Automated Individual Cattle Identification Using Video Data: A Unified Deep Learning Architecture Approach

Individual cattle identification is a prerequisite and foundation for precision livestock farming. Existing methods for cattle identification require radio frequency or visual ear tags, all of which are prone to loss or damage. Here, we propose and implement a new unified deep learning approach to c...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Yongliang Qiao, Cameron Clark, Sabrina Lomax, He Kong, Daobilige Su, Salah Sukkarieh
Formato:	article
Lenguaje:	EN
Publicado:	Frontiers Media S.A. 2021
Materias:	cattle identification deep learning BiLSTM self-attention precision livestock farming Veterinary medicine SF600-1100
Acceso en línea:	https://doaj.org/article/f57e5f1f94bb4aedb72a854ad8899104
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:f57e5f1f94bb4aedb72a854ad8899104
record_format	dspace
spelling	oai:doaj.org-article:f57e5f1f94bb4aedb72a854ad88991042021-11-30T14:40:43ZAutomated Individual Cattle Identification Using Video Data: A Unified Deep Learning Architecture Approach2673-622510.3389/fanim.2021.759147https://doaj.org/article/f57e5f1f94bb4aedb72a854ad88991042021-11-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/fanim.2021.759147/fullhttps://doaj.org/toc/2673-6225Individual cattle identification is a prerequisite and foundation for precision livestock farming. Existing methods for cattle identification require radio frequency or visual ear tags, all of which are prone to loss or damage. Here, we propose and implement a new unified deep learning approach to cattle identification using video analysis. The proposed deep learning framework is composed of a Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) with a self-attention mechanism. More specifically, the Inception-V3 CNN was used to extract features from a cattle video dataset taken in a feedlot with rear-view. Extracted features were then fed to a BiLSTM layer to capture spatio-temporal information. Then, self-attention was employed to provide a different focus on the features captured by BiLSTM for the final step of cattle identification. We used a total of 363 rear-view videos from 50 cattle at three different times with an interval of 1 month between data collection periods. The proposed method achieved 93.3% identification accuracy using a 30-frame video length, which outperformed current state-of-the-art methods (Inception-V3, MLP, SimpleRNN, LSTM, and BiLSTM). Furthermore, two different attention schemes, namely, additive and multiplicative attention mechanisms were compared. Our results show that the additive attention mechanism achieved 93.3% accuracy and 91.0% recall, greater than multiplicative attention mechanism with 90.7% accuracy and 87.0% recall. Video length also impacted accuracy, with video sequence length up to 30-frames enhancing identification performance. Overall, our approach can capture key spatio-temporal features to improve cattle identification accuracy, enabling automated cattle identification for precision livestock farming.Yongliang QiaoCameron ClarkSabrina LomaxHe KongDaobilige SuSalah SukkariehFrontiers Media S.A.articlecattle identificationdeep learningBiLSTMself-attentionprecision livestock farmingVeterinary medicineSF600-1100ENFrontiers in Animal Science, Vol 2 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	cattle identification deep learning BiLSTM self-attention precision livestock farming Veterinary medicine SF600-1100
spellingShingle	cattle identification deep learning BiLSTM self-attention precision livestock farming Veterinary medicine SF600-1100 Yongliang Qiao Cameron Clark Sabrina Lomax He Kong Daobilige Su Salah Sukkarieh Automated Individual Cattle Identification Using Video Data: A Unified Deep Learning Architecture Approach
description	Individual cattle identification is a prerequisite and foundation for precision livestock farming. Existing methods for cattle identification require radio frequency or visual ear tags, all of which are prone to loss or damage. Here, we propose and implement a new unified deep learning approach to cattle identification using video analysis. The proposed deep learning framework is composed of a Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) with a self-attention mechanism. More specifically, the Inception-V3 CNN was used to extract features from a cattle video dataset taken in a feedlot with rear-view. Extracted features were then fed to a BiLSTM layer to capture spatio-temporal information. Then, self-attention was employed to provide a different focus on the features captured by BiLSTM for the final step of cattle identification. We used a total of 363 rear-view videos from 50 cattle at three different times with an interval of 1 month between data collection periods. The proposed method achieved 93.3% identification accuracy using a 30-frame video length, which outperformed current state-of-the-art methods (Inception-V3, MLP, SimpleRNN, LSTM, and BiLSTM). Furthermore, two different attention schemes, namely, additive and multiplicative attention mechanisms were compared. Our results show that the additive attention mechanism achieved 93.3% accuracy and 91.0% recall, greater than multiplicative attention mechanism with 90.7% accuracy and 87.0% recall. Video length also impacted accuracy, with video sequence length up to 30-frames enhancing identification performance. Overall, our approach can capture key spatio-temporal features to improve cattle identification accuracy, enabling automated cattle identification for precision livestock farming.
format	article
author	Yongliang Qiao Cameron Clark Sabrina Lomax He Kong Daobilige Su Salah Sukkarieh
author_facet	Yongliang Qiao Cameron Clark Sabrina Lomax He Kong Daobilige Su Salah Sukkarieh
author_sort	Yongliang Qiao
title	Automated Individual Cattle Identification Using Video Data: A Unified Deep Learning Architecture Approach
title_short	Automated Individual Cattle Identification Using Video Data: A Unified Deep Learning Architecture Approach
title_full	Automated Individual Cattle Identification Using Video Data: A Unified Deep Learning Architecture Approach
title_fullStr	Automated Individual Cattle Identification Using Video Data: A Unified Deep Learning Architecture Approach
title_full_unstemmed	Automated Individual Cattle Identification Using Video Data: A Unified Deep Learning Architecture Approach
title_sort	automated individual cattle identification using video data: a unified deep learning architecture approach
publisher	Frontiers Media S.A.
publishDate	2021
url	https://doaj.org/article/f57e5f1f94bb4aedb72a854ad8899104
work_keys_str_mv	AT yongliangqiao automatedindividualcattleidentificationusingvideodataaunifieddeeplearningarchitectureapproach AT cameronclark automatedindividualcattleidentificationusingvideodataaunifieddeeplearningarchitectureapproach AT sabrinalomax automatedindividualcattleidentificationusingvideodataaunifieddeeplearningarchitectureapproach AT hekong automatedindividualcattleidentificationusingvideodataaunifieddeeplearningarchitectureapproach AT daobiligesu automatedindividualcattleidentificationusingvideodataaunifieddeeplearningarchitectureapproach AT salahsukkarieh automatedindividualcattleidentificationusingvideodataaunifieddeeplearningarchitectureapproach
_version_	1718406540610764800

Automated Individual Cattle Identification Using Video Data: A Unified Deep Learning Architecture Approach

Ejemplares similares