Fish feeding behavior recognition model based on the fusion of visual and water quality features
ObjectiveTo improve the accuracy of fish feeding behavior recognition in industrial aquaculture environment. MethodA fish feeding behavior recognition model was proposed based on the fusion of visual and water quality features, namely MC-ConvNeXtV2. To better capture the global features of different...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
South China Agricultural University
2025-07-01
|
| Series: | Huanan Nongye Daxue xuebao |
| Subjects: | |
| Online Access: | https://journal.scau.edu.cn/article/doi/10.7671/j.issn.1001-411X.202410002 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849411876000104448 |
|---|---|
| author | Zheng ZHANG Bosheng ZOU |
| author_facet | Zheng ZHANG Bosheng ZOU |
| author_sort | Zheng ZHANG |
| collection | DOAJ |
| description | ObjectiveTo improve the accuracy of fish feeding behavior recognition in industrial aquaculture environment. MethodA fish feeding behavior recognition model was proposed based on the fusion of visual and water quality features, namely MC-ConvNeXtV2. To better capture the global features of different aggregation levels and the detailed features of feeding behavior, a context-aware local attention mechanism (Cloatt) was introduced in each convolution stage of ConvNeXtV2-T. To improve the behavior recognition performance of the model in high-density aquaculture, a multimodal feature fusion module (MFFM) was designed to achieve adaptive fusion of visual features and dissolve oxygen, temperature, and pH of water quality features. The model test was conducted in a Micropterus salmoides culture factory with a culture density of 160 fish/m3. ResultThe test results showed that for the task of four feeding behaviors classification of fish school, the recognition accuracy, precision and recall of MC-ConvNeXtV2 model were 96.89%, 96.34%, and 96.59%, respectively. Compared with ConvNeXtV2-T, these indicators increased by 3.11, 2.42, and 2.72 percentage points, respectively. ConclusionThe proposed fish feeding behavior recognition model offers a new approach for intelligent aquaculture management. |
| format | Article |
| id | doaj-art-aac600529d154166a88617de1bb036ed |
| institution | Kabale University |
| issn | 1001-411X |
| language | zho |
| publishDate | 2025-07-01 |
| publisher | South China Agricultural University |
| record_format | Article |
| series | Huanan Nongye Daxue xuebao |
| spelling | doaj-art-aac600529d154166a88617de1bb036ed2025-08-20T03:34:37ZzhoSouth China Agricultural UniversityHuanan Nongye Daxue xuebao1001-411X2025-07-0146453854810.7671/j.issn.1001-411X.202410002202504zhangzhengFish feeding behavior recognition model based on the fusion of visual and water quality featuresZheng ZHANG0Bosheng ZOU1College of Engineering Science and Technology, Shanghai Ocean University, Shanghai 201306, ChinaCollege of Engineering Science and Technology, Shanghai Ocean University, Shanghai 201306, ChinaObjectiveTo improve the accuracy of fish feeding behavior recognition in industrial aquaculture environment. MethodA fish feeding behavior recognition model was proposed based on the fusion of visual and water quality features, namely MC-ConvNeXtV2. To better capture the global features of different aggregation levels and the detailed features of feeding behavior, a context-aware local attention mechanism (Cloatt) was introduced in each convolution stage of ConvNeXtV2-T. To improve the behavior recognition performance of the model in high-density aquaculture, a multimodal feature fusion module (MFFM) was designed to achieve adaptive fusion of visual features and dissolve oxygen, temperature, and pH of water quality features. The model test was conducted in a Micropterus salmoides culture factory with a culture density of 160 fish/m3. ResultThe test results showed that for the task of four feeding behaviors classification of fish school, the recognition accuracy, precision and recall of MC-ConvNeXtV2 model were 96.89%, 96.34%, and 96.59%, respectively. Compared with ConvNeXtV2-T, these indicators increased by 3.11, 2.42, and 2.72 percentage points, respectively. ConclusionThe proposed fish feeding behavior recognition model offers a new approach for intelligent aquaculture management.https://journal.scau.edu.cn/article/doi/10.7671/j.issn.1001-411X.202410002aquaculturefish feeding behaviormultimodalconvnextv2 |
| spellingShingle | Zheng ZHANG Bosheng ZOU Fish feeding behavior recognition model based on the fusion of visual and water quality features Huanan Nongye Daxue xuebao aquaculture fish feeding behavior multimodal convnextv2 |
| title | Fish feeding behavior recognition model based on the fusion of visual and water quality features |
| title_full | Fish feeding behavior recognition model based on the fusion of visual and water quality features |
| title_fullStr | Fish feeding behavior recognition model based on the fusion of visual and water quality features |
| title_full_unstemmed | Fish feeding behavior recognition model based on the fusion of visual and water quality features |
| title_short | Fish feeding behavior recognition model based on the fusion of visual and water quality features |
| title_sort | fish feeding behavior recognition model based on the fusion of visual and water quality features |
| topic | aquaculture fish feeding behavior multimodal convnextv2 |
| url | https://journal.scau.edu.cn/article/doi/10.7671/j.issn.1001-411X.202410002 |
| work_keys_str_mv | AT zhengzhang fishfeedingbehaviorrecognitionmodelbasedonthefusionofvisualandwaterqualityfeatures AT boshengzou fishfeedingbehaviorrecognitionmodelbasedonthefusionofvisualandwaterqualityfeatures |