Improved HRR-ATR using hybridization of HMM and eigen-template-matched filtering

Paul, Anindya S.; Shaw, A.K.; Das, Kishore Kumar; Mitra, Atindra K.

doi:10.1109/icassp.2003.1202382

Cited by 8 publications

(7 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Paul et al [16] combine outputs from eigen-template based matched filters and hidden Markov models (HMM)-based clustering using a product-of-classification-probabilities rule. More recently, Gomes et al [17] proposed simple voting combinations of individual classifier decisions.…”

Section: A Atr Algorithms: a Reviewmentioning

confidence: 99%

“…This has spurred interest in combining the complementary benefits of multiple classifiers. Fusion techniques have been developed [14][15][16][17][18] that combine decisions from multiple classifiers into an ensemble classifier. These approaches reveal the presence of complementary yet correlated information present in distinct feature sets, which is exploited to a first order by fusing classifier outputs that use these features.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SAR Automatic Target Recognition Using Discriminative Graphical Models

Srinivas

Monga

Raj

2014

IEEE Trans. Aerosp. Electron. Syst.

168

View full text Add to dashboard Cite

The problem of automatically classifying sensed imagery such as synthetic aperture radar (SAR) into a canonical set of target classes is widely known as automatic target recognition (ATR). A typical ATR algorithm comprises the extraction of a meaningful set of features from target imagery followed by a decision engine that performs class assignment. While ATR algorithms have significantly increased in sophistication over the past two decades, two outstanding challenges have been identified in the rich body of ATR literature: 1) the desire to mine complementary merits of distinct feature sets (also known as feature fusion), and 2) the ability of the classifier to excel even as training SAR images are limited. We propose to apply recent advances in probabilistic graphical models to address these challenges. In particular we develop a two-stage target recognition framework that combines the merits of distinct SAR image feature representations with discriminatively learned graphical models. The first stage projects the SAR image chip to informative feature spaces that yield multiple complementary SAR image representations. The second stage models each individual representation using graphs and combines these initially disjoint and simple graphs into a thicker probabilistic graphical model by leveraging a recent advance in discriminative graph learning. Experimental results on the benchmark moving and stationary target acquisition and recognition (MSTAR) data set confirm the benefits of our framework over existing ATR algorithms in terms of improvement in recognition rates. The proposed graphical classifiers Manuscript are particularly robust when feature dimensionality is high and number of training images is small, a commonly observed constraint in SAR imagery-based target recognition.

show abstract

Section: A Atr Algorithms: a Reviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

SAR Automatic Target Recognition Using Discriminative Graphical Models

Srinivas

Monga

Raj

2014

IEEE Trans. Aerosp. Electron. Syst.

168

View full text Add to dashboard Cite

show abstract

“…The SAR cross range resolution is given by Equation 9 and the range resolution is found in Equation 10 where λ is the wavelength, L is the synthetic aperture extent, BW is the radar bandwidth, and c is the speed of light.…”

Section: Figure 3 Image Quality Vs Sensor Resolutionmentioning

confidence: 99%

“…Multi-look fusion can be accomplished a number of ways [9,10], such as using a decision-level fusion algorithm [11] to combine single look recognition results in a serial fashion or by averaging 1-D signatures from all looks to create a mean signature that is then run through an ATR algorithm [12]. More recently, other sensor modalities, including hyperspectral (HSI), multispectral (MSI) [13,14], laser radar (LADAR) [15,16], and acoustic have been applied to the CID problem with some results appearing in the literature.…”

Section: Introductionmentioning

confidence: 99%

Predicted radar/optical feature fusion gains for target identification

Kahler

Blasch

2010

Proceedings of the IEEE 2010 National Aerospace &Amp; Electronics Conference

View full text Add to dashboard Cite

Airborne tracking and identification (ID) of high value ground targets is a difficult task impacted by sensor, target, and environmental conditions. Layered sensing, using a combination of standoff and short-range sensors, maintains target track and identification in cluttered environments such as cities or densely vegetated areas through sensor diversity. Data, feature, decision, or information fusion is necessary for high confidence target classification to be achieved using multiple sensors and sensor modalities. Target identification performance is improved by exploiting the extra information gained from independent sensing modalities through information fusion for automatic target recognition (ATR). Increased target ID has been demonstrated using spatial-temporal multi-look sensor fusion and decision level fusion. To further enhance target ID performance and increase decision confidence, feature level fusion techniques are being investigated. A fusion performance model for feature level fusion was applied to a combination of sensor types and features to provide estimates of a fusion gain. This paper presents a fusion performance gain for Synthetic Aperture Radar (SAR), electro-optical (EO), and infrared (IR) video stationary target identification.Keywords: decision level fusion, feature level fusion, electro-optical (EO), infrared (IR), synthetic aperture radar (SAR), information fusion, automatic target recognition (ATR), national imagery interpretability rating scale (NIIRS).

show abstract

“…A number of papers have been published that evaluate 1D HRR ATR solutions [7,8,9]. Classifiers have been developed for correlation [10], Bayes and Dempster Shafer information fusion approaches [11], and Neuro-Fuzzy methods [12].…”

Section: Introductionmentioning

confidence: 99%

Target identification performance improvement from enhanced HRR radar clutter suppression

Kahler

Blasch

2009

Proceedings of the IEEE 2009 National Aerospace &Amp; Electronics Conference (NAECON)

View full text Add to dashboard Cite

Airborne radar tracking in moving ground vehicle scenarios is impacted by sensor, target, and environmental dynamics. Moving targets can be assessed with 1-D High Range Resolution (HRR) Radar profiles with sufficient signal-to-noise (SNR) present which contain enough feature information to discern one target from another to help maintain track or to identify the vehicle. Typical radar clutter suppression algorithms developed for processing moving ground target data not only remove the surrounding clutter but also a portion of the target signature. Enhanced clutter suppression can be achieved using a multi-channel signal subspace (MSS) algorithm which preserves target features. In this paper, we exploit extra information from enhanced clutter suppression for automatic target recognition (ATR), present a gain comparison using displaced phase center antenna (DPCA) and MSS clutter suppressed HRR data, and generate confusion-matrix identification results. The results show that more channels for MSS increase ID over DCPA, result in a slightly noisier clutter suppressed image, and preserve more target features after clutter cancellation

show abstract

Improved HRR-ATR using hybridization of HMM and eigen-template-matched filtering

Cited by 8 publications

References 6 publications

SAR Automatic Target Recognition Using Discriminative Graphical Models

SAR Automatic Target Recognition Using Discriminative Graphical Models

Predicted radar/optical feature fusion gains for target identification

Target identification performance improvement from enhanced HRR radar clutter suppression

Contact Info

Product

Resources

About