Proc. SPIE. 8711, Sensors, and Command, Control, Communications, and Intelligence (C3I) Technologies for Homeland Security and Homeland Defense XII
KEYWORDS: Digital signal processing, Video acceleration, Detection and tracking algorithms, Sensors, Video, Field programmable gate arrays, Video surveillance, Surveillance, Surveillance systems, Infrared radiation
Long-range surveillance systems are typically used in rural areas for detecting and tracking illegal border crossings, trafficking and drug activity. These systems commonly deploy mast or tower-based surveillance systems equipped with thermal infrared cameras, which have the advantage of providing early warnings and increasing the range of observation. However, these systems are subject to high frequency vibration due to slight wind or wind gusts, which is difficult to correct mechanically. In order to identify the border activity, it is critical for the vision system to robustly detect the objects in the scene, classify the objects and track the detected targets. The performance of these post-processing algorithms is known to suffer if the video is not properly stabilized.
Surveillance systems in rural areas, particularly in thermal band, pose several unique challenges to video stabilization algorithms. First, the scene rarely contains man-made objects. Water surface, trees and forests present very low contrast and ambiguous textures such that stabilization algorithms struggle to consistently and repeatedly extract distinctive corners and features. Second, even if the system captures certain human activities or structural objects in the scene, the video typically lacks sharpness in the background due to the motion blur at the long range. In this research paper, we propose a biologically-inspired, robust and compact video motion stabilization algorithm, which is ideal for rural areas. Our novel algorithm is compared quantitatively with other competing algorithm (SURF) in terms of robustness and performance. Finally, we evaluate the resource usage on FPGA platforms.
Military forces and law enforcement agencies are facing new challenges for persistent surveillance as the area
of interest shifts towards urban environments. Some of the challenges include tracking vehicles and dismounts
within complex road networks, traffic patterns and building structures. Under these conditions, conventional
video tracking algorithms suffer from target occlusion, lost tracks and stop-and-start. Furthermore, these
algorithms typically depend solely on pixel-based features to detect and locate potential targets, which are
computationally intensive and time consuming.
This research paper investigates the fusion of geographic information into video-based target tracking algorithms
for persistent surveillance. A geographic information system (GIS) has the capability to store attributes
about a target's surroundings - such as road direction and boundaries, intersections and speed limit - and can be
used as a decision-making tool in prediction and analysis. Fusing this prediction capability into conventional
video-centric target tracking algorithms provides geographical context to the target feature space improves
occlusion of targets and reduces the search area for tracking. The GIS component specifically improves the
performance of target tracking by minimizing the search area a target is likely to be located. We present the
results from our simulations to demonstrate the feasibility of the proposed technique with video collected from
a prototype persistent surveillance system. Our approach maintains compatibility with existing GIS databases
and provides an integrated solution for multi-source target tracking algorithms.
This research paper investigates the use of the wavelet transform to extract spatially-invariant wavelet-based
shape signatures for automatic target recognition (ATR). Target signatures based on shape information can
be generally categorized as either contour-based or region-based. The wavelet-based shape signatures facilitate
detection and localization of important edge and texture information aiding in discrimination between targets.
To demonstrate the advantages of both edge and region information, we present an approach that combines
region-based shape methods and the wavelet transform for generating target signatures. Our approach generates
a rotationally invariant class of wavelet signatures based on the spatial ground pixel coverage of the target
is determined from the region-of-interest (ROI) in the wavelet domain. This process results in a multiresolution
representation of the target, and provides a hierarchical approach to target signature-matching. We
demonstrate this methodology using signatures from aircraft targets utilizing the Angular Radial Transform
(ART) as the region-based shape signature. Region-based signatures are shown more robust than contourbased
signatures in the presence of noise and disconnected target regions providing greater confidence in target
identification. Our research results show the value of combining the rotational invariance of the ART signatures
with the localization and edge discrimination properties of the wavelet transform.
The three-dimensional (3-D) nature and the unorganized structure of topographic LIDAR data pose several challenges
for target recognition tasks. In the past, several approaches have applied two-dimensional transformations such as spinimages
or Digital Elevation Maps (DEMs) as an intermediate step for analyzing the 3-D data with two-dimensional
(2-D) methods. However, these techniques are computationally intensive and often sacrifice some of the overall
geometrical relationship of the target points.
