Wound Depth Measurement System in Forensic Case using Image Processing and Machine Learning DOI Creative Commons

Elvira Sukma Wahyuni,

Kern Cesarean Ahnaf,

Firdaus Firdaus

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 25, 2024

Abstract Accurate evaluation of wound depth is vital for forensic investigations, as it can greatly influence case assessments and outcomes. This research presents a new method classifying using Support Vector Machine (SVM) model, comparing its efficacy against Decision Tree Logistic Regression models by utilizing color-based features derived from HSV LAB color spaces. The dataset included 76 images, divided into three categories: stage 2 (36 images), 3 (12 4 (28 images). Performance was measured confusion matrices, well F1-Score, precision, recall. SVM model attained an overall accuracy 85%, showing high precision recall across all stages, in contrast to the models, which achieved 50% 70%, respectively. results illustrated strong performance especially distinguishing wounds, while highlighting difficulties differentiating between stages compared other approaches. Furthermore, ROC curves statistical tests, including paired t-tests, validated significant improvements over baseline methods. indicate that proposed system could improve efficiency evaluations offering quick objective classification tool. Future should aim broaden dataset, integrate additional features, investigate machine learning enhance system's robustness applicability various scenarios.

Language: Английский

State Incentives for Solar Energy in the Context of Energy Transition in Developed and Developing Countries DOI Creative Commons
Minh Phuong Nguyen, Tatiana Ponomarenko

Energies, Journal Year: 2025, Volume and Issue: 18(5), P. 1227 - 1227

Published: March 3, 2025

The world economy is developing rapidly, and many countries in the are facing pressure of energy demand due to economic development. Vietnam one that has advantages solar development, so it plans develop renewable structure balance. article analyzes policies a number have significantly developed field (USA, Germany, Japan) or some similarities with (China, Thailand, Australia). Based on results analysis, we recommendations for strategy sustainable development sector Vietnam.

Language: Английский

Citations

0
Wound Depth Measurement System in Forensic Case using Image Processing and Machine Learning DOI Creative Commons

Elvira Sukma Wahyuni,

Kern Cesarean Ahnaf,

Firdaus Firdaus

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 25, 2024

Abstract Accurate evaluation of wound depth is vital for forensic investigations, as it can greatly influence case assessments and outcomes. This research presents a new method classifying using Support Vector Machine (SVM) model, comparing its efficacy against Decision Tree Logistic Regression models by utilizing color-based features derived from HSV LAB color spaces. The dataset included 76 images, divided into three categories: stage 2 (36 images), 3 (12 4 (28 images). Performance was measured confusion matrices, well F1-Score, precision, recall. SVM model attained an overall accuracy 85%, showing high precision recall across all stages, in contrast to the models, which achieved 50% 70%, respectively. results illustrated strong performance especially distinguishing wounds, while highlighting difficulties differentiating between stages compared other approaches. Furthermore, ROC curves statistical tests, including paired t-tests, validated significant improvements over baseline methods. indicate that proposed system could improve efficiency evaluations offering quick objective classification tool. Future should aim broaden dataset, integrate additional features, investigate machine learning enhance system's robustness applicability various scenarios.

Language: Английский

Citations

0