Big Map R-CNN for object detection in large-scale remote sensing images

Abstract

Detecting sparse and multi-sized objects in very high resolution (VHR) remote sensing images remains a significant challenge in satellite imagery applications and analytics. Difficulties include broad geographical scene distributions and high pixel counts in each image: a large-scale satellite image contains tens to hundreds of millions of pixels and dozens of complex backgrounds. Furthermore, the scale of the same category object can vary widely (e.g., ships can measure from several to thousands of pixels). To address these issues, here we propose the Big Map R-CNN method to improve object detection in VHR satellite imagery. Big Map R-CNN introduces mean shift clustering for quadric detecting based on the existing Mask R-CNN architecture. Big Map R-CNN considers four main aspects: 1) big map cropping to generate small size sub-images; 2) detecting these sub-images using the typical Mask R-CNN network; 3) screening out fragmented low-confidence targets and collecting uncertain image regions by clustering; 4) quadric detecting to generate prediction boxes. We also introduce a new large-scale and VHR remote sensing imagery dataset containing two categories (RSI LS-VHR-2) for detection performance verification. Comprehensive evaluations on RSI LS-VHR-2 dataset demonstrate the effectiveness of the proposed Big Map R-CNN algorithm for object detection in large-scale remote sensing images.

Mask R‐CNN‐based building extraction from VHR satellite data in operational humanitarian action: An example related to Covid‐19 response in Khartoum, Sudan

“…Sampling focuses therefore on target objects only; no background or negative samples are needed, which is an advantage over other approaches in a time‐critical, operational setting. Mask R‐CNN has been successfully applied in similar satellite image analysis tasks, for example for sparse and multi‐sized object detection in VHR images (Wang, Tao, Wang, Wang, & Li, 2019), building extraction (Wen et al., 2019) or within the DeepGlobe Building Extraction Challenge (Zhao, Kang, Jung, & Sohn, 2018).…”

Self-compositional data augmentation for scene text detection

“…Current data augmentation for STD can be classified into two categories. The first category is to use image-level data augmentation, such as using random rotation, random flipping, random clipping, and color or contrast jitter on images [2][3] [4]. However, such a manner requires changing the content of the whole image and belongs to image-level variation; it ignores the possible usefulness of the instance-level variations.…”

A Contextual Bidirectional Enhancement Method for Remote Sensing Image Object Detection

“…[Available]: http://jiong.tea.ac.cn/people/JunweiHan/NWPUVHR 10dataset.html2 Online. [Available]: https://github.com/RSIA-LIESMARS-WHU/RSOD-Dataset-3 Online. [Available]: http://www.escience.cn/people/gongcheng/DIOR.html University.…”