This PDF file contains the front matter associated with SPIE Proceedings Volume 7341, including the Title Page, Copyright information, Table of Contents, and the Conference Committee listing.

Imaging over long distances is crucial to a number of defense and security applications, such as homeland security and launch tracking. However, the image quality obtained from current long-range optical systems can be severely degraded by the turbulent atmosphere in the path between the region under observation and the imager. While this obscured image information can be recovered using post-processing techniques, the computational complexity of such approaches has prohibited deployment in real-time scenarios. To overcome this limitation, we have coupled a state-of-the-art atmospheric compensation algorithm, the average-bispectrum speckle method, with a powerful FPGA-based embedded processing board. The end result is a light-weight, lower-power image processing system that improves the quality of long-range imagery in real-time, and uses modular video I/O to provide a flexible interface to most common digital and analog video transport methods. By leveraging the custom, reconfigurable nature of the FPGA, a 20x speed increase over a modern desktop PC was achieved in a form-factor that is compact, low-power, and field-deployable.

Automatic selection of key-shots is an important step for video data processing. Depending on the purpose, key-shot selection provides user feed back on recorded data, storage reduction and viewpoint selection and it can be used for panoramic image stitching and 3D-reconstruction. In particular, investigating scenes of crime or accidental investigations involves large amount of data, containing information on physical arrangement of objects, details on surface geometry and appearances. This paper proposes an efficient method for automatic selection of key-shot, providing onsite feedback on recorded segments and automatic selection of view-points for 3D-reconstruction. It uses appearance based object and scene modeling for a freely moving, hand held camera. The camera motion is determined on two levels, comparing appearances of local image regions and full 3D reconstruction. On the lower level, the 2D-warp between subsequent video frames is used to determine local change of image appearance and derive a set of motion key frames. These keyframes than are used to determine full 3D motion and to reconstruct objects. Furthermore, key-frames are used for fast indexation and detection of loop closures. Examples for automatic key-frame selection are given for an re-enacted crime scene, and compared to manual selection.

Transmitting video over wireless with good quality is a challenging problem since video may get distorted by bit errors caused by fading and thermal noise. To address this problem, one can use adaptive transmission strategies, e.g., adapt the transmission power or the transmission data rate according to the channel condition. To achieve this adaptive control, we need to have the closed-form relationship between the video layer performance (the control objective) and the parameters in each layer (the control parameters). This paper is to analyze video distortion caused by fading and thermal noise, which we call "transmission distortion". The major contribution of this paper is that for the first time, we provide a method to predict how the transmission distortion process evolves over time, i.e., we obtain the analytical form of instantaneous transmission distortion as a function of system parameters in real-time video transmission over fading channels. We also obtain the analytical form of instantaneous transmission distortion under various conditions, e.g., data partitioning and the condition that one frame consists of multiple packets transmitted over the wireless channel; these results do not exist in the literature. A nice property of our prediction method is that it is not required for a receiver to acknowledge whether a packet is correctly received. Experimental results show that the transmission distortion predicted by our formula agrees well with the true distortion. The prediction formulae derived in this paper are crucial for real-time control in wireless video transmission.

Environmental monitoring through the method of traditional ship sampling is time consuming and requires a high survey cost. The objective of this study is to evaluate the feasibility of Landsat TM imagery for total suspended solids (TSS) mapping using a newly developed algorithm over Penang Island. The study area is the seawater region around Penang Island, Malaysia. Water samples were collected during a 3-hour period simultaneously with the satellite image acquisition and later analyzed in the laboratory above the study area. The samples locations were determined using a handheld GPS. The satellite image was geometrically corrected using the second order polynomial transformation. The satellite image also was atmospheric corrected by using ATCOR2 image processing software. The digital numbers for each band corresponding to the sea-truth locations were extracted and then converted into reflectance values for calibration of the water quality algorithm. The proposed algorithm is based on the reflectance model that is a function of the inherent optical properties of water, which can be related to its constituent's concentrations. The generated algorithm was developed for three visible wavelenghts, red, green and blue for this study. Results indicate that the proposed developed algorithm was superior based on the correlation coefficient (R) and root-mean-square deviation (RMS) values. Finally the proposed algorithm was used for TSS mapping at Penang Island, Malaysia. The generated TSS map was colour-coded for visual interpretation and image smoothing was performed on the map to remove random noise. This preliminary study has produced a promising result. This study indicates that the empirical algorithm is suitable for TSS mapping around Penang Island by using satellite Landsat TM data.

