A Novel Method of Aircraft Detection under Complex Background Based on Circular Intensity Filter and Rotation Invariant Feature.

Xin Chen,Lihua Cao,Yujia Zuo,Fang Xu,Zhihua Xie,Jinghong Liu

doi:10.3390/s22010319

Abstract

Aircraft detection in remote sensing images (RSIs) has drawn widespread attention in recent years, which has been widely used in the military and civilian fields. While the complex background, variations of aircraft pose and size bring great difficulties to the effective detection. In this paper, we propose a novel aircraft target detection scheme based on small training samples. The scheme is coarse-to-fine, which consists of two main stages: region proposal and target identification. First, in the region proposal stage, a circular intensity filter, which is designed based on the characteristics of the aircraft target, can quickly locate the centers of multi-scale suspicious aircraft targets in the RSIs pyramid. Then the target regions can be extracted by adding bounding boxes. This step can get high-quality but few candidate regions. Second, in the stage of target identification, we proposed a novel rotation-invariant feature, which combines rotation-invariant histogram of oriented gradient and vector of locally aggregated descriptors (VLAD). The feature can characterize the aircraft target well by avoiding the impact of its rotation and can be effectively used to remove false alarms. Experiments are conducted on Remote Sensing Object Detection (RSOD) dataset to compare the proposed method with other advanced methods. The results show that the proposed method can quickly and accurately detect aircraft targets in RSIs and achieve a better performance.

Highlights

With the rapid development of image sensing technique [1] and aerospace technology, the acquisition of remote sensing images (RSIs) has become more convenient
Candidate regions can be generated by adding specific bounding boxes
One is RICNN [39] that is based on Convolutional Neural Networks (CNN) model and rotation invariant analysis, and the oobtafhsfeeardiisronnYOeHsLOsO,Gvo2aun[4dr0]pF.oTruoorpvieeorrisHfeyOdtGherfeeefgafteiucortneofaptrhereocFoponoudrsuiacetrleHdmOteoGtm-hVaoLkdAeDaisfceouamtsuperader,itsihonensm.tFeeotahrdothdoesf the sliding windows msakeethofofdairunessesd, opurrpervoipoousesdlyr.egFioonrptrhoepopsaalrmaemtheotdeirs suesetdtiinngstseaodfotfhtheesselidailnggowritnh- ms, please refer to the odroiwgsimnaetlhcoidtuesdedliptreervaiotuusrlye..FTorhtheePpaRracmuertevressettoinfgtshoef thdeisfefealrgeonritthmmse,tphleoadsesreofenr the Remote Sensing Object Detection (RSOD) dataset are staohretohswehoonwrigniinninaFlFciiiggtueudrerlei1t0e1,raa0tnu,draeTn.aTdbhleeTP3aRcboclruerrev3sepscooonfdrtirhneegsdlypiflofiesnrtsednthtinemgqeutlhyaondltiistsaottnisvtethhreeeRsuSqOltusDaindnattetiartmsaesttive results in terms of AofPAsPsaanndd mmeeaannrurnuninnng itnimgest.imes

Summary

Introduction

With the rapid development of image sensing technique [1] and aerospace technology, the acquisition of RSIs has become more convenient. RSIs contain a large amount of useful information, so it is important to fully extract and utilize the information. Owing to the important applications in dynamic airport surveillance and military reconnaissance, aircraft detection in RSIs has attracted much attention [2]. For civilian use, effective detection of aircraft targets can improve the utilization of airports while providing guidance on parking areas for aircraft to be landed. Unlike the common natural images, target detection in RSIs has the following specificities: more complex geographical environmental information, variations of target poses and sizes. All these factors contribute to the degradation of detection algorithm performance.

Methods

Results

Discussion

Conclusion