We investigate the relationship between paired SAR and optical images. SAR sensors have the capabilities of penetrating clouds and capturing data at night, whereas optical sensors cannot. We are interested in the case where we have access to both modalities during training, but only the SAR during test time. To that end, we developed a framework that inputs a SAR image and predicts a Canny edge map of the optical image, which retains structural information, while removing superfluous details. Our experiments show that by additionally using this predicted edge map for downstream tasks, we can outperform the same model that only uses the SAR image.
The topic of constructing data-dependent dictionaries, referred to as dictionary learning, has received considerable
interest in the past decade. In this work, we compare the ability of two dictionary learning algorithms,
K-SVD and geometric multi-resolution analysis (GMRA), to perform image reconstruction using a fixed number
of coefficients. K-SVD is an algorithm originating from the compressive sensing community and relies on
optimization techniques. GMRA is a multi-scale technique that is based on manifold approximation of highdimensional
point clouds of data. The empirical results of this work using a synthetic dataset of images of
vehicles with diversity in viewpoint and lighting show that the K-SVD algorithm exhibits better generalization
reconstruction performance with respect to test images containing lighting diversity that were not present in the
construction of the dictionary, while GMRA exhibits superior reconstruction on the training data.
The human brain has the capability to process high quantities of data quickly for detection and recognition tasks. These tasks are made simpler by the understanding of data, which intentionally removes redundancies found in higher dimensional data and maps the data onto a lower dimensional space. The brain then encodes manifolds created in these spaces, which reveal a specific state of the system. We propose to use a recurrent neural network, the nonlinear line attractor (NLA) network, for the encoding of these manifolds as specific states, which will draw untrained data towards one of the specific states that the NLA network has encoded. We propose a Gaussian-weighted modular architecture for reducing the computational complexity of the conventional NLA network. The proposed architecture uses a neighborhood approach for establishing the interconnectivity of neurons to obtain the manifolds. The modified NLA network has been implemented and tested on the Electro-Optic Synthetic Vehicle Model Database created by the Air Force Research Laboratory (AFRL), which contains a vast array of high resolution imagery with several different lighting conditions and camera views. It is observed that the NLA network has the capability for representing high dimensional data for the recognition of the objects of interest through its new learning strategy. A nonlinear dimensionality reduction scheme based on singular value decomposition has found to be very effective in providing a low dimensional representation of the dataset. Application of the reduced dimensional space on the modified NLA algorithm would provide fast and more accurate recognition performance for real time applications.
Target detection is limited based on a specific sensors capability; however, the combination of multiple sensors will
improve the confidence of target detection. Confidence of detection, tracking and identifying a target in a multi-sensor
environment depends on intrinsic and extrinsic sensor qualities, e.g. target geo-location registration, and environmental
conditions 1. Determination of the optimal sensors and classification algorithms, required to assist in specific target
detection, has largely been accomplished with empirical experimentation. Formulation of a multi-sensor effectiveness
metric (MuSEM) for sensor combinations is presented in this paper. Leveraging one or a combination of sensors should
provide a higher confidence of target classification. This metric incorporates the Dempster-Shafer Theory for decision
analysis. MuSEM is defined for weakly labeled multimodal data and is modeled and trained with empirical fused sensor
detections; this metric is compared to Boolean algebra algorithms from decision fusion research. Multiple sensor
specific classifiers are compared and fused to characterize sensor detection models and the likelihood functions of the
models. For area under the curve (AUC), MuSEM attained values as high as .97 with an average difference of 5.33%
between Boolean fusion rules. Data was collected from the Air Force Research Lab’s Minor Area Motion Imagery
(MAMI) project. This metric is efficient and effective, providing a confidence of target classification based on sensor
combinations.
KEYWORDS: Image classification, Detection and tracking algorithms, Light sources and illumination, Principal component analysis, Reconstruction algorithms, Associative arrays, Evolutionary algorithms, Databases, Liquid crystals, Chemical elements
The Sparse Representation for Classification (SRC) algorithm has been demonstrated to be a state-of-the-art algorithm for facial recognition applications. Wright et al. demonstrate that under certain conditions, the SRC algorithm classification performance is agnostic to choice of linear feature space and highly resilient to image corruption. In this work, we examined the SRC algorithm performance on the vehicle recognition application, using images from the semi-synthetic vehicle database generated by the Air Force Research Laboratory. To represent modern operating conditions, vehicle images were corrupted with noise, blurring, and occlusion, with representation of varying pose and lighting conditions. Experiments suggest that linear feature space selection is important, particularly in the cases involving corrupted images. Overall, the SRC algorithm consistently outperforms a standard k nearest neighbor classifier on the vehicle recognition task.
