Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/2067/49318
Titolo: An innovative technique for faecal score classification based on RGB images and Artificial Intelligence algorithms
Autori: Ortenzi, Luciano 
Violino, S.
Costa, Corrado
Figorilli, Simone
Vasta, Simone
Tocci, Francesco
Moscovini, Lavinia
Basiric√≤, Loredana 
Evangelista, C.
Pallottino, Federico
Bernabucci, Umberto 
Rivista: THE JOURNAL OF AGRICULTURAL SCIENCE 
Data pubblicazione: 2023
Abstract: 
The milk production is strongly influenced by the dairy cow welfare related to a good nutrition and the analysis of the digestibility of feeds allows to evaluate the health status of the animals. Through faeces visual examination it is possible to estimate the quality of diet fed in terms of lacking in fibre or too high in non-structural carbohydrates. The study was carried out in 2021, on 4 dairy farms in Central Italy. The purpose of this work is the classification and evaluation of dairy cow faeces using RGB image analysis through an artificial intelligence (CNN) algorithm. The main features to analyse are pH, colour and consistency. For the latter two RGB imaging was used combined with deep learning and Artificial Intelligence (AI) to reach objectivity in samples evaluation. The images have been captured with several smartphones and cameras, in various light conditions, collecting a dataset of 441 images. Image acquired by RGB cameras are then analysed through Convolutional Neural Networks (CNNs) technology that extract features and data previously standardized by a faecal score index (FCI) assigned after a visual analysis and based on 5 classes. The results achieved with different training strategy show a training accuracy of 90% and a validation accuracy of 78% of the model which allow to identify problems in bovine digestion and to intervene promptly in feed variation. The method used in this study eliminates subjectivity in field analysis and allows future improvement increasing the data set to strengthen the model.
URI: http://hdl.handle.net/2067/49318
ISSN: 1469-5146
DOI: 10.1017/S0021859623000114
Diritti: Attribution 4.0 International
È visualizzato nelle collezioni:A1. Articolo in rivista

File in questo documento:
File Descrizione DimensioniFormato
2023_Ortenzi et al_FecalScore_Pubb.pdfManoscritto577.32 kBAdobe PDFVisualizza/apri
Visualizza tutti i metadati del documento

Page view(s)

37
Last Week
0
Last month
4
controllato il 29-nov-2023

Download(s)

5
controllato il 29-nov-2023

Google ScholarTM

Check

Altmetric


Questo documento è distribuito in accordo con Licenza Creative Commons Creative Commons