EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Development of Machine Learning Frameworks for Identifying Lumpy Skin Disease

Olusola Bamidele Ayoade, Mumini Oyetunji Raji, Aminat Adejoke Akindele, Kemi Jemilat Yusuf-Mashopa, Muinat Folake Abdulrauff, Ibrahim Adebayo Raji, Fatima Bolanle Musah, Abiodun Timothy Adegbiji, Oluwafemi Micheal Amuda

Shodh Sari-An International Multidisciplinary Journal · 30.10.2026 · 5 (4) · 654

DOI: 10.59231/sari8002 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Shodh Sari-An International Multidisciplinary Journal, online erschienen am 30.10.2026. Verfasst von Olusola Bamidele Ayoade und 8 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Lumpy skin disease (LSD), a viral disease affecting cattle, can cause substantial health and production losses, thereby creating significant economic challenges for livestock producers and the wider cattle industry. Prompt and accurate diagnosis is essential for reducing disease transmission and supporting timely veterinary intervention. This study investigated the effectiveness of optimisation-based support vector machine (SVM) models for automated detection of LSD using a dataset of 3,000 cattle skin images, consisting of 1,500 LSD-infected images and 1,500 healthy skin images obtained from Kaggle village datasets. The image-processing pipeline began with spatial normalisation of the input images, followed by the removal of unwanted hair structures through Black-Hat morphological filtering and image inpainting, together with noise suppression. The resulting images were then subjected to Sobel-based edge analysis to delineate the regions of affected skin. A combination of texture, geometric, and chromatic features was subsequently generated using Gray Level Spatial Dependence (GLSD) matrices, region-based shape descriptors, and Colour Moment descriptors, respectively. For classification, Support Vector Machine (SVM) models optimised using the Aquila Optimisation (AO) Algorithm and Harris Hawks Optimisation (HHO) were implemented as AO-SVM and HHO-SVM, respectively. Their classification performance was assessed using false positive rate (FPR), specificity, accuracy, precision, and sensitivity. Both models successfully differentiated infected cattle from healthy animals. However, the HHO-SVM model produced superior performance, achieving an accuracy of 95.95% compared with 95.50% for AO-SVM, representing a 0.45% improvement in classification accuracy. The results demonstrate that optimisation strategies can enhance the predictive capability of SVM classifiers for LSD detection, with HHO-SVM providing the most reliable performance. The proposed approach offers a practical decision-support tool for early disease diagnosis and could strengthen veterinary surveillance, improve livestock health management, and contribute to more effective control of lumpy skin disease. Keywords: Aquila Optimisation, Cattle Skin Images, Harris Hawk Optimisation Algorithm, Image Processing, Metaheuristic Optimisation

Zitieren

APA

Ayoade, O. B., Raji, M. O., Akindele, A. A., Yusuf-Mashopa, K. J., Abdulrauff, M. F., Raji, I. A., Musah, F. B., Adegbiji, A. T., & Amuda, O. M. (2026). Development of Machine Learning Frameworks for Identifying Lumpy Skin Disease. Shodh Sari-An International Multidisciplinary Journal, 5(4), 654. https://doi.org/10.59231/sari8002

Vancouver

Ayoade OB, Raji MO, Akindele AA, Yusuf-Mashopa KJ, Abdulrauff MF, Raji IA, et al. Development of Machine Learning Frameworks for Identifying Lumpy Skin Disease. Shodh Sari-An International Multidisciplinary Journal. 2026;5(4):654. doi:10.59231/sari8002

BibTeX
@article{Ayoade2026,
  author = {Ayoade, Olusola Bamidele and Raji, Mumini Oyetunji and Akindele, Aminat Adejoke and Yusuf-Mashopa, Kemi Jemilat and Abdulrauff, Muinat Folake and Raji, Ibrahim Adebayo and Musah, Fatima Bolanle and Adegbiji, Abiodun Timothy and Amuda, Oluwafemi Micheal},
  title = {Development of Machine Learning Frameworks for Identifying Lumpy Skin Disease},
  journal = {Shodh Sari-An International Multidisciplinary Journal},
  year = {2026},
  volume = {5},
  number = {4},
  pages = {654},
  doi = {10.59231/sari8002}
}
Kostenlos starten →