New technologies for Secondary Ion Mass Spectrometry (SIMS) produce three-dimensional hyperspectral chemical images with high spatial resolution and fine mass-spectral precision. SIMS imaging of biological tissues and cells promises to provide an informational basis for important advances in a wide variety of applications, including cancer treatments. However, the volume and complexity of data pose significant challenges for interactive visualization and analysis. This paper describes new methods and tools for computer-based visualization and analysis of SIMS data, including a coding scheme for efficient storage and fast access, interactive interfaces for visualizing and operating on three-dimensional hyperspectral images, and spatio-spectral clustering and classification.

Each year, hundred millions hectares of forests burn causing human and economic losses. For efficient fire fighting, the personnel in the ground need tools permitting the prediction of fire front propagation. In this work, we present a new technique for automatically tracking fire spread in three-dimensional space. The proposed approach uses a stereo system to extract a 3D shape from fire images. A new segmentation technique is proposed and permits the extraction of fire regions in complex unstructured scenes. It works in the visible spectrum and combines information extracted from YUV and RGB color spaces. Unlike other techniques, our algorithm does not require previous knowledge about the scene. The resulting fire regions are classified into different homogenous zones using clustering techniques. Contours are then extracted and a feature detection algorithm is used to detect interest points like local maxima and corners. Extracted points from stereo images are then used to compute the 3D shape of the fire front. The resulting data permits to build the fire volume. The final model is used to compute important spatial and temporal fire characteristics like: spread dynamics, local orientation, heading direction, etc. Tests conducted on the ground show the efficiency of the proposed scheme. This scheme is being integrated with a fire spread mathematical model in order to predict and anticipate the fire behaviour during fire fighting. Also of interest to fire-fighters, is the proposed automatic segmentation technique that can be used in early detection of fire in complex scenes.

There is a continual need for faster and accurate stereo vision algorithms. While scan-line based algorithms are fast they are less accurate and algorithms based on probabilistic models are accurate but more time consuming. In this work, a new scan-line based algorithm that defines feature fronts according to the level set method is introduced. The feature fronts are matched using a new matching criterion that also compares the feature strengths in addition to an end-point SAD score. Feature fronts are sorted based on their strengths and strongest fronts are matched first. The algorithm seeks to identify visible fronts where a front is defined to be visible if it turns out to be its match's match. Continuity of feature fronts is imposed by requiring neighboring feature fronts in one image match to neighboring feature fronts in the other image or to no front.

The effectiveness of autonomous munitions systems can be enhanced by transmitting target images to a man-in-the-loop (MITL) as the system deploys. Based on the transmitted images, the MITL could change target priorities or conduct damage assessment in real-time. One impediment to this enhancement realization is the limited bandwidth of the system data-link. In this paper, an innovative pattern-based image compression technology is presented for enabling efficient image transmission over the ultra-low bandwidth system data link, while preserving sufficient details in the decompressed images for the MITL to perform the required assessments. Based on a pattern-driven image model, our technology exploits the structural discontinuities in the image by extracting and prioritizing edge segments with their geometric and intensity profiles. Contingent on the bit budget, only the most salient segments are encoded and transmitted, therefore achieving scalable bit-streams. Simulation results corroborate the technology efficiency and establish its subjective quality superiority over JPEG/JPEG2000 as well as feasibility for real-time implementation. Successful technology demonstrations were conducted using images from surrogate seekers in an aircraft and from a captive-carry test-bed system. The developed technology has potential applications in a broad range of network-enabled weapon systems.

Multi-objective evolutionary algorithms (MOEAs) have been utilized in many fields to optimize designs and constraints using biologically inspired methods. In this research, MOEAs are used to determine more optimal DCT based filter coefficient sets in order to enhance images under various image processing attacks and functions. The filter coefficients are adapted to minimize the mean squared error and to remove noise-induced artifacts. The capabilities of the proposed enhanced image filters are demonstrated on multiple digital images.

