Synthetic SAR/IR database generation for sensor fusion-based A.T.R.

Won, Jin-Ju; Kim, Sungho; Cho, Youngrea; Song, Woo‐Jin; Kim, So-Hyun

doi:10.1109/urai.2015.7358893

Cited by 6 publications

(3 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The SAR/IR database used in the experiment consists of images of 16 different target classes. This database contains a total 4,608 images per sensor, and the size of each image is 64×64 ( The SAR/IR images are generated by using the SE-RAY-SAR and SE-RAY-IR simulators, which can generate very realistic images via the ray-tracing technique [42]- [46]. For training and evaluation, we separated the SAR/IR database depending on the depression angle ( Table 2).…”

Section: Experiments a Databasementioning

confidence: 99%

Multistage Fusion With Dissimilarity Regularization for SAR/IR Target Recognition

et al. 2019

Self Cite

View full text Add to dashboard Cite

In this paper, we propose dissimilarity regularization with a multistage fusion stream for a synthetic aperture radar (SAR) and infrared (IR) sensor fusion using deep learning. The multistage fusion structures are composed of multiple layers for fusing all the feature maps generated by the convolutional neural networks. The proposed structure combines feature maps of equivalent levels, ensuring that the spatial information of the corresponding levels can be utilized for fusion. Dissimilarity regularization is the sum of the normalized cross-correlation between the features generated in two different single-sensor streams. The proposed regularization is added to the conventional learning problem of a single-sensor stream, and each single-sensor stream is promoted to learn the disparate types of features for fusion. To evaluate the proposed algorithm, we compare the recognition rate of the proposed algorithm with that of the conventional fusion approaches using the SAR and IR image databases. Finally, the effects of the proposed architecture and regularization on the fusion result are analyzed.

show abstract

Section: Experiments a Databasementioning

confidence: 99%

Multistage Fusion With Dissimilarity Regularization for SAR/IR Target Recognition

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Adaption of this approach to buildings with complex shapes, e.g., round shape. [7] J. Won and S. Kim et al: The SAR-IR sensor fusion research has been in progress for effective ground target detection.…”

Section: Literature Surveymentioning

confidence: 99%

A Review on Automatic Target Recognition and Detection Approaches for SAR Images

Kushwah¹

2019

IJRASET

View full text Add to dashboard Cite

This paper present a comparative or survey study of object detection or target recognition techniques for synthetic aperture radar images. As know, SAR is an artificial aperture microwave radar or remote sensing equipment which has capability to capture the scene of object by millions of kilometer or where human beings not supposed to reached, generally SAR images known as satellites imaginary. Here in this paper discusses the different techniques and approaches for the target recognition and object detection for SAR imaginary and find the common problem face by the researchers during implementation of such kind of artificial intelligence. Basically ATR i.e. automatic target recognition such as oil spills, missing air-bourn, ships and object identification or recognition at polar surface where human being not supposed to be present or reached has an interested area for the any researchers. In this paper discusses the various techniques such as prescreening method, CFAR, neural networks algorithms, and supervised classifier and many other methods and find the optimal solution or method for fast automatic target recognition and object detection according of their geometry and size. Finally a unique discussion has to be done in this paper and concluded the paper topic.

show abstract

“…Infrared target dataset was prepared using a synthesis tool, OKTAL-SE [18,19]. OKTAL-SE can synthesize a desired image using user's desired parameters such as season, weather, time, temperature, atmosphere environment, and thermal noise.…”

Section: Infrared Target Datasetmentioning

confidence: 99%

Rotation Invariant Automatic Infrared Target Recognition using G-Radon

Won

Kim

2016

MATEC Web of Conferences

Self Cite

View full text Add to dashboard Cite

Abstract. In recent research, automatic target recognition (ATR) of infrared targets has been taking a lot of interest to the researchers. A rotation invariant method is useful in target recognition, classification and image analysis to reduce the number of training data. In particular, rotation invariant method, Radon transform, is an effective technique that is used for medical care such as computerized tomography (CT) image. This paper proposes a new rotation invariant algorithm for target recognition. The proposed method combines the gradient information and radon transform. The propose method, called gradient Radon (G-Radon), is applied to synthesized infrared images and compared with traditional radon transform and Zernike moments for validation.

show abstract

Synthetic SAR/IR database generation for sensor fusion-based A.T.R.

Cited by 6 publications

References 4 publications

Multistage Fusion With Dissimilarity Regularization for SAR/IR Target Recognition

Multistage Fusion With Dissimilarity Regularization for SAR/IR Target Recognition

A Review on Automatic Target Recognition and Detection Approaches for SAR Images

Rotation Invariant Automatic Infrared Target Recognition using G-Radon

Contact Info

Product

Resources

About