AI /ML Enabled by HPC for Edge Camera Devices for the Next Generation Hen Farms
Presentation of the problem and objective of the experiment
Researchers from this consortium have engaged a large number of poultry farms of different sizes. The main goal of the engagement was to understand their business challenges and to present an IoT based poultry farm management solution, supported by a set of sensors for environmental monitoring. The experiment targets the use of HPC and deep learning AI to create prediction models that can be deployed on the edge devices equipped with camera sensors for the use in IoT/AI solutions in the poultry sector.
Short description of the experiment
The experiment uses HPC and AI/ML to calculate prediction models that allow us to create a new type of precision agriculture sensors that combine camera, edge computing, and IoT platform in order to support the next generation of hen farms. The project leverages the existing poultry farm management platform and the use of HPC resources to develop unique and innovative solutions based on the AI-enhanced edge IoT camera nodes. AI/ML algorithms are being used to create prediction models such as detection of dead chickens, estimation of the weight of the chicken as well as monitoring behavioral patterns of the chicken which could indicate the appearance of certain diseases. The validation process will include feedback from the end-users and local NCC for HPC.
The idea of this experiment is to monitor the number of chickens, their weight, and dead chickens with the help of artificial intelligence (AI) and a camera in which AI models will be integrated, and placed on poultry farms. Currently, partners are working on the following ML models: a) detection of chicken, and b) segmentation of each chicken in the image. These models will be used for the development of IoT edge sensors to count the number of chickens in an image and/or video, and possibly assess their growth (weight and homogeneity). The goal of this experiment is to validate the use of HPC in AI/ML training and evaluate acceleration in the development of these prediction models.
End users: Radinović Company and Meso-Promet Franca
HPC Provider: Yotta Advanced Computing
Domain Expert: University of Donja Gorica
HPC & AI Expert: DigitalSmart
Partner University of Donja Gorica - UDG is part of the NCC Montenegro.