Oriented Object Detection in Remote Sensing Images with Anchor-Free Oriented Region Proposal Network

Jianxiang Li,Zili Zhang,Yiping Xu,Yan Tian

doi:10.3390/rs14051246

Jianxiang Li, Zili Zhang + Show 2 more

Open Access

https://doi.org/10.3390/rs14051246

Copy DOI

Journal: Remote Sensing	Publication Date: Mar 3, 2022
Citations: 7	License type: CC BY 4.0

Affiliation: Huazhong University of Science and Technology

Abstract

Oriented object detection is a fundamental and challenging task in remote sensing image analysis that has recently drawn much attention. Currently, mainstream oriented object detectors are based on densely placed predefined anchors. However, the high number of anchors aggravates the positive and negative sample imbalance problem, which may lead to duplicate detections or missed detections. To address the problem, this paper proposes a novel anchor-free two-stage oriented object detector. We propose the Anchor-Free Oriented Region Proposal Network (AFO-RPN) to generate high-quality oriented proposals without enormous predefined anchors. To deal with rotation problems, we also propose a new representation of an oriented box based on a polar coordinate system. To solve the severe appearance ambiguity problems faced by anchor-free methods, we use a Criss-Cross Attention Feature Pyramid Network (CCA-FPN) to exploit the contextual information of each pixel and its neighbors in order to enhance the feature representation. Extensive experiments on three public remote sensing benchmarks—DOTA, DIOR-R, and HRSC2016—demonstrate that our method can achieve very promising detection performance, with a mean average precision (mAP) of 80.68%, 67.15%, and 90.45%, respectively, on the benchmarks.

Highlights

2806 images ranging from 800 × 800 to 4000 × 4000 pixels and 188,282 instances labeled by arbitrarily oriented quadrilaterals over 15 categories: plane (PL), baseball diamond (BD), bridge (BR), ground track field (GTF), small vehicle (SV), large vehicle (LV), ship (SH), tennis court (TC), basketball court (BC), storage tank (ST), soccer-ball field (SBF), roundabout (RA), harbor (HA), swimming pool (SP), and helicopter (HC)
We used the training set for network training and the validation set for evaluation in the ablation experiments
In a comparison with state-of-the-art object detectors, the training set and validation set were both used for network training, and the corresponding results on the test set were submitted to the official evaluation server at https://captain-whu.github.io/

Summary

Introduction

Extensive experiments on three public remote sensing benchmarks—DOTA, DIOR-R, and HRSC2016—demonstrate that our method can achieve very promising detection performance, with a mean average precision (mAP) of. Traditional object detection methods [10], like object-based image analysis (OBIA) [11], usually take two steps to accomplish object detection: firstly, extract regions that may contain potential objects, extract hand-designed features and apply classifiers to obtain the class information. Their detection performance is unsatisfactory because the handcrafted features have limited representational power with insufficient semantic information.

Methods

Results

Discussion

Conclusion