Along with developing deep learning methods, Convolutional Neural Networks (CNNs) have emerged as a powerful tool by providing both remarkable performances in image processing and the ability to work in a wide variety of applications in the vision community. In the past few years, biologically inspired CNNs have emerged and proven effective in the image processing field, from social media to precision medicine and robotics. A beneficial characteristic of CNNs is data processing in multiple arrays and automatic feature extraction ability, which have received acknowledgment in the geoscience and remote sensing community.
Moreover, the inherent characteristics of CNNs, such as local connectivity and weight sharing, allow this deep learning method to tackle the drawbacks of artificial feature extraction, by considering the 2-D structures and reducing network parameters using convolutional filters. CNN-based models have benefited from the recent exponential advances in imaging technologies, such as the availability of various image types (optical, RADAR, temperature and microwave radiometer, altimeter, etc.) with complex characteristics (high dimensionality, multiple scales, and nonstationary). CNNs are composed of a set of blocks that make them particularly suitable for image analysis. The multiple layers of operations, such as convolution, pooling, and nonlinear activation functions, allow for a hierarchical extraction of high-level abstract features. Accordingly, CNNs have been successfully used in image preprocessing, scene classification, pixel-based classification, image segmentation, and object detection. CNNs have been used in numerous studies, for instance: to improve image classification results to extract buildings and non-building regions automatically; to detect areas of build-up; to assess the quality of OpenStreetMap data; to detect oil spills, ships, and icebergs. Although CNNs can be considered newly introduced algorithms in geoscience and remote sensing, they are now clearly among the top performers in most of the applications.
Despite this progress, the study of CNN-based approaches in the field of remote sensing and geoscience is currently at its beginning stages, and there is still much potential for new developments. In this perspective, the design of new network architectures for specific tasks, the generation of large-scale datasets for network training, the integration of conventional techniques for various remote sensing data, the advancement and analysis of existing networks concerning their architectures, optimization techniques, and the regularization strategies are still open topics, which are in close relation with each other and should be jointly considered.