Deep learning in the last decade has been very successful in computer vision and machine learning applications. Deep learning networks provide state-of-the-art performance in almost all of the applications where they have been employed. In this review, we aim to summarize the essential deep learning techniques and then apply them to COVID-19, a highly contagious viral infection that wreaks havoc on everyone’s lives in various ways. According to the World Health Organization and scientists, more testing potentially helps contain the virus’s spread. The use of chest radiographs is one of the early screening tests for determining disease, as the infection affects the lungs severely. To detect the COVID-19 infection, this experimental survey investigates and automates the process of testing by employing state-of-the-art deep learning classifiers. Moreover, the viruses are of many types, such as influenza, hepatitis, and COVID. Here, our focus is on COVID-19. Therefore, we employ binary classification, where one class is COVID-19 while the other viral infection types are treated as non-COVID-19 in the radiographs. The classification task is challenging due to the limited number of scans available for COVID-19 and the minute variations in the viral infections. We aim to employ current state-of-the-art CNN architectures, compare their results, and determine whether deep learning algorithms can handle the crisis appropriately and accurately. We train and evaluate 34 models. We also provide the limitations and future direction.

Fine-grained categorization is an essential field in classification, a subfield of object recognition that aims to differentiate subordinate classes. Fine-grained image classification concentrates on distinguishing between similar, hard-to-differentiate types or species, for example, flowers, birds, or specific animals such as dogs or cats, and identifying airplane makes or models. An important step towards fine-grained classification is the acquisition of datasets and baselines; hence, we propose a holistic system and two novel datasets, including reef fish and butterflies, for fine-grained classification. The butterflies and fish can be imaged at various locations in the image plane; thus, causing image variations due to translation, rotation, and deformation in multiple directions can induce variations, and depending on the image acquisition device’s position, scales can be different. We evaluate the traditional algorithms based on quantized rotation and scale-invariant local image features and the convolutional neural networks (CNN) using their pre-trained models to extract features. The comprehensive evaluation shows that the CNN features calculated using the pre-trained models outperform the rest of the image representations. The proposed system can prove instrumental for various purposes, such as education, conservation, and scientific research. The codes, models, and dataset are publicly available.