Heat stress assessment by swine related vocalizations

Ferrari, Sara; Costa, Annamaria; Guarino, Marcella

doi:10.1016/j.livsci.2012.10.013

Cited by 21 publications

(13 citation statements)

References 19 publications

(16 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For instance, cough and sneeze audio data can provide information for recognizing in advance respiratory diseases in poultry, pig, and cattle, with precision scores (percentage of samples predicted to be coughing that are true coughing) ranging between 88.4 and 97.6% in the best scenarios (Carpentier et al, 2018(Carpentier et al, , 2019Yin et al, 2020). The sound analysis also has been used for identifying heat and management related distress conditions using pigs' vocalization generated data, achieving accuracies over 80% (Cordeiro et al, 2013(Cordeiro et al, , 2018Ferrari et al, 2013) and for acoustic monitoring of intake behavior in broilers, cattle, and sheep, also delivering reliable predictions (Clapham et al, 2011;Galli et al, 2011;Aydin et al, 2014;Aydin and Berckmans, 2016;Chelotti et al, 2016), as evidenced by the good determination coefficients achieved in those studies (higher than 0.89 and up to 0.995). These results reinforce thereof that the possibilities of exploring soundgenerated data in the animal science discipline are vast.…”

Section: Discussionmentioning

confidence: 99%

Integrating Audio Signal Processing and Deep Learning Algorithms for Gait Pattern Classification in Brazilian Gaited Horses

et al. 2021

View full text Add to dashboard Cite

This study focused on assessing the usefulness of using audio signal processing in the gaited horse industry. A total of 196 short-time audio files (4 s) were collected from video recordings of Brazilian gaited horses. These files were converted into waveform signals (196 samples by 80,000 columns) and divided into training (N = 164) and validation (N = 32) datasets. Twelve single-valued audio features were initially extracted to summarize the training data according to the gait patterns (Marcha Batida—MB and Marcha Picada—MP). After preliminary analyses, high-dimensional arrays of the Mel Frequency Cepstral Coefficients (MFCC), Onset Strength (OS), and Tempogram (TEMP) were extracted and used as input information in the classification algorithms. A principal component analysis (PCA) was performed using the 12 single-valued features set and each audio-feature dataset—AFD (MFCC, OS, and TEMP) for prior data visualization. Machine learning (random forest, RF; support vector machine, SVM) and deep learning (multilayer perceptron neural networks, MLP; convolution neural networks, CNN) algorithms were used to classify the gait types. A five-fold cross-validation scheme with 10 repetitions was employed for assessing the models' predictive performance. The classification performance across models and AFD was also validated with independent observations. The models and AFD were compared based on the classification accuracy (ACC), specificity (SPEC), sensitivity (SEN), and area under the curve (AUC). In the logistic regression analysis, five out of the 12 audio features extracted were significant (p < 0.05) between the gait types. ACC averages ranged from 0.806 to 0.932 for MFCC, from 0.758 to 0.948 for OS and, from 0.936 to 0.968 for TEMP. Overall, the TEMP dataset provided the best classification accuracies for all models. The most suitable method for audio-based horse gait pattern classification was CNN. Both cross and independent validation schemes confirmed that high values of ACC, SPEC, SEN, and AUC are expected for yet-to-be-observed labels, except for MFCC-based models, in which clear overfitting was observed. Using audio-generated data for describing gait phenotypes in Brazilian horses is a promising approach, as the two gait patterns were correctly distinguished. The highest classification performance was achieved by combining CNN and the rhythmic-descriptive AFD.

show abstract

Section: Discussionmentioning

confidence: 99%

Integrating Audio Signal Processing and Deep Learning Algorithms for Gait Pattern Classification in Brazilian Gaited Horses

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In these studies on high frequency vocalisations during different situations were analysed such as diverse castration practices [ 10 – 12 ], cold [ 13 ] or warm [ 14 ] temperatures. Other examples were the simulated crushing of piglets [ 15 ] or an electric shock or anticipation to the electric shock [ 16 ].…”

