Developing Edge AI Computer Vision for Smart Poultry Farms Using Deep Learning and HPC.

This research describes the use of high-performance computing (HPC) and deep learning to create prediction models that could be deployed on edge AI devices equipped with camera and installed in poultry farms. The main idea is to leverage an existing IoT farming platform and use HPC offline to run deep learning to train the models for object detection and object segmentation, where the objects are chickens in images taken on farm. The models can be ported from HPC to edge AI devices to create a new type of computer vision kit to enhance the existing digital poultry farm platform. Such new sensors enable implementing functions such as counting chickens, detection of dead chickens, and even assessing their weight or detecting uneven growth. These functions combined with the monitoring of environmental parameters, could enable early disease detection and improve the decision-making process. The experiment focused on Faster R-CNN architectures and AutoML was used to identify the most suitable architecture for chicken detection and segmentation for the given dataset. For the selected architectures, further hyperparameter optimization was carried out and we achieved the accuracy of AP = 85%, AP50 = 98%, and AP75 = 96% for object detection and AP = 90%, AP50 = 98%, and AP75 = 96% for instance segmentation. These models were installed on edge AI devices and evaluated in the online mode on actual poultry farms. Initial results are promising, but further development of the dataset and improvements in prediction models is needed.

Throughput of an IEEE 802.11 Wireless Network in the Presence of Wireless Audio Transmission: A Laboratory Analysis.

Wireless networks, including IEEE 802.11-based or Wi-Fi networks, are inexpensive and easy to install and therefore serve as useful connectivity alternatives in areas lacking wired-network infrastructure. However, IEEE 802.11 networks may not always provide the seamless connectivity and minimal throughput required for Industry 4.0 communications because of their susceptibility to interference from other devices operating in the unlicensed "Industrial, Scientific, and Medical" frequency band. Here we analyzed how a wireless audio transmitter operating on this band influences the throughput of an IEEE 802.11 b/g/n network under laboratory conditions. Wireless audio transmission reduced mean throughput by 85%, rendering the IEEE 802.11 b/g/n network nearly unusable. Our analysis suggests that in order for IEEE 802.11 wireless networks to support Industrial 4.0 applications, attention should be paid to the physical layer as well as the data or upper layers, and critical services should not transmit on the 2.4 GHz band. These findings may contribute to understanding and managing IEEE 802.11 wireless networks in various Industry 4.0 contexts.