An advanced Bangladeshi local fish classification system based on the combination of deep learning and the internet of things (IoT)

Ahmed, Md. Asif; Hossain, Md. Shakil; Rahman, Wahidur; Uddin, Abdul Hasib; Islam, Md. Tarequl

doi:10.1016/j.jafr.2023.100663

Cited by 9 publications

(2 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, it is possible to evaluate multiple times (k-fold) the generalization error of the classifiers on data not seen during the training stage [46]. This method also allows for a more detailed evaluation of the goodness-of-fit as opposed to a single evaluation, as when using the holdout technique (e.g., 66% for training and 34% for testing) [47]. The performance metrics selected for studying the classification results were Accuracy (ACC), Precision (PRE), Recall (REC), and F1-value (F1).…”

Section: Performance Evaluationmentioning

confidence: 99%

A Spatial-Spectral Classification Method Based on Deep Learning for Controlling Pelagic Fish Landings in Chile

Pezoa,

Ramírez,

Godoy

et al. 2023

Sensors

View full text Add to dashboard Cite

Fishing has provided mankind with a protein-rich source of food and labor, allowing for the development of an important industry, which has led to the overexploitation of most targeted fish species. The sustainable management of these natural resources requires effective control of fish landings and, therefore, an accurate calculation of fishing quotas. This work proposes a deep learning-based spatial-spectral method to classify five pelagic species of interest for the Chilean fishing industry, including the targeted Engraulis ringens, Merluccius gayi, and Strangomera bentincki and non-targeted Normanichthtys crockeri and Stromateus stellatus fish species. This proof-of-concept method is composed of two channels of a convolutional neural network (CNN) architecture that processes the Red–Green–Blue (RGB) images and the visible and near-infrared (VIS-NIR) reflectance spectra of each species. The classification results of the CNN model achieved over 94% in all performance metrics, outperforming other state-of-the-art techniques. These results support the potential use of the proposed method to automatically monitor fish landings and, therefore, ensure compliance with the established fishing quotas.

show abstract

Section: Performance Evaluationmentioning

confidence: 99%

A Spatial-Spectral Classification Method Based on Deep Learning for Controlling Pelagic Fish Landings in Chile

Pezoa,

Ramírez,

Godoy

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…By carefully evaluating the efficacy of deep learning algorithms in raising energy efficiency, raising manufacturing quality, and streamlining operational procedures, this empirical study seeks to close this gap. This study tackles the knowledge gap and provides firms aiming to adopt Industry 5.0 with practical insights by means of empirical data [11]- [15].…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Algorithms in Industry 5.0: A Comprehensive Experimental Study

Shchepkina,

Chandramauli,

Ahuja

et al. 2024

BIO Web Conf.

View full text Add to dashboard Cite

This extensive experimental research provides strong empirical proof of the revolutionary power of deep learning algorithms when integrated into Industry 5.0. Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), Generative Adversarial Networks (GAN), and Transformers are a few examples of deep learning algorithms that have shown remarkable accuracy rates of 92.3%, 88.7%, and 95.1%, respectively. Furthermore, the processing durations, which vary between 15 and 25 milliseconds, confirm their ability to make decisions in real time. The abundance of various data accessible in Industry 5.0 is highlighted by data collection sources such as picture databases (300 GB), text corpora (150 GB), equipment records (250 GB), and IoT sensor data (500 GB). The significant energy savings, shown by 20% reductions across a range of machine types, highlight the financial and ecological advantages of deep learning integration. Moreover, the noteworthy improvements in production quality, exhibiting up to 50% reductions in defect rates, highlight the potential of deep learning in quality assurance. These results provide tangible proof of the critical roles deep learning algorithms play in streamlining production lines, increasing energy economy, and boosting product quality in the ever-changing Industry 5.0 environment.

show abstract

Marine Resources: Identification, Restoring, and Monitoring of Fisheries Food Resources Using Deep Learning and Image Processing

Ahamed,

Ayesha

2024

The Springer Series in Applied Machine Learning

View full text Add to dashboard Cite

An advanced Bangladeshi local fish classification system based on the combination of deep learning and the internet of things (IoT)

Cited by 9 publications

References 22 publications

A Spatial-Spectral Classification Method Based on Deep Learning for Controlling Pelagic Fish Landings in Chile

A Spatial-Spectral Classification Method Based on Deep Learning for Controlling Pelagic Fish Landings in Chile

Deep Learning Algorithms in Industry 5.0: A Comprehensive Experimental Study

Marine Resources: Identification, Restoring, and Monitoring of Fisheries Food Resources Using Deep Learning and Image Processing

Contact Info

Product

Resources

About