Fine-grained image classification on bats using VGG16-CBAM: a practical example with 7 horseshoe bats taxa (CHIROPTERA: Rhinolophidae: Rhinolophus) from Southern China
Abstract Background Rapid identification and classification of bats are critical for practical applications. However, species identification of bats is a typically detrimental and time-consuming manual task that depends on taxonomists and well-trained experts. Deep Convolutional Neural Networks (DCN...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2024-04-01
|
| Series: | Frontiers in Zoology |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12983-024-00531-5 |
| _version_ | 1797219696166043648 |
|---|---|
| author | Zhong Cao Kunhui Wang Jiawei Wen Chuxian Li Yi Wu Xiaoyun Wang Wenhua Yu |
| author_facet | Zhong Cao Kunhui Wang Jiawei Wen Chuxian Li Yi Wu Xiaoyun Wang Wenhua Yu |
| author_sort | Zhong Cao |
| collection | DOAJ |
| description | Abstract Background Rapid identification and classification of bats are critical for practical applications. However, species identification of bats is a typically detrimental and time-consuming manual task that depends on taxonomists and well-trained experts. Deep Convolutional Neural Networks (DCNNs) provide a practical approach for the extraction of the visual features and classification of objects, with potential application for bat classification. Results In this study, we investigated the capability of deep learning models to classify 7 horseshoe bat taxa (CHIROPTERA: Rhinolophus) from Southern China. We constructed an image dataset of 879 front, oblique, and lateral targeted facial images of live individuals collected during surveys between 2012 and 2021. All images were taken using a standard photograph protocol and setting aimed at enhancing the effectiveness of the DCNNs classification. The results demonstrated that our customized VGG16-CBAM model achieved up to 92.15% classification accuracy with better performance than other mainstream models. Furthermore, the Grad-CAM visualization reveals that the model pays more attention to the taxonomic key regions in the decision-making process, and these regions are often preferred by bat taxonomists for the classification of horseshoe bats, corroborating the validity of our methods. Conclusion Our finding will inspire further research on image-based automatic classification of chiropteran species for early detection and potential application in taxonomy. |
| first_indexed | 2024-04-24T12:37:45Z |
| format | Article |
| id | doaj.art-d71e6a3634454d4d84a85924d1c96dc1 |
| institution | Directory Open Access Journal |
| issn | 1742-9994 |
| language | English |
| last_indexed | 2024-04-24T12:37:45Z |
| publishDate | 2024-04-01 |
| publisher | BMC |
| record_format | Article |
| series | Frontiers in Zoology |
| spelling | doaj.art-d71e6a3634454d4d84a85924d1c96dc12024-04-07T11:25:58ZengBMCFrontiers in Zoology1742-99942024-04-0121111310.1186/s12983-024-00531-5Fine-grained image classification on bats using VGG16-CBAM: a practical example with 7 horseshoe bats taxa (CHIROPTERA: Rhinolophidae: Rhinolophus) from Southern ChinaZhong Cao0Kunhui Wang1Jiawei Wen2Chuxian Li3Yi Wu4Xiaoyun Wang5Wenhua Yu6School of Electronics and Communication Engineering, Guangzhou UniversitySchool of Electronics and Communication Engineering, Guangzhou UniversitySchool of Electronics and Communication Engineering, Guangzhou UniversitySchool of Electronics and Communication Engineering, Guangzhou UniversitySchool of Life Sciences, Guangzhou UniversitySchool of Life Sciences, Guangzhou UniversitySchool of Life Sciences, Guangzhou UniversityAbstract Background Rapid identification and classification of bats are critical for practical applications. However, species identification of bats is a typically detrimental and time-consuming manual task that depends on taxonomists and well-trained experts. Deep Convolutional Neural Networks (DCNNs) provide a practical approach for the extraction of the visual features and classification