Human being can perceive natural scenes very well under various illumination conditions. Partial reasons are due to the contrast enhancement of center/surround networks and opponent analysis on the human retina. In this paper, we propose an image enhancement model to simulate the color processes in the human retina. Specifically, there are two center/surround layers, bipolar/horizontal and ganglion/amacrine; and four color opponents, red (R), green (G), blue (B), and yellow (Y). The central cell (bipolar or ganglion) takes the surrounding information from one or several horizontal or amacrine cells; and bipolar and ganglion both have ON and OFF sub-types. For example, a +R/-G bipolar (red-center- ON/green-surround-OFF) will be excited if only the center is illuminated, or inhibited if only the surroundings (bipolars) are illuminated, or stay neutral if both center and surroundings are illuminated. Likewise, other two color opponents with ON-center/OFF-surround, +G/-R and +B/-Y, follow the same rules. The yellow (Y) channel can be obtained by averaging red and green channels. On the other hand, OFF-center/ON-surround bipolars (i.e., -R/+G and -G/+R, but no - B/+Y) are inhibited when the center is illuminated. An ON-bipolar (or OFF-bipolar) only transfers signals to an ONganglion (or OFF-ganglion), where amacrines provide surrounding information. Ganglion cells have strong spatiotemporal responses to moving objects. In our proposed enhancement model, the surrounding information is obtained using weighted average of neighborhood; excited or inhibited can be implemented with pixel intensity increase or decrease according to a linear or nonlinear response; and center/surround excitations are decided by comparing their intensities. A difference of Gaussian (DOG) model is used to simulate the ganglion differential response. Experimental results using natural scenery pictures proved that, the proposed image enhancement model by simulating the two-layer center/surrounding retinal networks can effectively enhance color images in terms of color contrast and image details.

The aim of image registration is to align two or more images taken from different viewpoints, at different time instances, or by different modalities. Image registration methods are divided into two main categories, feature based and intensity based methods. Recently intensity based methods have gained popularity since they aim at finding a dense correspondence between the images needed to be aligned without calculating correspondence between salient features. In this work, a new intensity based image registration method has been proposed and tested. This method models the source and target image as a single image displaced over time and calculates the optical flow fields in a multiresolution framework. In order to have the ability to represent complex fields, the deformation has been modelled as locally affine but globally smooth. Multiresolution image representation by steerable pyramid decomposition is integrated with the differential image registration technique in order to find accurate image deformations. The usage of steerable pyramid overcomes traditional problems in other pyramidal methods namely aliasing across different bands, lack of translation and rotation invariance. The new algorithm was validated using torso images for volunteers at the University of Alberta in addition to images captured of a cast model of the human torso. Experiments have demonstrated promising results in terms of root mean square error and average pixel error.

Image Registration is the process of fusing or overlaying of two images of the same scene that were taken at different times and/or from different viewing angles and/or by sensors with different modality or resolution. The variations in the imaging environment induce the difference between the images of the same scene. In our situation, we have two images of the same scene taken with different sensors, one image is in the visible domain and the other is an IR image. The images are captured using a visible camera and a thermal camera by placing the two devices adjacent to each other and taking the images simultaneously without a significant change in time and spatial location. Using image registration we project the visible image into the infrared image to rectify both images to the same co-ordinate system. This is done to match the sensor output and then produce an information product from the two which can be used to further analyze and assess the scene. In this paper we use the conformal log polar mapping (CLPM) for image registration. The CLPM is invariant to changes in rotation and scale: rotational changes map to a shift along one of the axes and scale changes a shift long the other. Thus the CLPM for the two images mentioned above should essentially be the same even though the two were taken from different sensors. The amount of shift can be used to determine the angle of rotation of the rotated image with respect to the original image. The same holds for the change in scale of the two images were the shift along the x-axis or the magnitude axis in the log polar mappings of the two images can be used to find the scaling factor between the two images. This method calculates the transformation parameters for rectifying the reference image to the original image or vice-versa, and can be used to register the two images that vary in phase and scale. In this paper, we present a robust approach for image registration that uses these shifts to compute the parameters representing rotation and scale change, and registers a pair of images. An application of this approach is in detecting wave height for efficient navigation of a watercraft where the images to be fused are taken from multiple sensors which are fixed on the craft. The wave parameters can be determined by looking at the common features in the edge profile of the images taken from different sensors. The goal here is to provide real time assessment of the sea state from the information obtained from the sensor suite for possible route and speed changes to increase safety and improve ride quality.

Super resolution combines a sequence of low-resolution noisy blurred images and produces either a higher resolution image or sequence. It can significantly increase image resolution without changing optical/mechanical/electrical imaging characteristics of the camera device. Existing restoration based super-resolution methods require enhancement factors (magnification) to be integer values. When the number of low resolution image frames is limited, the existing methods estimate spatial information from neighborhood so that the resulting high resolution image is blurred. Also, in real-time systems, a fixed object size for every image sequence frame is often desired. In such cases, resolution enhancement factors must be an arbitrary real number. In order to tackle these problems, we propose an alternate approach based upon a modified mathematics model and modified Maximum Likelihood (ML) estimator. Using our new model and modified ML, resolution enhancement factor can be any real number and traditional regularization operation of image restoration is not necessary. Therefore sharp edge and other high frequency contents are highly preserved. In this proposed method, L2 norm minimization is applied for data fusion without any regularization so that optimal and robust results are achieved and computation complexity is low. Also, in this paper an optimal "enhancement factor" algorithm is proposed.

