Categorizing Low Resolution Aerial Photographs by Hessian-Regularized Perceptual Feature Selection

• Published in: IEEE access 2024-04, p.1-1
• Main Authors: Wang, Guifeng, Xiao, Jianzhang, Yang, Yi
• Format: Article
• Language: English
• Subjects: active feature; Feature extraction; Image reconstruction; Image resolution; low-resolution; manifold regularizer; perception learning; Satellites; selecting features; Semantics; Solid modeling; Visualization
• ISSN: 2169-3536
• DOI: 10.1109/ACCESS.2024.3384836

There are plenty of high- and low-altitude earth observation satellites asynchronously capture massive-scale aerial images everyday. In practice, high-altitude satellites take low-resolution (LR) aerial pictures, each covers a considerably large area. Comparatively, low-altitude satellites capture high-resolution (HR) aerial photos, each depicts a relatively small area. Effectively identifying LR aerial images' semantic categories is an indispensable module in many AI systems. However, it is also a challenging task due to: 1) the inefficiency to label adequate training samples, and 2) the difficulty to describe how humans preserving the world. To handle these problems, this work presents a so-called active perception learning coupled with a Hessian-regularized feature selection (HRFS), aiming at acquiring perceptual and discriminative visual representation to classify LR aerial photos. Particularly, by stimulating how humans sequentially perceiving different salient regions, we deploy an active learning paradigm to divide an LR aerial image into a few attractive regions as well as a rich set of non-attractive regions. Theoretically, the active learning technique ensures that the selected attractive regions can maximally reconstruct each LR aerial image, which well mimics human visual perception. Subsequently, a novel HRFS is designed to select high quality features from the actively detected attractive regions. HRFS has several competitive attributes: 1) it conducts feature selection in a semi-supervised mode; 2) a linear classifier can be jointly trained; 3) sample geometric distribution in the feature space can be maximally maintained. Empirical results showed the robustness of our proposed classification pipeline.