of objects, with potential application for bat classification. Results In this study, we investigated the capability of deep learning models to classify 7 horseshoe bat taxa (CHIROPTERA: Rhinolophus) from Southern China. We constructed an image dataset of 879 front, oblique, and lateral targeted facial images of live individuals collected during surveys between 2012 and 2021. All images were taken using a standard photograph protocol and setting aimed at enhancing the effectiveness of the DCNNs classification. The results demonstrated that our customized VGG16-CBAM model achieved up to 92.15% classification accuracy with better performance than other mainstream models. Furthermore, the Grad-CAM visualization reveals that the model pays more attention to the taxonomic key regions in the decision-making process, and these regions are often preferred by bat taxonomists for the classification of horseshoe bats, corroborating the validity of our methods. Conclusion Our finding will inspire further research on image-based automatic classification of chiropteran species for early detection and potential application in taxonomy.https://doi.org/10.1186/s12983-024-00531-5Attention mechanismBatConvolutional block attention module (CBAM)Species identificationVGG16 |
| spellingShingle | Zhong Cao Kunhui Wang Jiawei Wen Chuxian Li Yi Wu Xiaoyun Wang Wenhua Yu Fine-grained image classification on bats using VGG16-CBAM: a practical example with 7 horseshoe bats taxa (CHIROPTERA: Rhinolophidae: Rhinolophus) from Southern China Frontiers in Zoology Attention mechanism Bat Convolutional block attention module (CBAM) Species identification VGG16 |
| title | Fine-grained image classification on bats using VGG16-CBAM: a practical example with 7 horseshoe bats taxa (CHIROPTERA: Rhinolophidae: Rhinolophus) from Southern China |
| title_full | Fine-grained image classification on bats using VGG16-CBAM: a practical example with 7 horseshoe bats taxa (CHIROPTERA: Rhinolophidae: Rhinolophus) from Southern China |
| title_fullStr | Fine-grained image classification on bats using VGG16-CBAM: a practical example with 7 horseshoe bats taxa (CHIROPTERA: Rhinolophidae: Rhinolophus) from Southern China |
| title_full_unstemmed | Fine-grained image classification on bats using VGG16-CBAM: a practical example with 7 horseshoe bats taxa (CHIROPTERA: Rhinolophidae: Rhinolophus) from Southern China |
| title_short | Fine-grained image classification on bats using VGG16-CBAM: a practical example with 7 horseshoe bats taxa (CHIROPTERA: Rhinolophidae: Rhinolophus) from Southern China |
| title_sort | fine grained image classification on bats using vgg16 cbam a practical example with 7 horseshoe bats taxa chiroptera rhinolophidae rhinolophus from southern china |
| topic | Attention mechanism Bat Convolutional block attention module (CBAM) Species identification VGG16 |
| url | https://doi.org/10.1186/s12983-024-00531-5 |
| work_keys_str_mv | AT zhongcao finegrainedimageclassificationonbatsusingvgg16cbamapracticalexamplewith7horseshoebatstaxachiropterarhinolophidaerhinolophusfromsouthernchina AT kunhuiwang finegrainedimageclassificationonbatsusingvgg16cbamapracticalexamplewith7horseshoebatstaxachiropterarhinolophidaerhinolophusfromsouthernchina AT jiaweiwen finegrainedimageclassificationonbatsusingvgg16cbamapracticalexamplewith7horseshoebatstaxachiropterarhinolophidaerhinolophusfromsouthernchina AT chuxianli finegrainedimageclassificationonbatsusingvgg16cbamapracticalexamplewith7horseshoebatstaxachiropterarhinolophidaerhinolophusfromsouthernchina AT yiwu finegrainedimageclassificationonbatsusingvgg16cbamapracticalexamplewith7horseshoebatstaxachiropterarhinolophidaerhinolophusfromsouthernchina AT xiaoyunwang finegrainedimageclassificationonbatsusingvgg16cbamapracticalexamplewith7horseshoebatstaxachiropterarhinolophidaerhinolophusfromsouthernchina AT wenhuayu finegrainedimageclassificationonbatsusingvgg16cbamapracticalexamplewith7horseshoebatstaxachiropterarhinolophidaerhinolophusfromsouthernchina |