Accès libre

Research on deep learning image segmentation method based on attention mechanism

Haibo Li

Haibo Li

School of Information Technology, Jiangsu Open UniversityNanjing, China
Recherchez cet auteur sur
Sciendo|Google Scholar
Li, Haibo
  
17 mars 2025
Research on deep learning image segmentation method based on attention mechanism's Cover Image
À propos de cet article
Article précédent
Article suivant
Citez
Télécharger la couverture
Publié en ligne: 17 mars 2025
Reçu: 24 oct. 2024
Accepté: 12 févr. 2025
DOI: https://doi.org/10.2478/amns-2025-0210
Mots clés
© 2025 Haibo Li, published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1.

The loss function parameters compare the results of the experiment
The loss function parameters compare the results of the experiment

Figure 2.

Training curves of different addition strategies
Training curves of different addition strategies

Figure 3.

The loss curve of the different addition strategies on the test' set
The loss curve of the different addition strategies on the test' set

Figure 4.

The segmentation performance of each model on the PH2 test set
The segmentation performance of each model on the PH2 test set

Figure 5.

Segmentation performance in the ISIC 2018 validation set
Segmentation performance in the ISIC 2018 validation set

Data set information

Name Format Label type Training Data partitioning validation Testing
ISIC 2017 800 350 0
PH2 jpg Pixel level 0 0 220
ISIC 2018 1865 635 0

The hardware and software environment of the experiment

Central processor Intel(R)Core i7-8600K CPU@3.60 GHz
Graphics card Nvidia TITAN Xp 24GB
Memory 64GB
Operating system Windows 10
Python 3.9.15
CUDA 11.6
torch 1.13.0+cu116
torchvision 0.14.0+cu116
Simulation platform PyCharm

Ablation research results in DRIVE and the CHASEDB data set

LKM ETM MFM DRIVE CHASEDB
Acc Sen Sp AUC Acc Sen Sp AUC
✓ 0.925 0.836 0.972 0.966 0.958 0.875 0.975 0.982
✓ ✓ 0.964 0.841 0.966 0.973 0.944 0.869 0.963 0.991
✓ ✓ 0.952 0.869 0.979 0.987 0.957 0.873 0.967 0.985
✓ ✓ ✓ 0.973 0.824 0.983 0.989 0.970 0.889 0.986 0.992
Langue:
Anglais
Périodicité:
1 fois par an
Sujets de la revue:
Sciences de la vie, Sciences de la vie, autres, Mathématiques, Mathématiques appliquées, Mathématiques générales, Physique, Physique, autres
RSS Feed de la revue