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

Čakić, Stevan; Popović, Tomo; Krčo, Srdjan; Nedic, Daliborka; Babić, Dejan; Jovović, Ivan

doi:10.3390/s23063002

Cited by 10 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The system uses edge computing to analyze and process the data captured from environmental sensors and thermal imagers. Cakic et al 127 have proposed an edge−IoTbased computer vision system to monitor the farm, weight of birds, and early disease detection in birds. Although edge computing has many advantages, it cannot replace cloud-based solutions.…”

Section: Data Analysis Methodsmentioning

confidence: 99%

A Literature Review on the Role of Internet of Things, Computer Vision, and Sound Analysis in a Smart Poultry Farm

Goyal,

Yadav,

Mukherjee

2024

ACS Agric. Sci. Technol.

View full text Add to dashboard Cite

The rapid growth of human population and frequent changes in environmental conditions pose challenges in providing nutrient-rich food to the current population. The demand for poultry products is increasing exponentially as they are a good source of low-cost proteins. Ensuring the well-being of birds and delivering nutritious poultry products is essential for both current and future food security needs and sustainable agriculture. This article is motivated by the potential of emerging digital technologies in agriculture to provide innovative solutions to challenges faced by farmers, particularly in poultry farming. It aims to shed light on various issues encountered by farmers like diseases on poultry farms and offers valuable insights for researchers seeking to address these challenges using the advancements in digital technologies. Precision farming in poultry involves leveraging technologies such as the internet of things, artificial intelligence, and edge computing to enhance animal health management. A smart poultry farm maintains the farm environment and detects diseases at early stage in the chickens. The farm resources are utilized optimally in the smart poultry farm. This research work provides a systematic literature review of intelligent systems designed for (i) poultry birds' health and welfare management using technologies like the internet of things (IoT), artificial intelligence, cloud computing, and edge computing and (ii) early disease detection, estimation of weight, and feeding behavior of birds with the use of computer vision and vocalization analysis. This article also explores various sensors employed in the development of IoT infrastructure. It has been demonstrated in the literature that modern digital technologies automate the management operations of poultry farms. As a result, the farmer's income increases, and high-quality food products are available to people at low cost.

show abstract

Section: Data Analysis Methodsmentioning

confidence: 99%

A Literature Review on the Role of Internet of Things, Computer Vision, and Sound Analysis in a Smart Poultry Farm

Goyal,

Yadav,

Mukherjee

2024

ACS Agric. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…Para el entrenamiento de los datos, los investigadores utilizaron un aproximado de 4000 imágenes y se amplió con la herramienta RoboFlow a 9000, de las cuales se usaron para el entrenamiento 7550, para la validación 725 y para las pruebas del modelo 725. También, para el desarrollo de modelos se utilizó el lenguaje de programación Python y el paquete Detectron2 que incluye variantes de redes neuronales profundas (Cakic et al, 2023).…”

Section: Con Respecto a La Investigación "Developing Edge Ai Computer...unclassified

Machine learning aplicado al análisis de un set de datos de parámetros ambientales en galpones de pollos de engorde

Batista-Mendoza,

Cedeño Herrera,

Cedeño-Batista

2023

Vis. Antataura

View full text Add to dashboard Cite

Este artículo trata sobre el desarrollo de un modelo de Machine Learning aplicado a un conjunto de datos recopilados en una granja avícola. Su objetivo es obtener un modelo predictivo basado en variables ambientales para anticipar eventos futuros. Este modelo predictivo busca optimizar decisiones relacionadas con el bienestar ambiental de las aves y reducir costos de producción en proyectos avícolas. Para esta investigación, se obtuvo el set de datos del sistema "Smart Poultry Farm" como guía del desarrollo, se utilizó la metodología SEMMA y el lenguaje de programación Python en el entorno IDE Google Colaboratory. El modelo se construyó mediante el algoritmo de regresión logística binomial en el contexto de aprendizaje supervisado. La evaluación del modelo predictivo incluyó la matriz de confusión y métricas como: el Índice General de Calidad, Precisión, Sensibilidad, Especificidad y F1-Score. Se aplicaron distintos escenarios para anticipar la activación/desactivación de los ventiladores del galpón, en función de los parámetros ambientales: humedad, temperatura e índice de calor.

show abstract

“…As a result, the role of AI in poultry farming has gained significant prominence [6]. In the poultry industry, the application of image processing technologies has produced impressive results [7][8][9][10]. However, data collection and capture commonly rely on IoT technologies.…”

Section: Introductionmentioning

confidence: 99%

Towards Early Poultry Health Prediction through Non-Invasive and Computer Vision-Based Dropping Classification

Nakrosis,

Paulauskaite-Taraseviciene,

Raudonis

et al. 2023

Animals

View full text Add to dashboard Cite

The use of artificial intelligence techniques with advanced computer vision techniques offers great potential for non-invasive health assessments in the poultry industry. Evaluating the condition of poultry by monitoring their droppings can be highly valuable as significant changes in consistency and color can be indicators of serious and infectious diseases. While most studies have prioritized the classification of droppings into two categories (normal and abnormal), with some relevant studies dealing with up to five categories, this investigation goes a step further by employing image processing algorithms to categorize droppings into six classes, based on visual information indicating some level of abnormality. To ensure a diverse dataset, data were collected in three different poultry farms in Lithuania by capturing droppings on different types of litter. With the implementation of deep learning, the object detection rate reached 92.41% accuracy. A range of machine learning algorithms, including different deep learning architectures, has been explored and, based on the obtained results, we have proposed a comprehensive solution by combining different models for segmentation and classification purposes. The results revealed that the segmentation task achieved the highest accuracy of 0.88 in terms of the Dice coefficient employing the K-means algorithm. Meanwhile, YOLOv5 demonstrated the highest classification accuracy, achieving an ACC of 91.78%.

show abstract

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

Cited by 10 publications

References 26 publications

A Literature Review on the Role of Internet of Things, Computer Vision, and Sound Analysis in a Smart Poultry Farm

A Literature Review on the Role of Internet of Things, Computer Vision, and Sound Analysis in a Smart Poultry Farm

Machine learning aplicado al análisis de un set de datos de parámetros ambientales en galpones de pollos de engorde

Towards Early Poultry Health Prediction through Non-Invasive and Computer Vision-Based Dropping Classification

Contact Info

Product

Resources

About