Nature

MedNet: a lightweight attention-augmented CNN for medical image classification

We propose MedNet, a lightweight and efficient CNN architecture that integrates depthwise separable convolutions and CBAM to improve classification accuracy on medical image datasets. We design a ...
Nature

Multi-resolution transfer learning for tampered image classification using SE-enhanced fused-MBConv and optimized CNN heads

Recognition of forged images is a critical task in fields such as forensics and cybersecurity. Conventional techniques often struggle to accurately identify tampered images due to complex artifacts ...