Seung Ho Doo

Bio: Seung Ho Doo is an academic researcher from Ohio State University. The author has contributed to research in topics: Feature vector & Radar imaging. The author has an hindex of 4, co-authored 7 publications receiving 59 citations.

Target classification performance as a function of measurement uncertainty

Abstract: In this paper, we demonstrate target classification using the proposed features in previously reported research under measurement uncertainty conditions. The MSTAR dataset is widely used real target measurements in automatic target recognition society. Extremely high classification results of the dataset, which are over 90% correct classification, have been reported from some literatures. However, this high classification results could be acquired not only by the classification system, but also the cleanness of the dataset. Therefore, in this paper, more realistic target classification scenarios including target aspect angle estimation error, strong white Gaussian noise, and different combination of test and training targets are applied for classification and its corresponding results are examined. The proposed target feature extraction techniques show the robustness of the measurement uncertainties and excellent classification results.

Reliable target feature extraction and classification using potential target information

Abstract: A reliable target feature extraction process is proposed in this paper. The locations of dominant scatterers have been widely adopted as target features in automatic target recognition (ATR) systems. However, the direct use of the locations shows high variability and results in a negative effect on target classification performance. Here, we propose a novel grid cell structure that uses information regarding potential targets to be classified. The grid cell structure extracts stable features from SAR images with a relatively lower computational complexity. A novel target feature that uses information about the variability of scatterers is also proposed. Simulation results, using real target measurements taken from the MSTAR dataset, demonstrate that the new feature vectors improve classification performance.

Aspect invariant features for radar target recognition

Abstract: In this study, the authors introduce a novel method of target feature extraction that in one form attempts to reduce variability inherent in target signatures and in a second form attempts to exploit the variability. It is well known that radar target signatures can be subject to significant fluctuations as a function of viewing geometry and that this behaviour severely degrades the performance of automatic target recognition algorithms. Here, they propose one variability reduction technique which groups strong scatterers in target signatures. A second method exploits variability by using the difference between two reduced resolution images to extract target angular stability information. In both cases, a grid cell structure enables extraction of these relatively stable features from synthetic aperture radar images with low computational complexity. Results, using measured target data taken from the MSTAR dataset, demonstrate that the proposed method of selecting feature vectors significantly improve the overall classification performance. A correct classification rate of 96% is reached in testing.

Scatterer stability in high range resolution profiles

Abstract: Radar target classification typically uses feature sets derived from high range resolution profiles as an input to a classifier. However, such feature sets, when projected in feature space, show both substantial overlap for different targets and substantial variability for a single target. This results in degraded classification performance, as targets cannot be reliably distinguished from one another. In this paper we show that it is a combination of target scattering and signal processing that causes such variability. Specifically we demonstrate that the output of the matched filter is a sensitive function of the relative position of scatters distributed along the length of a target and how this effects the variability of particular features.

Point scatterer modeling of complex targets

Abstract: In this paper, a novel point scatterer model is introduced that is able to accurately represent complex targets. It is well known that radar measurement data are highly sensitive to minute azimuth angle shifts or trivial target structure variations. Signal fluctuations caused by relative phase differences among scatterers is commonly regarded as the source of this sensitivity. This sensitivity can be demonstrated through interference of just two point scatterers. Less well known are the effects of “phase jumps”, and the results of matched filtering. All of these are included in our model. Results for 3-D representations of T-72 tank models are generated. Target length and range profile variability patterns are extracted from the target models and compared with real target measurement data, the MSTAR dataset. They show remarkably close similarity in both comparisons.

Automatic Target Recognition of Military Vehicles With Krawtchouk Moments

Abstract: The challenge of automatic target recognition of military targets within a synthetic aperture radar scene is addressed in this paper. The proposed approach exploits the discrete-defined Krawtchouk moments, which are able to represent a detected extended target with few features, allowing its characterization. The proposed algorithm provides robust performance for target recognition, identification, and characterization, with high reliability in the presence of noise and reduced sensitivity to discretization errors. The effectiveness of the proposed approach is demonstrated using the MSTAR dataset.

Target Reconstruction Based on 3-D Scattering Center Model for Robust SAR ATR

Abstract: This paper proposes a robust synthetic aperture radar (SAR) automatic target recognition method based on the 3-D scattering center model. The 3-D scattering center model is established offline from the CAD model of the target using a forward method, which can efficiently predict the 2-D scattering centers as well as the scattering filed of the target at arbitrary poses. For the SAR images to be classified, the 2-D scattering centers are extracted based on the attributed scattering center model and matched with the predicted scattering center set using a neighbor matching algorithm. The selected model scattering centers are used to reconstruct an SAR image based on the 3-D scattering center model, which is compared with the test image to reach a robust similarity. The designed similarity measure comprehensively considers the image correlation between the test image and the model reconstructed image and the model redundancy as for describing the test image. As for target recognition, the model with the highest similarity is determined to the target type of the test SAR image when it is denied to be an outlier. Experiments are conducted on both the data simulated by an electromagnetic code and the data measured in the moving and stationary target acquisition recognition program under standard operating condition and various extended operating conditions to validate the effectiveness and robustness of the proposed method.

Fusing Deep Learning and Sparse Coding for SAR ATR

Abstract: We propose a multimodal and multidiscipline data fusion strategy appropriate for automatic target recognition (ATR) on synthetic aperture radar imagery. Our architecture fuses a proposed clustered version of the AlexNet convolutional neural network with sparse coding theory that is extended to facilitate an adaptive elastic net optimization concept. Evaluation on the MSTAR dataset yields the highest ATR performance reported yet, which is 99.33% and 99.86% for the three- and ten-class problems, respectively.

Target recognition in synthetic aperture radar images using binary morphological operations

Abstract: Feature extraction and matching are two important steps in synthetic aperture radar automatic target recognition. This paper uses the binary target region as the feature and proposes a matching scheme for the target regions using binary morphological operations. The residuals between the testing target region and its corresponding template target regions are processed by the morphological opening operation. Then, a similarity measure is defined based on the residual remains to evaluate the similarities between different targets. Afterward, a Bayesian decision fusion is employed to fuse the similarities gained by different structuring elements to further enhance the recognition performance. The nonlinearity of the opening operation as well as the Bayesian decision fusion makes the proposed method robust to the nonlinear deformations of the target region. Experimental results on the moving and stationary target acquisition and recognition dataset demonstrate the validity of the proposed method.