An Improved Deep Learning Model for Underwater Species Recognition in Aquaculture
The ability to differentiate between various fish species plays an essential role in aquaculture. It helps to protect their populations and monitor their health situations and their nutrient systems. However, old machine learning methods are unable to detect objects in images with complex background...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-10-01
|
| Series: | Fishes |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2410-3888/8/10/514 |
| _version_ | 1797573904890331136 |
|---|---|
| author | Mahdi Hamzaoui Mohamed Ould-Elhassen Aoueileyine Lamia Romdhani Ridha Bouallegue |
| author_facet | Mahdi Hamzaoui Mohamed Ould-Elhassen Aoueileyine Lamia Romdhani Ridha Bouallegue |
| author_sort | Mahdi Hamzaoui |
| collection | DOAJ |
| description | The ability to differentiate between various fish species plays an essential role in aquaculture. It helps to protect their populations and monitor their health situations and their nutrient systems. However, old machine learning methods are unable to detect objects in images with complex backgrounds and especially in low-light conditions. This paper aims to improve the performance of a YOLO v5 model for fish recognition and classification. In the context of transfer learning, our improved model FishDETECT uses the pre-trained FishMask model. Then it is tested in various complex scenes. The experimental results show that FishDETECT is more effective than a simple YOLO v5 model. Using the evaluation metrics Precision, Recall, and mAP50, our new model achieved accuracy rates of 0.962, 0.978, and 0.995, respectively. |
| first_indexed | 2024-03-10T21:15:40Z |
| format | Article |
| id | doaj.art-5d487619beb548699672bb2d029ccd27 |
| institution | Directory Open Access Journal |
| issn | 2410-3888 |
| language | English |
| last_indexed | 2024-03-10T21:15:40Z |
| publishDate | 2023-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Fishes |
| spelling | doaj.art-5d487619beb548699672bb2d029ccd272023-11-19T16:28:02ZengMDPI AGFishes2410-38882023-10-0181051410.3390/fishes8100514An Improved Deep Learning Model for Underwater Species Recognition in AquacultureMahdi Hamzaoui0Mohamed Ould-Elhassen Aoueileyine1Lamia Romdhani2Ridha Bouallegue3Innov’COM Laboratory, Higher School of Communication of Tunis (SUPCOM), Technopark Elghazala, Raoued, Ariana 2083, TunisiaInnov’COM Laboratory, Higher School of Communication of Tunis (SUPCOM), Technopark Elghazala, Raoued, Ariana 2083, TunisiaCore Curriculum Program, Deanship of General Studies, University of Qatar, Doha P.O. Box 2713, QatarInnov’COM Laboratory, Higher School of Communication of Tunis (SUPCOM), Technopark Elghazala, Raoued, Ariana 2083, TunisiaThe ability to differentiate between various fish species plays an essential role in aquaculture. It helps to protect their populations and monitor their health situations and their nutrient systems. However, old machine learning methods are unable to detect objects in images with complex backgrounds and especially in low-light conditions. This paper aims to improve the performance of a YOLO v5 model for fish recognition and classification. In the context of transfer learning, our improved model FishDETECT uses the pre-trained FishMask model. Then it is tested in various complex scenes. The experimental results show that FishDETECT is more effective than a simple YOLO v5 model. Using the evaluation metrics Precision, Recall, and mAP50, our new model achieved accuracy rates of 0.962, 0.978, and 0.995, respectively.https://www.mdpi.com/2410-3888/8/10/514aquaculturefish speciescomputer visiondeep learningtransfer learning |
| spellingShingle | Mahdi Hamzaoui Mohamed Ould-Elhassen Aoueileyine Lamia Romdhani Ridha Bouallegue An Improved Deep Learning Model for Underwater Species Recognition in Aquaculture Fishes aquaculture fish species computer vision deep learning transfer learning |
| title | An Improved Deep Learning Model for Underwater Species Recognition in Aquaculture |
| title_full | An Improved Deep Learning Model for Underwater Species Recognition in Aquaculture |
| title_fullStr | An Improved Deep Learning Model for Underwater Species Recognition in Aquaculture |
| title_full_unstemmed | An Improved Deep Learning Model for Underwater Species Recognition in Aquaculture |
| title_short | An Improved Deep Learning Model for Underwater Species Recognition in Aquaculture |
| title_sort | improved deep learning model for underwater species recognition in aquaculture |
| topic | aquaculture fish species computer vision deep learning transfer learning |
| url | https://www.mdpi.com/2410-3888/8/10/514 |
| work_keys_str_mv | AT mahdihamzaoui animproveddeeplearningmodelforunderwaterspeciesrecognitioninaquaculture AT mohamedouldelhassenaoueileyine animproveddeeplearningmodelforunderwaterspeciesrecognitioninaquaculture AT lamiaromdhani animproveddeeplearningmodelforunderwaterspeciesrecognitioninaquaculture AT ridhabouallegue animproveddeeplearningmodelforunderwaterspeciesrecognitioninaquaculture AT mahdihamzaoui improveddeeplearningmodelforunderwaterspeciesrecognitioninaquaculture AT mohamedouldelhassenaoueileyine improveddeeplearningmodelforunderwaterspeciesrecognitioninaquaculture AT lamiaromdhani improveddeeplearningmodelforunderwaterspeciesrecognitioninaquaculture AT ridhabouallegue improveddeeplearningmodelforunderwaterspeciesrecognitioninaquaculture |