Deep Learning Approaches for Detecting Forgery Attacks on Digital Images
R Sucith, P. Mathiazhagan, M Ajith, G Lingesh · 2025
Malaria is a life-threatening disease that has affected millions across the world, and quick and accurate detection is essential for proper management. This paper presents an innovative automated technique to detect malaria parasites using digital image processing techniques. The proposed method, integrating preprocessing, segmentation, feature extraction, and classification in a unified framework, is implemented using the publicly available Cell Images Dataset. Images are normalized and denoised, and then Otsu’s thresholding and morphological operations are applied for the exact segmentation of cells. Texture, shape, and color features of segmented cells are extracted; and a Support Vector Machine classifier with an RBF kernel is used to classify them as infected or uninfected. The proposed method achieves very high classification accuracy of 97.3%, with robust precision and recall rates, thus exhibiting the effectiveness of the proposed algorithm under different imaging conditions. This automated system presents a reliable, efficient, and scalable solution for malaria detection, especially in resource-limited settings, which could reduce diagnostic workloads and improve patient outcomes. Real-time deployment and multi-stage malaria diagnosis will be explored in future work.