SUR-Net: A Deep Network for Fish Detection and Segmentation With Limited Training Data

Lin, Huei-Yung; Tseng, Shih-Lun; Li, Junyi

doi:10.1109/jsen.2022.3195365

Cited by 14 publications

(3 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the recent progress in computer vision technology , object detection methods based on deep learning have been rapidly developed and widely used for fish detection and analysis (Yang et al, 2021;Li J. et al, 2022). Sun et al (2021) proposed a DRN-Faster R-CNN-based multi-target fish detection model for complex backgrounds with limited generalization capability, which reduces missed and false detections.…”

Section: Species Detection Methods Based On Deep Learning Informationmentioning

confidence: 99%

RA-UNet: an intelligent fish phenotype segmentation method based on ResNet50 and atrous spatial pyramid pooling

Li,

Liu,

Yang

et al. 2023

Front. Environ. Sci.

View full text Add to dashboard Cite

Introduction: Changes in fish phenotypes during aquaculture must be monitored to improve the quality of fishery resources. Therefore, a method for segmenting and measuring phenotypes rapidly and accurately without harming the fish is essential. This study proposes an intelligent fish phenotype segmentation method based on the residual network, ResNet50, and atrous spatial pyramid pooling (ASPP).Methods: A sufficient number of fish phenotypic segmentation datasets rich in experimental research was constructed, and diverse semantic segmentation datasets were developed. ResNet50 was then built as the backbone feature extraction network to prevent the loss of fish phenotypic feature information and improve the precision of fish phenotypic segmentation. Finally, an ASPP module was designed to improve the phenotypic segmentation accuracy of different parts of fish.Results: The test algorithm based on the collected fish phenotype segmentation datasets showed that the proposed algorithm (RA-UNet) yielded the best results among several advanced semantic segmentation models. The mean intersection over union (mIoU) and mean pixel accuracy (mPA) were 87.8% and 92.3%, respectively.Discussion: Compared with the benchmark UNet algorithm, RA-UNet demonstrated improvements in the mIoU and mPA by 5.0 and 1.8 percentage points, respectively. Additionally, RA-UNet exhibited superior fish phenotype segmentation performance, with a low false detection rate and clear and complete edge segmentation. Conclusively, the RA-UNet proposed in this study has high accuracy and edge segmentation ability and can, therefore, directly improve the efficiency of phenotypic monitoring in fish farming.

show abstract

Section: Species Detection Methods Based On Deep Learning Informationmentioning

confidence: 99%

RA-UNet: an intelligent fish phenotype segmentation method based on ResNet50 and atrous spatial pyramid pooling

Li,

Liu,

Yang

et al. 2023

Front. Environ. Sci.

View full text Add to dashboard Cite

show abstract

“…The obtained features are then input to the ConVNNs for voice recognition and, therefore, detect the corresponding mammals. The authors in [82]…”

Section: A Underwater Animals/moving Objects Detectionmentioning

confidence: 99%

“…Training the network with the voices of the mammals is cumbersome as the required dataset has limited existence 97.8 [82] Performs fish detection and segmentation with limited training data by combining various convolutional layers and residual blocks Does not require the usually large datasets required for training the model Addition of more layers adds complexity to the implementation 95.04 [83] Uses reflection from the moving fish target and processes them with ConVNNs by ignoring the clutter reflection for detection. Also proposed a recurrent neural network version for online processing with low detection accuracy Real-time moving target detection In deep water, detecting moving object becomes challenging x [84] Detects fish in sonar images using improved faster R-CoNVNNs and FPN with distance over union intersection method Accuracy and speedy target detection High computational complexity x [85] Detects jellyfish in underwater images by first adjusting the brightness and contrast of the images and then using Resnet50 with the backbone of the faster R-ConVNNs with improved training speed Accurate detection of jellyfish Complexity of operations due to a number of algorithms involvement x [86] Processes time and frequency-time spectra of underwater objects through the convolutional neural networks to recognize them Better features detection than the time spectra of the target High computational processing is involved 98.29 [87] A mechanism is designed for divers and underwater intruders detection that obtains a background image using deep learning and its difference with the current image using plan position indication…”

Section: Effective For Recognizing and Detecting Mammal Speciesmentioning

confidence: 99%

Underwater Target Detection Using Deep Learning: Methodologies, Challenges, Applications, and Future Evolution

Khan,

Fouda,

et al. 2024

IEEE Access

View full text Add to dashboard Cite

This paper provides a study of the latest target (object) detection algorithms for underwater wireless sensor networks (UWSNs). To ensure selection of the latest and state-of-the-art algorithms, only algorithms developed in the last seven years are taken into account that are not entirely addressed by the existing surveys. These algorithms are classified based on their architecture and methodologies of operation and their applications are described that are helpful in their selection in a diverse set of applications. The merits and demerits of the algorithms are also addressed that are helpful to improve their performance in future investigation. Moreover, a comparative analysis of the described algorithms is also given that further provides an insight to their selection in various applications and future enhancement. A depiction of the addressed algorithms in various applications based on publication count over the latest decade (2023-2013) is also given using the IEEE database that is helpful in knowing their future application trend. Finally, the challenges associated with the underwater target detection are highlighted and the future research paradigms are identified. The conducted study is helpful in providing a thorough analysis of the underwater target detection algorithms, their feasibility in various applications with future challenges and defined strategies for further investigation. INDEX TERMSUnderwater target detection, deep learning, underwater object detection, YOLO, convolutional neural networks, ConVNNs.I. INTRODUCTION Underwater wireless sensor networks (UWSNs) is one of the latest realms of research that aims to explore the underwater environment for a number of applications. These applications include Tsunami detection and prediction [1], military surveillance [2], underwater navigation [3], secure communications [4], oil detection [5], fault monitoring in underwater cables [6], water quality monitoring [7], detection of crashed ships

show abstract

A novel approach for underwater fish segmentation in complex scenes based on multi-levels triangular atrous convolution

Yang,

Li,

Zhao

2024

Aquacult Int

View full text Add to dashboard Cite

SUR-Net: A Deep Network for Fish Detection and Segmentation With Limited Training Data

Cited by 14 publications

References 40 publications

RA-UNet: an intelligent fish phenotype segmentation method based on ResNet50 and atrous spatial pyramid pooling

RA-UNet: an intelligent fish phenotype segmentation method based on ResNet50 and atrous spatial pyramid pooling

Underwater Target Detection Using Deep Learning: Methodologies, Challenges, Applications, and Future Evolution

A novel approach for underwater fish segmentation in complex scenes based on multi-levels triangular atrous convolution

Contact Info

Product

Resources

About