Image Super-Resolution Through Compressive Sensing-based Recovery.

The primary aim of image super-resolution techniques is to produce a high resolution (HR) image from a low resolution (LR) image efficiently. Deep learning algorithms are being extensively used to address the ill-posed problem of single image super-resolution which requires extremely large data-sets and high processing power. When one does not have access to large data-sets or have limited processing power, an alternative technique may be in order. In this study, we have developed a novel positive scale image resizing method inspired by compressive sensing (CS). We have considered the image super-resolution as a CS recovery problem in which a low resolution image is assumed as a compressed measurement and the required interpolated image is treated as output of the CS-based recovery. In the proposed HR recovery method, a deterministic binary block diagonal measurement matrix, (DBBD), is used as measurement matrix since it maintains the visual similarity between the low and high resolution images. Then along with a sparsification matrix, the sparse representation of HR image is first recovered and subsequently the dense HR image is obtained. The proposed method is applied to medical and non-medical images. The HR images obtained using the traditional proximal, bilinear and bi-cubic interpolation techniques are compared with those obtained using the proposed method. The proposed CS inspired method delivers superior HR images than the traditional techniques. The superiority of the proposed method is attributed to the unique usage of the DBBD matrix and the CS recovery algorithm to obtain a high resolution image without any prior training data-set.

A new preprocessing approach to improve the performance of CNN-based skin lesion classification.

Skin lesion is one of the severe diseases which in many cases endanger the lives of patients on a worldwide extent. Early detection of disease in dermoscopy images can significantly increase the survival rate. However, the accurate detection of disease is highly challenging due to the following reasons: e.g., visual similarity between different classes of disease (e.g., melanoma and non-melanoma lesions), low contrast between lesions and skin, background noise, and artifacts. Machine learning models based on convolutional neural networks (CNN) have been widely used for automatic recognition of lesion diseases with high accuracy in comparison to conventional machine learning methods. In this research, we proposed a new preprocessing technique in order to extract the region of interest (RoI) of skin lesion dataset. We compare the performance of the most state-of-the-art CNN classifiers with two datasets which contain (1) raw, and (2) RoI extracted images. Our experiment results show that training CNN models by RoI extracted dataset can improve the accuracy of the prediction (e.g., InceptionResNetV2, 2.18% improvement). Moreover, it significantly decreases the evaluation (inference) and training time of classifiers as well.