Analysis of Pruned Neural Networks (MobileNetV2-YOLO v2) for Underwater Object Detection

Ayob, Ahmad Faisal Mohamad; Khairuddin, Khairul Farhah; Mustafah, Yasir Mohd; Salisa, A. R.; Kadir, Khairul Azmi Abd

doi:10.1007/978-981-15-5281-6_7

Cited by 11 publications

(7 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the value is not much different from the other models. Adding layers or creating a more sophisticated architecture does not always result in increased prediction accuracy [ 54 ], and here, there was only a slight performance rise with more layers. So, we can say that performance of methods is not always related to the complexity of the network.…”

Section: Resultsmentioning

confidence: 99%

Improved Analysis of COVID-19 Influenced Pneumonia from the Chest X-Rays Using Fine-Tuned Residual Networks

Ksibi

Zakariah

Ayadi

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

COVID-19 has remained a threat to world life despite a recent reduction in cases. There is still a possibility that the virus will evolve and become more contagious. If such a situation occurs, the resulting calamity will be worse than in the past if we act irresponsibly. COVID-19 must be widely screened and recognized early to avert a global epidemic. Positive individuals should be quarantined immediately, as this is the only effective way to prevent a global tragedy that has occurred previously. No positive case should go unrecognized. However, current COVID-19 detection procedures require a significant amount of time during human examination based on genetic and imaging techniques. Apart from RT-PCR and antigen-based tests, CXR and CT imaging techniques aid in the rapid and cost-effective identification of COVID. However, discriminating between diseased and normal X-rays is a time-consuming and challenging task requiring an expert’s skill. In such a case, the only solution was an automatic diagnosis strategy for identifying COVID-19 instances from chest X-ray images. This article utilized a deep convolutional neural network, ResNet, which has been demonstrated to be the most effective for image classification. The present model is trained using pretrained ResNet on ImageNet weights. The versions of ResNet34, ResNet50, and ResNet101 were implemented and validated against the dataset. With a more extensive network, the accuracy appeared to improve. Nonetheless, our objective was to balance accuracy and training time on a larger dataset. By comparing the prediction outcomes of the three models, we concluded that ResNet34 is a more likely candidate for COVID-19 detection from chest X-rays. The highest accuracy level reached 98.34%, which was higher than the accuracy achieved by other state-of-the-art approaches examined in earlier studies. Subsequent analysis indicated that the incorrect predictions occurred with approximately 100% certainty. This uncovered a severe weakness in CNN, particularly in the medical area, where critical decisions are made. However, this can be addressed further in a future study by developing a modified model to incorporate uncertainty into the predictions, allowing medical personnel to manually review the incorrect predictions.

show abstract

Section: Resultsmentioning

confidence: 99%

Improved Analysis of COVID-19 Influenced Pneumonia from the Chest X-Rays Using Fine-Tuned Residual Networks

Ksibi

Zakariah

Ayadi

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…This result shows that the performance of methods is related to the complexity of the network. The additional layers or more complex architecture will not ensure higher accuracy of prediction (Ayob et al, 2019).…”

Section: Resultsmentioning

confidence: 99%

Convolutional Neural Network Architectures Performance Evaluation for Fish Species Classification

ISMAIL¹,

Ayob²,

Muslim³

et al. 2021

JSSM

View full text Add to dashboard Cite

Fish image classification tool is important in the field of ichthyology. In this paper, we present a fish image classification benchmark comparison across different types of convolutional neural network (CNN). CNN extracts features from labeled image data to solve classification problems. CNN models were trained to classify fish images using transfer learning with data augmentation. CNN models consisting of AlexNet, GoogLeNet, and ResNet were incorporated in the benchmark tests. A dataset of 18,000 fish images across 18 categores, were split into 5,400 images for validation (30%) and 12,600 images for training (70%) dataset. Such neural network models show high accuracy up to 99.85% (AlexNet), 96.39% (GoogLeNet) and 99.51% (ResNet-50). To evaluate the performance of each framework, the analysis presented consists of classification accuracy, learning curve, validation test, top-five prediction and confusion matrix. The work presented here has shown its potential to contribute towards accurate development of state-of-the-art fish classification tools. It is envisioned that these CNN algorithms have the potential to assist in fish image classification problems with high accuracy despite visually similar features of images in the dataset.

show abstract

“…This coincides with their problem, which is similarly greatly constrained: apple flowers have a very limited amount of intra-class variance, and the background is very similar in all dataset images. Another example is the work of Ayob et al [ 19 ], combining depth-wise separable convolutions and pruning for underwater object detection from a static aquarium camera, which resulted in a reduction with a factor of 161 in model size and a speed-up with a factor of 4.7. Although the exact numbers are not comparable with our results, where we even combine more optimization techniques, these studies show indeed that for constrained problems great optimization factors can be achieved.…”

Section: Discussionmentioning

confidence: 99%

“…Wu et al [ 18 ] looked at pruning single-shot object detectors for the operational case of apple flower detection, but they only researched the pruning optimization technique on a single dataset. Ayob et al [ 19 ] investigated the combination of depth-wise separable convolutions and pruning for underwater object detection.…”

Section: Introductionmentioning

confidence: 99%

Investigating the Potential of Network Optimization for a Constrained Object Detection Problem

et al. 2021

View full text Add to dashboard Cite

Object detection models are usually trained and evaluated on highly complicated, challenging academic datasets, which results in deep networks requiring lots of computations. However, a lot of operational use-cases consist of more constrained situations: they have a limited number of classes to be detected, less intra-class variance, less lighting and background variance, constrained or even fixed camera viewpoints, etc. In these cases, we hypothesize that smaller networks could be used without deteriorating the accuracy. However, there are multiple reasons why this does not happen in practice. Firstly, overparameterized networks tend to learn better, and secondly, transfer learning is usually used to reduce the necessary amount of training data. In this paper, we investigate how much we can reduce the computational complexity of a standard object detection network in such constrained object detection problems. As a case study, we focus on a well-known single-shot object detector, YoloV2, and combine three different techniques to reduce the computational complexity of the model without reducing its accuracy on our target dataset. To investigate the influence of the problem complexity, we compare two datasets: a prototypical academic (Pascal VOC) and a real-life operational (LWIR person detection) dataset. The three optimization steps we exploited are: swapping all the convolutions for depth-wise separable convolutions, perform pruning and use weight quantization. The results of our case study indeed substantiate our hypothesis that the more constrained a problem is, the more the network can be optimized. On the constrained operational dataset, combining these optimization techniques allowed us to reduce the computational complexity with a factor of 349, as compared to only a factor 9.8 on the academic dataset. When running a benchmark on an Nvidia Jetson AGX Xavier, our fastest model runs more than 15 times faster than the original YoloV2 model, whilst increasing the accuracy by 5% Average Precision (AP).

show abstract

Analysis of Pruned Neural Networks (MobileNetV2-YOLO v2) for Underwater Object Detection

Cited by 11 publications

References 11 publications

Improved Analysis of COVID-19 Influenced Pneumonia from the Chest X-Rays Using Fine-Tuned Residual Networks

Improved Analysis of COVID-19 Influenced Pneumonia from the Chest X-Rays Using Fine-Tuned Residual Networks

Convolutional Neural Network Architectures Performance Evaluation for Fish Species Classification

Investigating the Potential of Network Optimization for a Constrained Object Detection Problem

Contact Info

Product

Resources

About