Classification of Beef, Goat, and Pork using GLCM Texture-Based Backpropagation Neural Network
Identifying different types of meat is crucial for preventing fraudulent activities and improving food safety. This research aims to create a classification system for various meat types (beef, goat, and pork) using the Gray Level Co-occurrence Matrix (GLCM) for extracting texture features, followed...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | Indonesian |
| Published: |
Islamic University of Indragiri
2024-11-01
|
| Series: | Sistemasi: Jurnal Sistem Informasi |
| Online Access: | https://sistemasi.ftik.unisi.ac.id/index.php/stmsi/article/view/4715 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Identifying different types of meat is crucial for preventing fraudulent activities and improving food safety. This research aims to create a classification system for various meat types (beef, goat, and pork) using the Gray Level Co-occurrence Matrix (GLCM) for extracting texture features, followed by classification through a Backpropagation Neural Network (BPNN). The methodology utilizes 60 images of beef, goat, and pork, achieving a remarkable accuracy of 100% in the training phase, which highlights the model's capability to effectively recognize patterns. However, when tested with new data, the system exhibits a sensitivity of 90% and a specificity of 95%, with some misclassifications occurring between goat and beef due to their similar textures. The findings of this study suggest that GLCM is an effective tool for deriving relevant statistical parameters necessary for classification. This research makes a significant contribution to developing a meat identification system that safeguards consumers and promotes awareness of food safety issues. The results are anticipated to provide a solid foundation for advancing meat type recognition and practical applications in the marketplace, ultimately boosting public trust in the meat products they purchase. |
|---|---|
| ISSN: | 2302-8149 2540-9719 |