Super-Resolution (SR) involves the registration of multiple images/frames and reconstruction of a single higher resolution image. The goal of this research is to use multiple, very-low resolution images, such as those produced from a video sequence in a wireless sensor network system, as input to the super-resolution process in a face recognition system. The algorithm used for face recognition is the Fisherfaces method with a nearest neighbor classifier used for the recognition decision. Super-resolution consists of two stages, a registration stage and a reconstruction stage. Testing images were segmented using a simple skin color detection approach. After cropping they were combined into groups of four to be used for the super-resolution algorithm using faces from three people. Each group of four images was used as an input to the Keren registration algorithm where the rotational and translation information was saved that was then entered into the robust super-resolution reconstruction algorithm to create a single high quality image, which was processed by the face recognition algorithm. An average of the same groups of four was tested as well as a centroid shifted average. Comparison was based on nearest neighbor classifier and on classification rates. The results were not in favor of the super-resolution method but instead, the centroid shifted average was the best in this study.

Face recognition in the infrared spectrum has attracted a lot of interest in recent years. Many of the techniques used in infrared are based on their visible counterpart, especially linear techniques like PCA (Principal Component Analysis) and LDA (Linear Discriminant Analysis). In this work, we introduce non linear dimensionality reduction approaches for multispectral face recognition. For this purpose, the following techniques were developed: global non linear techniques (Kernel-PCA, Kernel-LDA) and local non linear techniques (Local Linear Embedding, Locality Preserving Projection). The performances of these techniques were compared to classical linear techniques for face recognition like PCA and LDA. Two multispectral face recognition databases were used in our experiments: Equinox Face Recognition Database and Laval University Database. Equinox database contains images in the Visible, Short, Mid and Long waves infrared spectrums. Laval database contains images in the Visible, Near, Mid and Long waves infrared spectrums with variations in time and metabolic activity of the subjects. The obtained results are interesting and show the increase in recognition performance using local non linear dimensionality reduction techniques for infrared face recognition, particularly in near and short wave infrared spectrums.

Human-computer interfacing (HCI) describes a system or process with which two information processors, namely a human and a computer, attempt to exchange information. Computer-to-human (CtH) information transfer has been relatively effective through visual displays and sound devices. On the other hand, the human-tocomputer (HtC) interfacing avenue has yet to reach its full potential. For instance, the most common HtC communication means are the keyboard and mouse, which are already becoming a bottleneck in the effective transfer of information. The solution to the problem is the development of algorithms that allow the computer to understand human intentions based on their facial expressions, head motion patterns, and speech. In this work, we are investigating the feasibility of a stereo system to effectively determine the head position, including the head rotation angles, based on the detection of eye pupils.

For common computer interaction the mouse is established as a standard device. The recognition of freehand 3D-interaction has already been implemented by detecting the fingertips and the eyes of the users. This application is based on the stereo-photogrammetry approach with two webcams. Attempts with a single webcam have been performed as well to quit the synchronisation of two video streams. Using the range-imaging technology the user can move in front of the display from 30 cm up to the maximal ranging distance that is supported by the camera. The body, especially head and hand, can be detected in 3D within the operating range and an additional gesture-analysis tool is able to interpret the commands of the user. With this approach, the computer mouse is not needed anymore. The main topic of this paper is the multi-user interaction. Operating the computer at the same time with several users is not supported by the actual operating systems. The simultaneous detection of several users and their hands in 3D was achieved. A fast switching between the users to control the computer in turns is explained.

We present a spatially adaptive scheme for automatically searching a pair of images of a scene for unusual and interesting changes. Our motivation is to bring into play structural aspects of image features alongside the spectral attributes used for anomalous change detection (ACD). We leverage a small but informative subset of pixels, namely edge pixels of the images, as anchor points of a Delaunay triangulation to jointly decompose the images into a set of triangular regions, called trixels, which are spectrally uniform. Such decomposition helps in image regularization by simple-function approximation on a feature-adaptive grid. Applying ACD to this trixel grid instead of pixels offers several advantages. It allows: 1) edge-preserving smoothing of images, 2) speed-up of spatial computations by significantly reducing the representation of the images, and 3) the easy recovery of structure of the detected anomalous changes by associating anomalous trixels with polygonal image features. The latter facility further enables the application of shape-theoretic criteria and algorithms to characterize the changes and recognize them as interesting or not. This incorporation of spatial information has the potential to filter out some spurious changes, such as due to parallax, shadows, and misregistration, by identifying and filtering out those that are structurally similar and spatially pervasive. Our framework supports the joint spatial and spectral analysis of images, potentially enabling the design of more robust ACD algorithms.