The ability to accurately detect a target of interest in a hyperspectral imagery (HSI) is largely dependent on
the spatial and spectral resolution. While hyperspectral imaging provides high spectral resolution, the spatial
resolution is mostly dependent on the optics and distance from the target. Many times the target of interest
does not occupy a full pixel and thus is concealed within a pixel, i.e. the target signature is mixed with
other constituent material signatures within the field of view of that pixel. Extraction of spectral signatures
of constituent materials from a mixed pixel can assist in the detection of the target of interest. Hyperspectral
unmixing is a process to identify the constituent materials and estimate the corresponding abundances from the
mixture. In this paper, a framework based on non-negative matrix factorization (NMF) is presented, which is
utilized to extract the spectral signature and fractional abundance of human skin in a scene. The NMF technique
is employed in a supervised manner such that the spectral bases of each constituent are computed first, and then
these bases are applied to the mixed pixel. Experiments using synthetic and real data demonstrate that the
proposed algorithm provides an effective supervised technique for hyperspectral unmixing of skin signatures.
Vehicle tracking is an integral component in layered sensing exploitation applications. The utilization of a
combination of sensing modalities and processing techniques provides better insight about a situation than can
be achieved with a single sensing modality. In this work, several robust features are explored for vehicle tracking
using data captured in a remote sensing setting. A target area is surveyed by a sensor operating capturing
polarization information in the longwave infrared (LWIR) band. We here extend our previous work ([1]) to
experimental analysis of several feature sets including three classic features (Stokes images, DoLP, the Degree
of Linear Polarization, and AoP, the Angle of Polarization) and several geometry inspired features.1
Pattern recognition deals with the detection and identification of a specific target in an unknown input scene. Target
features such as shape, color, surface dynamics, and material characteristics are common target attributes used for
identification and detection purposes. Pattern recognition using multispectral (MS), hyperspectral (HS), and polarization-based
spectral (PS) imaging can be effectively exploited to highlight one or more of these attributes for more efficient
target identification and detection. In general, pattern recognition involves two steps: gathering target information from
sensor data and identifying and detecting the desired target from sensor data in the presence of noise, clutter, and other
artifacts. Multispectral and hyperspectral imaging (MSI/HSI) provide both spectral and spatial information about the
target. As the reflection or emission spectral signatures depend on the elemental composition of objects residing within
the scene, the polarization state of radiation is sensitive to the surface features such as relative smoothness or roughness,
surface material, shapes and edges, etc. Therefore, polarization information imparted by surface reflections of the target
yields unique and discriminatory signatures which could be used to augment spectral target detection techniques, through
the fusion of sensor data. Sensor data fusion is currently being used to effectively recognize and detect one or more of
the target attributes. However, variations between sensors and temporal changes within sensors can introduce noise in the
measurements, contributing to additional target variability that hinders the detection process. This paper provides a quick
overview of target identification and detection using MSI/HSI, highlighting the advantages and disadvantages of each. It
then discusses the effectiveness of using polarization-based imaging in highlighting some of the target attributes at single
and multiple spectral bands using polarization spectral imaging (PSI), known as spectropolarimetry imaging.
Spectral variability remains a challenging problem for target detection in hyperspectral (HS) imagery. In previous work,
we developed a target detection scheme using the kernel-based support vector data description (SVDD). We constructed
a first-order Markov-based Gaussian model to generate samples to describe the spectral variability of the target class.
However, the Gaussian-generated samples also require selection of the variance parameter σ 2 that dictates the level of
variability in the generated target class signatures. In this work, we have investigated the use of decision-level fusion
techniques for alleviating the problem of choosing a proper value of σ 2 . We have trained a collection of SVDDs with
unique variance parameters σ 2 for each of the target training sets and have investigated their combination using the
traditional AND, OR, and majority vote (MV) decision-level rules. We have inserted target signatures into an urban HS
scene with differing levels of spectral variability to explore the performance of the proposed scheme in these scenarios.
Experiments show that the MV fusion rule is the best choice, providing relatively low false positive rates (FPR) while
yielding high true positive rates (TPR). Detection results show that the proposed SVDD-based decision-level scheme
using the MV fusion rule is highly accurate and yields higher true positive rates (TPR) and lower false positive rates
(FPR) than the adaptive matched filter (AMF).
Spectral variability remains a challenging problem for target detection in hyperspectral (HS) imagery. In this paper, we
have applied the kernel-based support vector data description (SVDD) to perform full-pixel target detection. In target
detection scenarios, we do not have a collection of samples characterizing the target class; we are typically given a pure
target signature that is obtained from a spectral library. In our work, we use the pure target signature and first-order
Markov theory to generate N samples to model the spectral variability of the target class. We vary the value of N and
observe its effect to determine a value of N that provides acceptable detection performance.
