Deep Learning and Machine Vision Approaches for Posture Detection of Individual Pigs

Abstract

Posture detection targeted towards providing assessments for the monitoring of health and welfare of pigs has been of great interest to researchers from different disciplines. Existing studies applying machine vision techniques are mostly based on methods using three-dimensional imaging systems, or two-dimensional systems with the limitation of monitoring under controlled conditions. Thus, the main goal of this study was to determine whether a two-dimensional imaging system, along with deep learning approaches, could be utilized to detect the standing and lying (belly and side) postures of pigs under commercial farm conditions. Three deep learning-based detector methods, including faster regions with convolutional neural network features (Faster R-CNN), single shot multibox detector (SSD) and region-based fully convolutional network (R-FCN), combined with Inception V2, Residual Network (ResNet) and Inception ResNet V2 feature extractions of RGB images were proposed. Data from different commercial farms were used for training and validation of the proposed models. The experimental results demonstrated that the R-FCN ResNet101 method was able to detect lying and standing postures with higher average precision (AP) of 0.93, 0.95 and 0.92 for standing, lying on side and lying on belly postures, respectively and mean average precision (mAP) of more than 0.93.

Automated Video Behavior Recognition of Pigs Using Two-Stream Convolutional Networks

“…Zheng et al [2] and Yang et al [22] used Faster-RCNN to recognize pig postures and feeding behaviors. Nasirahmadi et al [16] proposed three detector methods including Faster R-CNN, SSD, and R-FCN to recognize postures of pigs. Real-time sow drinking, urination, and mounting behavior recognition has been achieved by using an optimized target detection method based on the SSD and the MobileNet [24].…”

“…Staff also need to record sensor readings frequently, which is more troublesome. So, contactless, low-cost, easy, and effective computer vision techniques [1] have been widely used in animal monitoring processes and play an essential role in assessment of animal behavior [16]. Viazzi et al [17] extracted the mean intensity of motion and the occupation index; then, they used the Linear Discriminant Analysis (LDA) to classify two features to identify aggressive behavior with an accuracy of 89%.…”

“…Sows were segmented from all frames through the Fully Convolutional Network (FCN) model, which could help recognize sows' nursing behavior with an accuracy of 97.6% [23]. Nasirahmadi et al [16] proposed three detector methods including Faster R-CNN, single shot multibox detector (SSD), and region-based fully convolutional network (R-FCN) to recognize the standing, lying on side, and lying on belly postures of pigs with a mean average precision (mAP) of over 0.93. Real-time sow drinking, urination, and mounting behavior recognition has been achieved by using an optimized target detection method based on the SSD and the MobileNet [24].…”

Calibration of In-Plane Center Alignment Errors in the Installation of a Circular Slide with Machine-Vision Sensor and a Reflective Marker

“…Prior to calibrating the IPCA using the previously selected machine-vision camera, as shown in Figure 10 , camera calibration should be performed. Here, error factors are identified and calibrated, such as the distortion of the lens (including the radial distortion and tangential distortion that arise when a point in a three-dimensional space is mapped onto a two-dimensional image plane) and installation uncertainties [ 25 , 26 , 27 ].…”

A Spatiotemporal Convolutional Network for Multi-Behavior Recognition of Pigs

“…Lying, feeding and drinking behaviors were detected by posture classification in deep images [7]. Faster regions-convolutional neural network (Faster R-CNN), single shot multibox detector (SSD) and region-based fully convolutional network (R-FCN) were applied for classification of lying and other kinds of standing behaviors [8]. Moreover, the behaviors could also be classified by spatiotemporal features.…”