Underwater Fish Detection and Counting Using Mask Regional Convolutional Neural Network

Khai, Teh Hong; Abdullah, Siti Norul Huda Sheikh; Hasan, Mohammad Kamrul; Tarmizi, Ahmad

doi:10.3390/w14020222

Cited by 32 publications

(10 citation statements)

References 29 publications

(43 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our method has outperformed [78,79] by both performance accuracy and number of species detected. The model shown in [80] could not identify different species which our proposed model has performed.…”

Section: Comparison With the State-of-the-artmentioning

confidence: 88%

Fish Species Identification on Low Resolution - A Study with Enhanced Super Resolution Generative Adversarial Network (ESRGAN), YOLO and VGG-16

Adhikary¹

2022

Preprint

View full text Add to dashboard Cite

A smart detection and recognition model for the species of a fish from camera footage is of urgent requirement as fishery contributes to a large portion of the world economy and these kinds of advanced models can aid fishermen on a large scale. Such models incorporating a pick-and-place machine can be very useful to sort different fish species in a bulk without human intervention and this can greatly reduce costs for large scale fishing industries. Existing methods for detecting and recognizing fish species have many limitations such as limited scalability, detection accuracy, failing to detect multiple species, degraded performance at a lower resolution, or pinpointing the exact location of the fish. By modifying the head of a very powerful deep learning model namely VGG-16 with pre-trained weights can be used to detect both the species of the fish and find the exact location of the fish in an image by implementing a modified YOLO to incorporate the bounding box regression head. We have proposed the usage of the ESRGAN algorithm along with the proposed neural network to amplify the image resolution by a factor of 4. With this method, the overall detection accuracy of 96.5% has been obtained. The experiment has been conducted based on a total of 9460 images spread across 9 species. After further improving the model, a pick-and-place machine could be integrated for very fast sorting of the fish according to their species in different large-scale fish industries.

show abstract

Section: Comparison With the State-of-the-artmentioning

confidence: 88%

Fish Species Identification on Low Resolution - A Study with Enhanced Super Resolution Generative Adversarial Network (ESRGAN), YOLO and VGG-16

Adhikary¹

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The two-stage weighted least square method for moving target localization, non-convex optimization problems, 21 tracking the multiple moving targets using standard joint probabilistic association (JPDA) statistical method, 22 semidefinite relaxation for non-line of sight between the sensors, 23 mask regional model for underwater fish. 24 The VIV effect on underwater sensors short on the literature survey shown in Table 1. The proposed CVV algorithm is quantify the VIV effect on sensors, low localization error, high localization accuracy compared to standard methods.…”

Section: Related Workmentioning

confidence: 99%

“…Crammer Rao lower bound (CRLB) traces the optimal deployment of anchor nodes in underwater, 20 minimizing the localization noise at higher noise conditions (21). The two‐stage weighted least square method for moving target localization, non‐convex optimization problems, 21 tracking the multiple moving targets using standard joint probabilistic association (JPDA) statistical method, 22 semidefinite relaxation for non‐line of sight between the sensors, 23 mask regional model for underwater fish 24 …”

Section: Introductionmentioning

confidence: 99%

Vortex‐induced vibration effects in underwater acoustic wireless sensor networks

Reddy,

Arya

2023

Trans Emerging Tel Tech

View full text Add to dashboard Cite

A randomly fluctuated vortex‐induced vibration (VIV) affects the localization of moving targets in an underwater acoustic wireless sensor network (UAWSN) to cause an offset time and time synchronization error which disturbs the localization of the non‐cooperative moving target. Definition 1, Theorem 1 has been introduced in the present work to describe the genesis and consequences of the VIV effect in UAWSN. An elliptic localization introduces the direct and indirect path measurements of the moving target. The standard Crammer Rao lower bound parameter quantifies the VIV effect. The compensation of vortex‐induced vibrations (CVV) algorithm is proposed to quantify the degradation of localization performance and to estimate the position of the moving target. The theoretical measurements of position error, velocity error with respect to the variation of noise, number of receiver sensors, and probability of line of sight errors are compared to the standard methods. The proposed CVV algorithm improves the localization performance by 32.24% and reduces 34.36% of position errors, and 37.46% of velocity errors up to 600 m. It can be useful for the sub‐aquatic internet of underwater things.

show abstract

“…In addition, four-neighborhood labeling [9], grayscale image analysis [10], image noise reduction and segmentation [11] are also employed to biological counting. Convolutional neural networks can greatly improve accuracy and generalization ability [12][13][14][15], which are widely used in the field of fish target detection [16][17][18][19][20][21]. An improved YOLOv3 model was proposed by Cui [22] to implement a counting system of puffer fish.…”

Section: Introductionmentioning

confidence: 99%

Bass detection model based on improved YOLOv5 in circulating water system

Deng

Cao

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

The feeding amount of bass farming is closely related to the number of bass. It is of great significance to master the number of bass to achieve accurate feeding and improve the economic benefits of the farm. In view of the interference caused by the problems of multiple targets and target occlusion in bass data for bass detection, this paper proposes a bass target detection model based on improved YOLOV5 in circulating water system. Firstly, acquiring by HD cameras, Mosaic-8, a data augmentation method, is utilized to expand datasets and improve the generalization ability of the model. And K-means clustering algorithm is applied to generate suitable coordinates of prior boxes to improve training efficiency. Secondly, Coordinate Attention mechanism (CA) is introduced into backbone feature extraction network and neck feature fusion network to enhance attention to targets of interest. Finally, Soft-NMS algorithm replaces Non-Maximum Suppression algorithm (NMS) to re-screen prediction boxes and keep targets with higher overlap, which effectively solves the problems of missed detection and false detection. The experiments show that the proposed model can reach 98.09% in detection accuracy and detection speed reaches 13.4ms. The proposed model can help bass farmers under the circulating water system to accurately grasp the number of bass, which has important application value to realize accurate feeding and water conservation.

show abstract

Underwater Fish Detection and Counting Using Mask Regional Convolutional Neural Network

Cited by 32 publications

References 29 publications

Fish Species Identification on Low Resolution - A Study with Enhanced Super Resolution Generative Adversarial Network (ESRGAN), YOLO and VGG-16

Fish Species Identification on Low Resolution - A Study with Enhanced Super Resolution Generative Adversarial Network (ESRGAN), YOLO and VGG-16

Vortex‐induced vibration effects in underwater acoustic wireless sensor networks

Bass detection model based on improved YOLOv5 in circulating water system

Contact Info

Product

Resources

About