We have inserted target signatures into an urban HS scene with varying levels of spectral variability to explore the
performance of the proposed SVDD target detection scheme in these scenarios. The proposed approach makes no
assumptions regarding the underlying distribution of the scene data as do traditional stochastic detectors such as the
adaptive matched filter (AMF). Detection results in the form of confusion matrices and receiver-operating-characteristic
(ROC) curves demonstrate that the proposed SVDD-based scheme is highly accurate and yields higher true positive rates
(TPR) and lower false positive rates (FPR) than the AMF.
A practical challenge that designers of hyperspectral (HS) target detection algorithms must confront is the variety of
spectral sampling properties exhibited by various HS imaging sensors. Examples of these variations include different
spectral resolutions and the possibility of regular or irregular sampling. To confront this problem, we propose
construction of a spectral synthetic discriminant signature (SSDS). The SSDS is constructed from q spectral training
signatures which are obtained by sampling the original target signature. Since the SSDS is formulated offline, it does not
impose any burden on the processing speed of the recognition process. Results on our HS scenery show that use of the
SSDS in conjunction with the spectral fringe-adjusted joint transform correlation (SFJTC) algorithm provides spectrallyinvariant
target detection, yielding area under ROC curve (AUROC) values above 0.993.
Recently, the use of imaging polarimetry has received considerable attention for use in automatic target recognition
(ATR) applications. In military remote sensing applications, there is a great demand for sensors that are capable of
discriminating between real targets and decoys. Accurate discrimination of decoys from real targets is a challenging
task and often requires the fusion of various sensor modalities that operate simultaneously. In this paper, we use a
simple linear fusion technique known as the high-boost fusion method for effective discrimination of real targets in the
presence of multiple decoys. The HBF assigns more weight to the polarization-based imagery in forming the final
fused image that is used for detection. We have captured both intensity and polarization-based imagery from an
experimental laboratory arrangement containing a mixture of sand/dirt, rocks, vegetation, and other objects for the
purpose of simulating scenery that would be acquired in a remote sensing military application. A target object and
three decoys that are identical in physical appearance (shape, surface structure and color) and different in material
composition have also been placed in the scene. We use the wavelet-filter joint transform correlation (WFJTC)
technique to perform detection between input scenery and the target object. Our results show that use of the HBF
method increases the correlation performance metrics associated with the WFJTC-based detection process when
compared to using either the traditional intensity or polarization-based images.
Spectral variability remains a challenging problem for target detection and classification in hyperspectral imagery (HSI).
In this paper, we have applied the nonlinear support vector data description (SVDD) to perform full-pixel target
detection. Using a pure target signature, we have developed a novel pattern recognition (PR) algorithm to train an SVDD
to characterize the target class. We have inserted target signatures into an urban hyperspectral (HS) scene with varying
levels of spectral variability to explore the performance of the proposed SVDD target detector in different scenarios. The
proposed approach makes no assumptions regarding the underlying distribution of the scene data as do traditional
statistical detectors such as the matched filter (MF). Detection results in the form of confusion matrices and receiver-operating-
characteristic (ROC) curves demonstrate that the proposed SVDD-based algorithm is highly accurate and
yields higher true positive rates (TPR) and lower false positive rates (FPR) than the MF.
Recently, the 1-D spectral fringe-adjusted joint transform correlation (SFJTC) technique has been combined with the
discrete wavelet transform (DWT) as an effective means for providing robust target detection in hyperspectral imagery.
This paper expands upon earlier work that demonstrates the utility of the DWT in conjunction with SFJTC for detection.
We show that using selected DWT coefficients at a given decomposition level can significantly improve the ROC curve
behavior of the detection process in comparison to using the original hyperspectral signatures. The DWT coefficients
that are selected for detection are based on a supervised training process that uses the pure target signature and randomly
selected samples from the scene. We illustrate this by conducting experiments on two different hyperspectral scenes
containing varying amounts of simulated noise. Results show that use of the selected DWT coefficients significantly
improves the ROC curve detection behavior in the presence of noise.
Recently, the 1-D spectral fringe-adjusted joint transform correlation (SFJTC) technique has been proposed as an
effective means for performing deterministic target detection in hyperspectral imagery. In this work, we explore the use
of the discrete wavelet transform (DWT) as a pre-processing tool for SFJTC-based target detection. To quantify
improvement and compare performance in the detection process, receiver operating characteristic (ROC) curves are
generated and the areas under the ROC curves (AUROC) are computed. The basic premise of this work is that selected
coefficients generated from a desired level of the DWT decomposition of the data can be used in place of the original
data for improved SFJTC-based detection. We illustrate this by conducting experiments on two different hyperspectral
scenes containing varying amounts of simulated noise. Results indicate that use of the DWT coefficients significantly
improves the detection performance, especially in the presence of noise.
We present a novel synthetic discriminant function (SDF) formulated from the Laplacian-enhanced (L) training images for the rotation and scale invariant target detection. It is shown that the proposed LSDF yields significantly improved correlation performance compared to the traditional SDF. Since the LSDF is formulated off line, it does not have any burden on the processing speed of the system.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.