In-situ classification of cattle behavior using accelerometry data

Arablouei, Reza; Currie, Lachlan; Kusý, Branislav; Ingham, Aaron; Greenwood, P. L.; Bishop-Hurley, Greg

doi:10.1016/j.compag.2021.106045

Cited by 43 publications

(34 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The first way consists in sending the raw information, through a radio transmitter device, to another system for post-processing; and the second one, called edge-computing, consists in processing the information in the device itself and sending the new resulting information [ 15 , 16 ]. The processing can be done using statistical models like [ 15 , 16 ] or using Embedded Neural Network (ENN) algorithms like [ 17 ]. This computing paradigm, which calls for processing the data at the edge of the device [ 18 ], has several advantages, such as real-time analysis, low latency, security, and reduced data delivery, which can be beneficial in the field of animal monitoring.…”

Section: Introductionmentioning

confidence: 99%

Wildlife Monitoring on the Edge: A Performance Evaluation of Embedded Neural Networks on Microcontrollers for Animal Behavior Classification

Domínguez-Morales

Durán-López

Gutiérrez-Galán

et al. 2021

Sensors

View full text Add to dashboard Cite

Monitoring animals’ behavior living in wild or semi-wild environments is a very interesting subject for biologists who work with them. The difficulty and cost of implanting electronic devices in this kind of animals suggest that these devices must be robust and have low power consumption to increase their battery life as much as possible. Designing a custom smart device that can detect multiple animal behaviors and that meets the mentioned restrictions presents a major challenge that is addressed in this work. We propose an edge-computing solution, which embeds an ANN in a microcontroller that collects data from an IMU sensor to detect three different horse gaits. All the computation is performed in the microcontroller to reduce the amount of data transmitted via wireless radio, since sending information is one of the most power-consuming tasks in this type of devices. Multiples ANNs were implemented and deployed in different microcontroller architectures in order to find the best balance between energy consumption and computing performance. The results show that the embedded networks obtain up to 97.96% ± 1.42% accuracy, achieving an energy efficiency of 450 Mops/s/watt.

show abstract

Section: Introductionmentioning

confidence: 99%

Wildlife Monitoring on the Edge: A Performance Evaluation of Embedded Neural Networks on Microcontrollers for Animal Behavior Classification

Domínguez-Morales

Durán-López

Gutiérrez-Galán

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Lin and others analyzed the CIAA project in Argentina, and their current work includes designing an integrated intelligent system based on the EDU-CIAA NXP version for educational and applied research purposes [10]. Arablouei show that this is achieved without causing any significant burden on intelligent system resources in terms of energy, computing, or memory [11].…”

Section: Related Workmentioning

confidence: 99%

Digital Development Strategy of Agricultural Planting and Breeding Enterprises Based on Intelligent Sensors

Zhang

2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

The digitalization of agricultural planting and breeding enterprises is the only way for agricultural development. Now with the development of various technologies, the digitalization of agricultural enterprises is becoming faster and faster. Today, the development of intelligent sensors provides platform support for the digitalization of agricultural enterprises. This article is aimed at introducing the application of intelligent sensors in agriculture to provide strategic research for the digital development of agricultural planting and breeding enterprises. This paper proposes the establishment of a system platform for network intelligent sensors and proposes the establishment of an agricultural short message management publishing platform. And the existing public information transmission methods are used to provide a cheap, simple, and fast way for agricultural producers to quickly obtain agricultural information, so as to provide a feasible plan for solving the agricultural “last mile of agriculture” problem. After inspection and analysis, the information management release platform can meet the design requirements, and the processing rate of short-term interest is above 98%, which can pave the way for the digital industrialization of agricultural enterprises.

show abstract

“…The first experiment took place in August 2018 at the Commonwealth Scientific and Industrial Research Organisation (CSIRO) FD McMaster Laboratory Chiswick, near Armidale, NSW, Australia (30 • 36'28.17"S, 151 • 32'39.12"E). There is a detailed description of the experiment and the equipment used including the collar tag in [13]. The accelerometry data was collected from ten steers wearing collar tags.…”

Section: Datasetsmentioning

confidence: 99%

“…Arablouei et al [13] devise a small set of computationally-efficient and informative features based on the accelerometry data's statistical and spectral properties. The validations show that resource-efficient discriminative models can yield good classification accuracy with the proposed features.…”

Section: Introductionmentioning

confidence: 99%

Animal Behavior Classification via Accelerometry Data and Recurrent Neural Networks

Wang¹,

Arablouei²,

Alvarenga³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

We study the classification of animal behavior using accelerometry data through various recurrent neural network (RNN) models. We evaluate the classification performance and complexity of the considered models, which feature long short-time memory (LSTM) or gated recurrent unit (GRU) architectures with varying depths and widths, using four datasets acquired from cattle via collar or ear tags. We also include two state-of-the-art convolutional neural network (CNN)-based time-series classification models in the evaluations.The results show that the RNN-based models can achieve similar or higher classification accuracy compared with the CNN-based models while having less computational and memory requirements. We also observe that the models with GRU architecture generally outperform the ones with LSTM architecture in terms of classification accuracy despite being less complex.A single-layer uni-directional GRU model with 64 hidden units appears to offer a good balance between accuracy and complexity making it suitable for implementation on edge/embedded devices.

show abstract

In-situ classification of cattle behavior using accelerometry data

Cited by 43 publications

References 24 publications

Wildlife Monitoring on the Edge: A Performance Evaluation of Embedded Neural Networks on Microcontrollers for Animal Behavior Classification

Wildlife Monitoring on the Edge: A Performance Evaluation of Embedded Neural Networks on Microcontrollers for Animal Behavior Classification

Digital Development Strategy of Agricultural Planting and Breeding Enterprises Based on Intelligent Sensors

Animal Behavior Classification via Accelerometry Data and Recurrent Neural Networks

Contact Info

Product

Resources

About