Bird Species Identification Using Spectrogram Based on Multi-Channel Fusion of DCNNs

Deep convolutional neural networks (DCNNs) have achieved breakthrough performance on bird species identification using a spectrogram of bird vocalization. Aiming at the imbalance of the bird vocalization dataset, a single feature identification model (SFIM) with residual blocks and modified, weighte...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Feiyu Zhang, Luyang Zhang, Hongxiang Chen, Jiangjian Xie
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	bird vocalization spectrogram feature multi-channel deep convolutional neural Science Q Astrophysics QB460-466 Physics QC1-999
Acceso en línea:	https://doaj.org/article/2ca38a7e0955475892901a44421f3ddc
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:2ca38a7e0955475892901a44421f3ddc
record_format	dspace
spelling	oai:doaj.org-article:2ca38a7e0955475892901a44421f3ddc2021-11-25T17:30:21ZBird Species Identification Using Spectrogram Based on Multi-Channel Fusion of DCNNs10.3390/e231115071099-4300https://doaj.org/article/2ca38a7e0955475892901a44421f3ddc2021-11-01T00:00:00Zhttps://www.mdpi.com/1099-4300/23/11/1507https://doaj.org/toc/1099-4300Deep convolutional neural networks (DCNNs) have achieved breakthrough performance on bird species identification using a spectrogram of bird vocalization. Aiming at the imbalance of the bird vocalization dataset, a single feature identification model (SFIM) with residual blocks and modified, weighted, cross-entropy function was proposed. To further improve the identification accuracy, two multi-channel fusion methods were built with three SFIMs. One of these fused the outputs of the feature extraction parts of three SFIMs (feature fusion mode), the other fused the outputs of the classifiers of three SFIMs (result fusion mode). The SFIMs were trained with three different kinds of spectrograms, which were calculated through short-time Fourier transform, mel-frequency cepstrum transform and chirplet transform, respectively. To overcome the shortage of the huge number of trainable model parameters, transfer learning was used in the multi-channel models. Using our own vocalization dataset as a sample set, it is found that the result fusion mode model outperforms the other proposed models, the best mean average precision (MAP) reaches 0.914. Choosing three durations of spectrograms, 100 ms, 300 ms and 500 ms for comparison, the results reveal that the 300 ms duration is the best for our own dataset. The duration is suggested to be determined based on the duration distribution of bird syllables. As for the performance with the training dataset of BirdCLEF2019, the highest classification mean average precision (cmAP) reached 0.135, which means the proposed model has certain generalization ability.Feiyu ZhangLuyang ZhangHongxiang ChenJiangjian XieMDPI AGarticlebird vocalizationspectrogram featuremulti-channeldeep convolutional neuralScienceQAstrophysicsQB460-466PhysicsQC1-999ENEntropy, Vol 23, Iss 1507, p 1507 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	bird vocalization spectrogram feature multi-channel deep convolutional neural Science Q Astrophysics QB460-466 Physics QC1-999
spellingShingle	bird vocalization spectrogram feature multi-channel deep convolutional neural Science Q Astrophysics QB460-466 Physics QC1-999 Feiyu Zhang Luyang Zhang Hongxiang Chen Jiangjian Xie Bird Species Identification Using Spectrogram Based on Multi-Channel Fusion of DCNNs
description	Deep convolutional neural networks (DCNNs) have achieved breakthrough performance on bird species identification using a spectrogram of bird vocalization. Aiming at the imbalance of the bird vocalization dataset, a single feature identification model (SFIM) with residual blocks and modified, weighted, cross-entropy function was proposed. To further improve the identification accuracy, two multi-channel fusion methods were built with three SFIMs. One of these fused the outputs of the feature extraction parts of three SFIMs (feature fusion mode), the other fused the outputs of the classifiers of three SFIMs (result fusion mode). The SFIMs were trained with three different kinds of spectrograms, which were calculated through short-time Fourier transform, mel-frequency cepstrum transform and chirplet transform, respectively. To overcome the shortage of the huge number of trainable model parameters, transfer learning was used in the multi-channel models. Using our own vocalization dataset as a sample set, it is found that the result fusion mode model outperforms the other proposed models, the best mean average precision (MAP) reaches 0.914. Choosing three durations of spectrograms, 100 ms, 300 ms and 500 ms for comparison, the results reveal that the 300 ms duration is the best for our own dataset. The duration is suggested to be determined based on the duration distribution of bird syllables. As for the performance with the training dataset of BirdCLEF2019, the highest classification mean average precision (cmAP) reached 0.135, which means the proposed model has certain generalization ability.
format	article
author	Feiyu Zhang Luyang Zhang Hongxiang Chen Jiangjian Xie
author_facet	Feiyu Zhang Luyang Zhang Hongxiang Chen Jiangjian Xie
author_sort	Feiyu Zhang
title	Bird Species Identification Using Spectrogram Based on Multi-Channel Fusion of DCNNs
title_short	Bird Species Identification Using Spectrogram Based on Multi-Channel Fusion of DCNNs
title_full	Bird Species Identification Using Spectrogram Based on Multi-Channel Fusion of DCNNs
title_fullStr	Bird Species Identification Using Spectrogram Based on Multi-Channel Fusion of DCNNs
title_full_unstemmed	Bird Species Identification Using Spectrogram Based on Multi-Channel Fusion of DCNNs
title_sort	bird species identification using spectrogram based on multi-channel fusion of dcnns
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/2ca38a7e0955475892901a44421f3ddc
work_keys_str_mv	AT feiyuzhang birdspeciesidentificationusingspectrogrambasedonmultichannelfusionofdcnns AT luyangzhang birdspeciesidentificationusingspectrogrambasedonmultichannelfusionofdcnns AT hongxiangchen birdspeciesidentificationusingspectrogrambasedonmultichannelfusionofdcnns AT jiangjianxie birdspeciesidentificationusingspectrogrambasedonmultichannelfusionofdcnns
_version_	1718412274846138368

Bird Species Identification Using Spectrogram Based on Multi-Channel Fusion of DCNNs

Ejemplares similares