Robust Ground Target Detection by SAR and IR Sensor Fusion Using Adaboost-Based Feature Selection.

Woo-Jin Song,So-Hyun Kim,Sungho Kim

doi:10.3390/s16071117

Abstract

Long-range ground targets are difficult to detect in a noisy cluttered environment using either synthetic aperture radar (SAR) images or infrared (IR) images. SAR-based detectors can provide a high detection rate with a high false alarm rate to background scatter noise. IR-based approaches can detect hot targets but are affected strongly by the weather conditions. This paper proposes a novel target detection method by decision-level SAR and IR fusion using an Adaboost-based machine learning scheme to achieve a high detection rate and low false alarm rate. The proposed method consists of individual detection, registration, and fusion architecture. This paper presents a single framework of a SAR and IR target detection method using modified Boolean map visual theory (modBMVT) and feature-selection based fusion. Previous methods applied different algorithms to detect SAR and IR targets because of the different physical image characteristics. One method that is optimized for IR target detection produces unsuccessful results in SAR target detection. This study examined the image characteristics and proposed a unified SAR and IR target detection method by inserting a median local average filter (MLAF, pre-filter) and an asymmetric morphological closing filter (AMCF, post-filter) into the BMVT. The original BMVT was optimized to detect small infrared targets. The proposed modBMVT can remove the thermal and scatter noise by the MLAF and detect extended targets by attaching the AMCF after the BMVT. Heterogeneous SAR and IR images were registered automatically using the proposed RANdom SAmple Region Consensus (RANSARC)-based homography optimization after a brute-force correspondence search using the detected target centers and regions. The final targets were detected by feature-selection based sensor fusion using Adaboost. The proposed method showed good SAR and IR target detection performance through feature selection-based decision fusion on a synthetic database generated by OKTAL-SE.

Highlights

Automatic target detection (ATD) is very important in military applications and there are challenging problems with ground surveillance [1]
The original method was modified by inserting a median local average filter and an asymmetric morphological closing filter to handle the simultaneous synthetic aperture radar (SAR) and IR target detection problem
This paper proposed a novel SAR/IR target detection by feature selection-based fusion from candidate targets generated using a modified Boolean map visual theory-based method

Summary

Introduction

Automatic target detection (ATD) is very important in military applications and there are challenging problems with ground surveillance [1]. A decision-level fusion scheme was adopted for SAR and IR-based target detection because independent processing can reduce the processing cost and less accurate sensor alignment is required. The Rectangles denotes the ground truths and the rounded edges denotes the detection results Based on these motivations, this paper proposes a single detection method to detect both SAR and IR targets simultaneously for a SAR and IR fusion study. The original method was modified by inserting a median local average filter and an asymmetric morphological closing filter (called modBMVT) to handle the simultaneous SAR and IR target detection problem Another contribution is the use of automatic SAR and IR image registration by RANdom SAmple.

Background of BMVT Theory and Its Limitations

Proposed Modified BMVT-Based Target Detection: modBMVT

Background

Iterate 1-3 until achieving maximum Region Consensus Score

Experimental Results

Conclusions and Discussion