Class-Aware Fish Species Recognition Using Deep Learning for an Imbalanced Dataset

Alaba, Simegnew; Nabi, M M; Shah, Chiranjibi; Prior, Jack; Campbell, Matthew D.; Wallace, Farron; Ball, John; Moorhead, Robert

doi:10.3390/s22218268

Cited by 41 publications

(29 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Computationally intelligent and energy-efficient data sharing among various onboard sensors need an advanced optimization framework to minimize the total transmit power of the vehicle-to-everything (V2X) networks [ 112 ]. Besides, sensor data can be corrupted by different noise models [ 113 ], so the noisy data need to be removed for better detection [ 114 ]. YOLO with adaptive frame control has been used for real-time object detection in AI-embedded systems [ 115 ].…”

Section: Discussionmentioning

confidence: 99%

Off-Road Detection Analysis for Autonomous Ground Vehicles: A Review

Islam¹,

Nabi²,

Ball³

2022

Sensors

Self Cite

View full text Add to dashboard Cite

When it comes to some essential abilities of autonomous ground vehicles (AGV), detection is one of them. In order to safely navigate through any known or unknown environment, AGV must be able to detect important elements on the path. Detection is applicable both on-road and off-road, but they are much different in each environment. The key elements of any environment that AGV must identify are the drivable pathway and whether there are any obstacles around it. Many works have been published focusing on different detection components in various ways. In this paper, a survey of the most recent advancements in AGV detection methods that are intended specifically for the off-road environment has been presented. For this, we divided the literature into three major groups: drivable ground and positive and negative obstacles. Each detection portion has been further divided into multiple categories based on the technology used, for example, single sensor-based, multiple sensor-based, and how the data has been analyzed. Furthermore, it has added critical findings in detection technology, challenges associated with detection and off-road environment, and possible future directions. Authors believe this work will help the reader in finding literature who are doing similar works.

show abstract

Section: Discussionmentioning

confidence: 99%

Off-Road Detection Analysis for Autonomous Ground Vehicles: A Review

Islam¹,

Nabi²,

Ball³

2022

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…Additionally, the dataset has a class-imbalance problem which means some classes have a higher number of samples and some classes have fewer samples. In order to solve the problem, Alaba et al 4 proposed a class-aware loss method that takes into account the inverse of the number of samples in each class to solve the class imbalance problem. They also incorporate the class-balanced loss into the object detection problem to re-weight the classification and localization losses.…”

Section: Related Workmentioning

confidence: 99%

MI-AFR: multiple instance active learning-based approach for fish species recognition in underwater environments

Shah

Alaba

Nabi³

et al. 2023

Ocean Sensing and Monitoring XV

Self Cite

View full text Add to dashboard Cite

Video surveys are commonly used to monitor the abundance and distribution of managed species to support management. However, considerable effort, time, and cost are required for human review and automated fish species recognition provides an effective solution to remove the bottleneck of post-processing. Implementing fish species detection techniques for underwater imagery is a challenging task. In this work, we present the Multiple Instance Active-learning for Fish-species Recognition (MI-AFR), which is formulated as an object detection-based approach to perform localization and classification of fish species. It can select the most informative fish images from unlabeled sets by estimating the uncertainty of unlabeled images by using adversarial classifiers trained on labeled sets. Moreover, we have analyzed the improved performance of MI-AFR by considering different backbone networks as a trade-off between speed and accuracy. For experiments, we have used the fine-grained and large-scale reef fish dataset obtained from the Gulf of Mexico – the Southeast Area Monitoring and Assessment Program Dataset 2021 (SEAMAPD21). The experimental results illustrate that the superiority of the proposed method can establish a solid foundation for active learning in fish species recognition, especially with a small number of labeled sets.

show abstract

“…In addition to the classification task, object detection works, such as [2], [3], [4], [5], [6], [7], [8], [9] used the convolutional network to get better performance over traditional machine learning. Various regularization and optimization techniques are used for the training.…”

Section: Introductionmentioning

confidence: 99%

Image Classification using Different Machine Learning Techniques

Alaba¹

2023

Preprint

View full text Add to dashboard Cite

<p>Artificial Neural Networks and Convolutional Neural Networks are becoming common tools for classification and object detection tasks due to their power to learn features without prior knowledge. The networks learn the parameters, weights, and biases through training. This paper proposes a simple Neural Network and Convolutional Neural Network (CNN) to do a classification task. Additionally, the Bayesian neural network work is reproduced to compare the result to my proposed networks and used as a baseline for comparison. The MNIST dataset is used for all experiments. </p> <p>The simple neural networks and the convolutional networks adjust the parameters based on the cost function during training, whereas the Bayesian convolutional neural network updates the parameters based on the backdrop that drives a variational approximation to the true posterior. I trained the two proposed networks by varying the hyperparameters: optimizer, learning rate, regularizers, dropout, epochs, and others. My proposed work gives better classification accuracy, approximately 99 \%, than the previously implemented Bayesian convolutional neural network, but it is difficult to predict how certain the prediction is, which is easy for Bayesian learning. </p>

show abstract

Class-Aware Fish Species Recognition Using Deep Learning for an Imbalanced Dataset

Cited by 41 publications

References 53 publications

Off-Road Detection Analysis for Autonomous Ground Vehicles: A Review

Off-Road Detection Analysis for Autonomous Ground Vehicles: A Review

MI-AFR: multiple instance active learning-based approach for fish species recognition in underwater environments

Image Classification using Different Machine Learning Techniques

Contact Info

Product

Resources

About