与漏检率,采用Transformer模型获取具有良好区分特性的目标外观特征;基于OC-SORT(observation-centric sort)的思想,通过集成特征匹配、IoU匹配和遮挡恢复匹配策略实现群养生猪的准确跟踪。基于PBVD(pigs behaviours video dataset)数据集的试验结果表明,PigsTrack跟踪器的HOTA(higher order tracking accuracy),MOTA(multiple object tracking accuracy)和IDF1得分(identification F1 score)分别为85.66%、98.59%和99.57%,相较于现有算法的最高精度,分别提高了3.71、0.03和2.05个百分点,证明了PigsTrack跟踪器在解决外观相似和遮挡交互引起的跟踪过程中身份跳变问题方面的有效性。随后,利用Slowfast网络对PigsTrack跟踪器的跟踪结果进行了典型行为统计,结果显示PigsTrack在群养生猪个体行为统计方面更准确。此外,通过在ABVD(aggressive-behavior video)数据集上的试验,PigsTrack跟踪器的HOTA、MOTA和IDF1得分分别为69.14%、94.82%和90.11%,相对于现有算法的最高精度,提高了5.33、0.57和8.60个百分点,验证了PigsTrack跟踪器在群养生猪跟踪任务中的有效性。总而言之,PigsTrack跟踪器能够有效应对外观相似和遮挡交互等挑战,实现了准确的生猪多目标跟踪,并在行为统计方面展现出更高的准确性,为生猪养殖领域的研究和实际应用提供了有价值的指导。
文摘Aims: Rotavirus-associated enteritis is a major problem in livestock, notably in young piglets and calves, and is also a zoonosis. It is also associated with diarrhoea mainly in children less than five years of age. In Tanzania however, no study has addressed Rotavirus in livestock species. Following our previous report on Rotavirus infection in children within urban and peri-urban Arusha, we sought to understand the disease situation in livestock in the same area. Study Design: Place and Duration of Study: In this study, we investigated the prevalence of Rotavirus in pigs of suckling, weaning and post weaning/grazing/fattening age categories in Lemara, Moshono and Sokoni I areas of Arusha peri-urban. Methodology: Molecular detection of Rotavirus in stool samples was done using conventional PCR with primers targeting Group A Rotavirus (GARV). Using a standardized questionnaire, we sought to find out risk factors associated with positive cases of Rotavirus including age, sex, location, diarrhoea status, recent diarrhoea case in the farm, breed, type of grazing system and type of feeding of individual pigs. Results: Out of a total of 110 pigs sampled (fecal samples), 41.8% were positive for Rotavirus. Chi Square’s (χ<sup>2</sup>) Fisher’s Exact Test was used to relate PCR test results with various possible risk factors. Recent diarrhoea case in the farm was significantly (p < 0.05) associated with Rotavirus infection in pigs indicating the possible role of cross-infection within farm and also the environmental resistance and persistence of the virus in the farm. Conclusions: This was the first study to report on Rotavirus infection in pigs in Tanzania. The information obtained should form the platform for further studies to address the molecular epidemiology and relatedness of Rotavirus from human and porcine positive cases.
