期刊
REMOTE SENSING
卷 13, 期 10, 页码 -出版社
MDPI
DOI: 10.3390/rs13101921
关键词
oriented object detection; optical remote sensing images; multi-sector; anchor-free; classification-to-regression
类别
资金
- National Natural Science Foundation of China [61701524]
- China Postdoctoral Science Foundation [2019M653742]
The paper proposes a novel multi-sector oriented object detection framework called MSO2-Det, which quantizes the scales and orientation prediction of targets in optical remote sensing images through an anchor-free classification-to-regression approach. It achieves more accurate localization information by dividing the scales and angle space into multiple discrete sectors and using a coarse-granularity classification to fine-grained regression strategy. To decrease angular-sector classification loss and accelerate network convergence, a smooth angular-sector label (SASL) and a localization-aided detection score (LADS) are designed.
Oriented object detection in optical remote sensing images (ORSIs) is a challenging task since the targets in ORSIs are displayed in an arbitrarily oriented manner and on small scales, and are densely packed. Current state-of-the-art oriented object detection models used in ORSIs primarily evolved from anchor-based and direct regression-based detection paradigms. Nevertheless, they still encounter a design difficulty from handcrafted anchor definitions and learning complexities in direct localization regression. To tackle these issues, in this paper, we proposed a novel multi-sector oriented object detection framework called MSO2-Det, which quantizes the scales and orientation prediction of targets in ORSIs via an anchor-free classification-to-regression approach. Specifically, we first represented the arbitrarily oriented bounding box as four scale offsets and angles in four quadrant sectors of the corresponding Cartesian coordinate system. Then, we divided the scales and angle space into multiple discrete sectors and obtained more accurate localization information by a coarse-granularity classification to fine-grained regression strategy. In addition, to decrease the angular-sector classification loss and accelerate the network's convergence, we designed a smooth angular-sector label (SASL) that smoothly distributes label values with a definite tolerance radius. Finally, we proposed a localization-aided detection score (LADS) to better represent the confidence of a detected box by combining the category-classification score and the sector-selection score. The proposed MSO2-Det achieves state-of-the-art results on three widely used benchmarks, including the DOTA, HRSC2016, and UCAS-AOD data sets.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据