Accurate fish detection is of great significance in aquaculture. However, the non-uniform strong reflection in aquaculture ponds will affect the precision of fish detection. This paper combines YOLOv4 and CVAE to accurately detect fishes in the image with non-uniform strong reflection, in which the reflection in the image is removed at first and then the reflection-removed image is provided for fish detecting. Firstly, the improved YOLOv4 is applied to detect and mask the strong reflective region, to locate and label the reflective region for the subsequent reflection removal. Then, CVAE is combined with the improved YOLOv4 for inferring the priori distribution of the Reflection region and restoring the Reflection region by the distribution so that the reflection can be removed. For further improving the quality of the reflection-removed images, the adversarial learning is appended to CVAE. Finally, YOLOV4 is used to detect fishes in the high quality image. In addition, a new image dataset of pond cultured takifugu rubripes is constructed,, which includes 1000 images with fishes annotated manually, also a synthetic dataset including 2000 images with strong reflection is created and merged with the generated dataset for training and verifying the robustness of the proposed method. Comprehensive experiments are performed to compare the proposed method with the state-of-the-art fish detecting methods without reflection removal on the generated dataset. The results show that the fish detecting precision and recall of the proposed method are improved by 2.7% and 2.4% respectively.
Meng ZHAO
Dalian Ocean University,Ministry of Education,Key Laboratory of Marine Information Technology of Liaoning Province
Junfeng WU
Dalian Ocean University,Ministry of Education,Key Laboratory of Marine Information Technology of Liaoning Province
Hong YU
Dalian Ocean University,Ministry of Education,Key Laboratory of Marine Information Technology of Liaoning Province
Haiqing LI
Dalian Ocean University,Ministry of Education,Key Laboratory of Marine Information Technology of Liaoning Province
Jingwen XU
Dalian Ocean University,Ministry of Education,Key Laboratory of Marine Information Technology of Liaoning Province
Siqi CHENG
Dalian Ocean University,Ministry of Education,Key Laboratory of Marine Information Technology of Liaoning Province
Lishuai GU
Dalian Ocean University,Ministry of Education,Key Laboratory of Marine Information Technology of Liaoning Province
Juan MENG
Dalian Ocean University,Ministry of Education,Key Laboratory of Marine Information Technology of Liaoning Province
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Meng ZHAO, Junfeng WU, Hong YU, Haiqing LI, Jingwen XU, Siqi CHENG, Lishuai GU, Juan MENG, "Fish Detecting Using YOLOv4 and CVAE in Aquaculture Ponds with a Non-Uniform Strong Reflection Background" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 5, pp. 715-725, May 2023, doi: 10.1587/transinf.2022DLK0001.
Abstract: Accurate fish detection is of great significance in aquaculture. However, the non-uniform strong reflection in aquaculture ponds will affect the precision of fish detection. This paper combines YOLOv4 and CVAE to accurately detect fishes in the image with non-uniform strong reflection, in which the reflection in the image is removed at first and then the reflection-removed image is provided for fish detecting. Firstly, the improved YOLOv4 is applied to detect and mask the strong reflective region, to locate and label the reflective region for the subsequent reflection removal. Then, CVAE is combined with the improved YOLOv4 for inferring the priori distribution of the Reflection region and restoring the Reflection region by the distribution so that the reflection can be removed. For further improving the quality of the reflection-removed images, the adversarial learning is appended to CVAE. Finally, YOLOV4 is used to detect fishes in the high quality image. In addition, a new image dataset of pond cultured takifugu rubripes is constructed,, which includes 1000 images with fishes annotated manually, also a synthetic dataset including 2000 images with strong reflection is created and merged with the generated dataset for training and verifying the robustness of the proposed method. Comprehensive experiments are performed to compare the proposed method with the state-of-the-art fish detecting methods without reflection removal on the generated dataset. The results show that the fish detecting precision and recall of the proposed method are improved by 2.7% and 2.4% respectively.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022DLK0001/_p
Copy
@ARTICLE{e106-d_5_715,
author={Meng ZHAO, Junfeng WU, Hong YU, Haiqing LI, Jingwen XU, Siqi CHENG, Lishuai GU, Juan MENG, },
journal={IEICE TRANSACTIONS on Information},
title={Fish Detecting Using YOLOv4 and CVAE in Aquaculture Ponds with a Non-Uniform Strong Reflection Background},
year={2023},
volume={E106-D},
number={5},
pages={715-725},
abstract={Accurate fish detection is of great significance in aquaculture. However, the non-uniform strong reflection in aquaculture ponds will affect the precision of fish detection. This paper combines YOLOv4 and CVAE to accurately detect fishes in the image with non-uniform strong reflection, in which the reflection in the image is removed at first and then the reflection-removed image is provided for fish detecting. Firstly, the improved YOLOv4 is applied to detect and mask the strong reflective region, to locate and label the reflective region for the subsequent reflection removal. Then, CVAE is combined with the improved YOLOv4 for inferring the priori distribution of the Reflection region and restoring the Reflection region by the distribution so that the reflection can be removed. For further improving the quality of the reflection-removed images, the adversarial learning is appended to CVAE. Finally, YOLOV4 is used to detect fishes in the high quality image. In addition, a new image dataset of pond cultured takifugu rubripes is constructed,, which includes 1000 images with fishes annotated manually, also a synthetic dataset including 2000 images with strong reflection is created and merged with the generated dataset for training and verifying the robustness of the proposed method. Comprehensive experiments are performed to compare the proposed method with the state-of-the-art fish detecting methods without reflection removal on the generated dataset. The results show that the fish detecting precision and recall of the proposed method are improved by 2.7% and 2.4% respectively.},
keywords={},
doi={10.1587/transinf.2022DLK0001},
ISSN={1745-1361},
month={May},}
Copy
TY - JOUR
TI - Fish Detecting Using YOLOv4 and CVAE in Aquaculture Ponds with a Non-Uniform Strong Reflection Background
T2 - IEICE TRANSACTIONS on Information
SP - 715
EP - 725
AU - Meng ZHAO
AU - Junfeng WU
AU - Hong YU
AU - Haiqing LI
AU - Jingwen XU
AU - Siqi CHENG
AU - Lishuai GU
AU - Juan MENG
PY - 2023
DO - 10.1587/transinf.2022DLK0001
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2023
AB - Accurate fish detection is of great significance in aquaculture. However, the non-uniform strong reflection in aquaculture ponds will affect the precision of fish detection. This paper combines YOLOv4 and CVAE to accurately detect fishes in the image with non-uniform strong reflection, in which the reflection in the image is removed at first and then the reflection-removed image is provided for fish detecting. Firstly, the improved YOLOv4 is applied to detect and mask the strong reflective region, to locate and label the reflective region for the subsequent reflection removal. Then, CVAE is combined with the improved YOLOv4 for inferring the priori distribution of the Reflection region and restoring the Reflection region by the distribution so that the reflection can be removed. For further improving the quality of the reflection-removed images, the adversarial learning is appended to CVAE. Finally, YOLOV4 is used to detect fishes in the high quality image. In addition, a new image dataset of pond cultured takifugu rubripes is constructed,, which includes 1000 images with fishes annotated manually, also a synthetic dataset including 2000 images with strong reflection is created and merged with the generated dataset for training and verifying the robustness of the proposed method. Comprehensive experiments are performed to compare the proposed method with the state-of-the-art fish detecting methods without reflection removal on the generated dataset. The results show that the fish detecting precision and recall of the proposed method are improved by 2.7% and 2.4% respectively.
ER -