Classification of Coronavirus Images using Shrunken Features

Saban Ozturk; Umut Ozkaya; Mucahid Barstugan

In addition, interest in hand-crafted methods has begun to decline with the introduction of CNN and other automated feature extraction techniques. Convolutional neural network architecture is a deep learning architecture that automatically extracts and classifies images from images [15] . Özyurt et al. [16] proposed a hybrid model called fused perceptual hash-based CNN to reduce the classifying time of liver CT images and maintain performance. Xu et al. [17] used a transfer learning strategy to deal with the medical image unbalance problem. Then, they compared GoogleNet, ResNet101, Xception, and MobileNetv2 performances. Lakshmanaprabu et al. [18] analyzed the CT scan of lung images using the assistance of optimal deep neural network and linear discriminate analysis. Gao et al. [19] converted raw CT images to low attenuation, raw images, and high attenuation pattern rescale. Then, they resampled these three samples and classified them using CNN.