Enhanced U-Net models with encoder and augmentation for phytoplankton segmentation
Open Access
- 1 December 2024
- journal article
- Published by Institute of Advanced Engineering and Science in International Journal of Advances in Applied Sciences
- Vol. 13 (4) , 1009-1018
Abstract
This study comprehensively analyzes U-Net models for semantic segmentation in phytoplankton image recognition, leveraging encoders such as EfficientNet-B5, MobileNetV2, ResNet50, and ResNeXt50 and employing the Adam optimizer. The research highlights the U-Net MobileNetV2 model with optical distortion, which achieves notable test scores with 93.69% Dice, 88.14% intersection over union (IoU), 99.89% Precision, and 100% Recall, underscoring the efficacy of the applied augmentation strategies, including geometric and distortion transforms, and color and blur techniques. The U-Net ResNet50 model with mix transform consistently demonstrates high accuracy in critical metrics, outperforming others, while EfficientNet-B5 with blur suggests increased model sensitivity with improved recall. These results underscore the crucial role of encoder-augmentation synergy in model performance. Training and testing times across models have remained under 250 seconds, reflecting methodological efficiency. Overall, these results demonstrate the model's excellent performance for the semantic segmentation task.Keywords
This publication has 43 references indexed in Scilit:
- Survey of automatic plankton image recognition: challenges, existing solutions and future perspectivesArtificial Intelligence Review, 2024
- Adaptive Learning Rate and Momentum for Training Deep Neural Networks2021
- Diatom Classification Including Morphological Adaptations Using CNNs2019
- Improving plankton image classification using context metadataLimnology and Oceanography: Methods, 2019
- Deep learning and transfer learning features for plankton classificationEcological Informatics, 2019
- Phytoplankton changes during SE monsoonal period in the Lembeh Strait of North Sulawesi, Indonesia, from 2012 to 2015Acta Oceanologica Sinica, 2018
- Classification of radiolarian images with hand-crafted and deep featuresComputers & Geosciences, 2017
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- LabelMe: A Database and Web-Based Tool for Image AnnotationInternational Journal of Computer Vision, 2007
- Conservation hotspots of biodiversity and endemism for Indo‐Pacific coral reef fishesAquatic Conservation: Marine and Freshwater Ecosystems, 2007