Convolutional Neural Network-Based Fish Posture Classification
Fish killing machines can effectively relieve the workers from the backbreaking labour. Generally, it is necessary to ensure the fish to be in unified posture before being input into the automatic fish killing machine. As such, how to detect the actual posture of fish in real time is a new and meani...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2021/9939688 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832560586714513408 |
|---|---|
| author | Xin Li Anzi Ding Shaojie Mei Wenjin Wu Wenguang Hou |
| author_facet | Xin Li Anzi Ding Shaojie Mei Wenjin Wu Wenguang Hou |
| author_sort | Xin Li |
| collection | DOAJ |
| description | Fish killing machines can effectively relieve the workers from the backbreaking labour. Generally, it is necessary to ensure the fish to be in unified posture before being input into the automatic fish killing machine. As such, how to detect the actual posture of fish in real time is a new and meaningful issue. Considering that in the actual situation, we only need to determine the four postures which are related to the head, tail, back, and belly of the fish, and we transfer this task into a four-kind classification problem. As such, the convolutional neural network (CNN) is introduced here to do classification and then to detect the fish’s posture. Before training the network, all sample images are preprocessed to make the fish be horizontal on the image according to the principal component analysis. Meanwhile, the histogram equalization is used to make the grey distribution of different images be close. After that, two kinds of strategies are taken to do classification. The first is a paired binary classification CNN and the second is a four-category CNN. In addition, three kinds of CNN are adopted. By comparison, the four-kind classification can obtain better results with error less than 1/1000. |
| format | Article |
| id | doaj-art-282838436bbf4e6b8901bc6b464e8f6d |
| institution | Kabale University |
| issn | 1076-2787 1099-0526 |
| language | English |
| publishDate | 2021-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj-art-282838436bbf4e6b8901bc6b464e8f6d2025-02-03T01:27:08ZengWileyComplexity1076-27871099-05262021-01-01202110.1155/2021/99396889939688Convolutional Neural Network-Based Fish Posture ClassificationXin Li0Anzi Ding1Shaojie Mei2Wenjin Wu3Wenguang Hou4Institute of Agricultural Products Processing and Nuclear Agricultural Technology, Hubei Academy of Agricultural Sciences, Wuhan 430064, Hubei, ChinaInstitute of Agricultural Products Processing and Nuclear Agricultural Technology, Hubei Academy of Agricultural Sciences, Wuhan 430064, Hubei, ChinaCollege of Life Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, Hubei, ChinaInstitute of Agricultural Products Processing and Nuclear Agricultural Technology, Hubei Academy of Agricultural Sciences, Wuhan 430064, Hubei, ChinaCollege of Life Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, Hubei, ChinaFish killing machines can effectively relieve the workers from the backbreaking labour. Generally, it is necessary to ensure the fish to be in unified posture before being input into the automatic fish killing machine. As such, how to detect the actual posture of fish in real time is a new and meaningful issue. Considering that in the actual situation, we only need to determine the four postures which are related to the head, tail, back, and belly of the fish, and we transfer this task into a four-kind classification problem. As such, the convolutional neural network (CNN) is introduced here to do classification and then to detect the fish’s posture. Before training the network, all sample images are preprocessed to make the fish be horizontal on the image according to the principal component analysis. Meanwhile, the histogram equalization is used to make the grey distribution of different images be close. After that, two kinds of strategies are taken to do classification. The first is a paired binary classification CNN and the second is a four-category CNN. In addition, three kinds of CNN are adopted. By comparison, the four-kind classification can obtain better results with error less than 1/1000.http://dx.doi.org/10.1155/2021/9939688 |
| spellingShingle | Xin Li Anzi Ding Shaojie Mei Wenjin Wu Wenguang Hou Convolutional Neural Network-Based Fish Posture Classification Complexity |
| title | Convolutional Neural Network-Based Fish Posture Classification |
| title_full | Convolutional Neural Network-Based Fish Posture Classification |
| title_fullStr | Convolutional Neural Network-Based Fish Posture Classification |
| title_full_unstemmed | Convolutional Neural Network-Based Fish Posture Classification |
| title_short | Convolutional Neural Network-Based Fish Posture Classification |
| title_sort | convolutional neural network based fish posture classification |
| url | http://dx.doi.org/10.1155/2021/9939688 |
| work_keys_str_mv | AT xinli convolutionalneuralnetworkbasedfishpostureclassification AT anziding convolutionalneuralnetworkbasedfishpostureclassification AT shaojiemei convolutionalneuralnetworkbasedfishpostureclassification AT wenjinwu convolutionalneuralnetworkbasedfishpostureclassification AT wenguanghou convolutionalneuralnetworkbasedfishpostureclassification |