Section: Introductionmentioning

confidence: 99%

Discerning Pig Screams in Production Environments

et al. 2015

Self Cite

View full text Add to dashboard Cite

Pig vocalisations convey information about their current state of health and welfare. Continuously monitoring these vocalisations can provide useful information for the farmer. For instance, pig screams can indicate stressful situations. When monitoring screams, other sounds can interfere with scream detection. Therefore, identifying screams from other sounds is essential. The objective of this study was to understand which sound features define a scream. Therefore, a method to detect screams based on sound features with physical meaning and explicit rules was developed. To achieve this, 7 hours of labelled data from 24 pigs was used. The developed detection method attained 72% sensitivity, 91% specificity and 83% precision. As a result, the detection method showed that screams contain the following features discerning them from other sounds: a formant structure, adequate power, high frequency content, sufficient variability and duration.

show abstract

“…For this reason, one possible way to make animal welfare assessment easier and faster could be 51 the application of audio and video data analysis. (Tefera, 2012) (Ferrari et al, 2013;Tullo et al, 2013). 52 Image analysis, in particular, was successfully used to estimate the body weight of the animals (Mollah et 53 al., 2010) while audio analysis have been widely used to better identify specific behaviours and vocalisation 54 patterns in different animals' species (Chan et al, 2011;Vandermeulen et al, 2013).…”

mentioning

confidence: 99%

An innovative approach to predict the growth in intensive poultry farming

Fontana

Tullo

Butterworth

et al. 2015

Computers and Electronics in Agriculture

View full text Add to dashboard Cite

General rightsThis document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Full terms of use are available: http://www.bristol.ac.uk/pure/about/ebr-terms Abstract 10Chicken weight provides information about growth and feed conversion of the flock in order to identify 11 deviations from the expected homogeneous growth trend of the birds. This paper proposes a novel method 12 to automatically measure the growth rate of broiler chickens by sound analysis. 13Through the application of process engineering, Precision Livestock Farming (PLF) can combine audio and 14 video information into on-line automated tools that can be used to control, monitor and model the 15 behaviour, health and production of animals and their biological response. 16The aim of this study was to record and analyse broiler vocalisations under normal farm conditions, to 17 identify the relation between animal sounds and growth trend. Recordings were made at regular intervals, 18 during the entire life of birds, in order to evaluate the variation of frequency and bandwidth of the sounds 19 emitted by the animals. 20Two experimental trials were carried out in an indoor reared broiler farm; the audio recording procedures 21 lasted for 38 days. The recordings were made, in an automated, non-invasive and non-intrusive way and 22 without disturbing the animals in to the broiler house. Once a week, 50 birds were selected at random and 23 their weight recorded in order to follow the growth trend in the birds. 24Sound recordings were manually analysed and labelled using the Adobe Audition CS6 software. 25 2 Analysing the sounds recorded, it was possible to find a significant correlation (P<0.001) between the 26 frequencies of the vocalisations recorded and the weight of the broilers. 27The results explained how the frequency of the sounds emitted by the animals was inversely proportional 28 to the age and to the weight of the broilers; the more they grow, the lower the frequency of the sounds 29 emitted by the animals. 30This preliminary study shows how this method based on the identification of specific frequencies of the 31 sounds, in an indoor reared broiler farm, linked to the age and to the weight of the birds, could be used as 32 an early warning method/system to evaluate the health and welfare status of the animals at farm level, 33 developing also an automated growth monitoring tool. 34

show abstract

Heat stress assessment by swine related vocalizations

Cited by 21 publications

References 19 publications

Integrating Audio Signal Processing and Deep Learning Algorithms for Gait Pattern Classification in Brazilian Gaited Horses

Integrating Audio Signal Processing and Deep Learning Algorithms for Gait Pattern Classification in Brazilian Gaited Horses

Discerning Pig Screams in Production Environments

An innovative approach to predict the growth in intensive poultry farming

Contact Info

Product

Resources

About