In this paper, we present a simple and efficient 3-D spatial transformation that preserves the geometrical attributes of the
LIDAR data in all its dimensions. This transformation permits the utilization of well established statistical and shapebased
descriptors for the implementation of an automatic target recognition algorithm. We evaluate our transformation
and analysis technique on a set of simulated LIDAR point clouds of ground vehicles with varied obstructions and noise
levels. Classification results demonstrate that our approach is efficient, tolerant to scale, rotation, and robust to noise and
We present a scalable registration algorithm for aligning large-frame imagery compressed with the JPEG2000 coding standard. Unlike traditional approaches, the proposed method registers the images in the compressed domain, which eliminates the need to reconstruct the full image prior to performing registration. Two forms of scalability are exploited during registration: resolution and quality. Resolution scalability results from the native multiresolution image representation of the discrete wavelet transform utilized as a building block in JPEG2000. Quality scalability relates to the embedded block coding with optimal truncation (EBCOT) used for compressing the wavelet coefficients. This combination allows registration on selectable resolution levels and quality layers, which enables registration of large-frame imagery at low bit rates over constrained bandwidth channels. Furthermore, the hierarchical nature of the algorithm provides a trade-off between registration accuracy and computational complexity. Experimental results show that the proposed algorithm exhibits consistent registration performance across a range of quality levels (3.5 to 0.5 bpp) for frames sizes of 2 K×4 K. We present simulation results with imagery collected from a prototype persistent surveillance system to demonstrate the feasibility of the proposed algorithm in real-world scenarios.
This paper describes a registration algorithm for aligning large frame imagery compressed with the JPEG2000
compression standard. The images are registered in the compressed domain using wavelet-based techniques.
Unlike traditional approaches, our proposed method eliminates the need to reconstruct the full image prior
to performing registration. The proposed method is highly scalable allowing registration to be performed on
selectable resolution levels, quality layers, and regions of interest. The use of the hierarchical nature of the wavelet
transform also allows for the trade-off between registration accuracy and processing speed. We present the
results from our simulations to demonstrate the feasibility of the proposed technique in real-world scenarios with
streaming sources. The wavelet-based approach maintains compatibility with JPEG2000 and enables additional
features not offered by traditional approaches.
Image decomposition using directional filter banks is useful in discovering and extracting edge orientation cues for
target detection in airborne surveillance images. Since images of interest are very large and the filtered images are not
downsampled in the application of interest, conventional filtering can be computationally extremely demanding and
there is a need to explore procedures to make the filtering efficient. In this paper a novel filter bank structure for
directional filtering of images is proposed and its design described. The design is carried out by imposing structural
constraints on the filters, which are implemented using a generalized notion of separable filtering. The structure uses
one-dimensional (1-D) filters as building blocks, which are employed in novel configurations to obtain filters with
narrow wedge-shaped passbands. Design procedures have been developed for constructing 16-band, 32-band, and 64-
band partitions starting with either built-in or user-specified 1-D prototypes. Implementations of filters using the
proposed method show significant improvement compared with conventional implementation, often more by an order of
magnitude, which is also supported by a theoretical analysis of the filter complexity.
Efficient processing of imagery derived from remote sensing systems has become ever more important due to increasing
data sizes, rates, and bit depths. This paper proposes a target detection method that uses a special class of wavelets based on
highly frequency-selective directional filter banks. The approach helps isolate object features in different directional filter
output components. These components lend themselves well to the application of powerful denoising and edge detection
procedures in the wavelet domain. Edge information is derived from directional wavelet decompositions to detect targets
of known dimension in electro optical imagery. Results of successful detection of objects using the proposed method are
presented in the paper. The approach highlights many of the benefits of working with directional wavelet analysis for
image denoising and detection.
This paper investigates the use of wavelets to automatically register remotely sensed images. The proposed algorithm is based on the Laplacian of Gaussian (LoG) filter to automatically extract ground control points and the discrete wavelet transform (DWT) for multiresolution analysis. The structural properties of the wavelet coefficients can be exploited in a unique fashion to reduce the search space for the control points. The inherent multiresolution
processing of the image data provides an efficient method for registering large image data sets because the full-size does not require processing.