As wide-area persistent imaging systems become cost effective, increasingly large areas of the earth can be imaged at relatively high frame rates. Efficient exploitation of the large geo-spatial-temporal datasets produced by these systems poses significant technical challenges for image and video analysis and for data mining. Significant progress in image stabilization, moving object detection and tracking, are allowing automated systems to generate hundreds to thousands of vehicle tracks from raw data, with little human intervention. However, tracking performance at this scale is unreliable, and average track length is much smaller than the average vehicle route. These are limiting factors for applications that depend heavily on track identity, i.e. tracking vehicles from their points of origin to their final destination. In this paper, we propose and evaluate a framework for wide-area motion imagery (WAMI) exploitation that minimizes the dependence on track identity. In its current form, this framework takes noisy, incomplete moving object detection tracks as input, and produces a small set of activities (e.g. multi-vehicle meetings) as output. The framework can be used to focus and direct human users and additional computation, and suggests a path towards high-level content extraction by learning from the human-in-the-loop.

Findings from the current UK national research programme, MEDUSA (Multi Environment Deployable Universal Software Application), are presented. MEDUSA brings together two approaches to facilitate the design of an automatic, CCTV-based firearm detection system: psychological-to elicit strategies used by CCTV operators; and machine vision-to identify key cues derived from camera imagery. Potentially effective human- and machine-based strategies have been identified; these will form elements of the final system. The efficacies of these algorithms have been tested on staged CCTV footage in discriminating between firearms and matched distractor objects. Early results indicate the potential for this combined approach.

Detecting complex targets, such as facilities, in commercially available satellite imagery is a difficult problem that human analysts try to solve by applying world knowledge. Often there are known observables that can be extracted by pixel-level feature detectors that can assist in the facility detection process. Individually, each of these observables is not sufficient for an accurate and reliable detection, but in combination, these auxiliary observables may provide sufficient context for detection by a machine learning algorithm. We describe an approach for automatic detection of facilities that uses an automated feature extraction algorithm to extract auxiliary observables, and a semi-supervised assisted target recognition algorithm to then identify facilities of interest. We illustrate the approach using an example of finding schools in Quickbird image data of Albuquerque, New Mexico. We use Los Alamos National Laboratory's Genie Pro automated feature extraction algorithm to find a set of auxiliary features that should be useful in the search for schools, such as parking lots, large buildings, sports fields and residential areas and then combine these features using Genie Pro's assisted target recognition algorithm to learn a classifier that finds schools in the image data.

Video, especially massive video archives, is by nature dense information medium. Compactly presenting the activities of targets of interest provides an efficient and cost saving way to analyze the content of the video. In this paper, we propose a video content analysis system to summarize and visualize the trajectories of targets from massive video archives. We first present an adaptive appearance-based algorithm to robustly track the targets in a particle filtering framework. It provides high performance while facilitating implementation of this algorithm in hardware with parallel processing. Phase correlation algorithm is used to estimate the motion of the observation platform which is then compensated in order to extract the independent trajectories of the targets. Based on the trajectory information, we develop the interface for browsing the videos which enables us to directly manipulate the video. The user could scroll over objects to view their trajectories. If interested, he/she could click on the object and drag it along the displayed path. The actual video will be played in synchronous to the mouse movement.

Moving object detection is of significant interest in temporal image analysis since it is a first step in many object identification and tracking applications. A key component in almost all moving object detection algorithms is a pixellevel classifier, where each pixel is predicted to be either part of a moving object or part of the background. In this paper we investigate a change detection approach to the pixel-level classification problem and evaluate its impact on moving object detection. The change detection approach that we investigate was previously applied to multi- and hyper-spectral datasets, where images were typically taken several days, or months apart. In this paper, we apply the approach to lowframe rate (1-2 frames per second) video datasets.

In order to control riots in crowds, it is helpful to get the ringleader under control. A great support to achieve this task is the capability to automatically track individual persons in a video sequence taken from a crowd. In this paper we address the robustness of such a tracking function. We start from the results of a previous evaluation of tracking methods, where a so-called Covariance-Tracker was found to be most appropriate. This tracker uses covariance matrices as object descriptors, as proposed by Porikli et al. The set of all covariance matrices describes a Riemannian manifold that is used to compare and update the covariance descriptors during tracking. We propose Covariance-Tracker adaptations to improve its performance. Furthermore, we summarize the performance evaluation results of the original method and compare these with the results of the adapted one. The result is a robust method for tracking people in crowds which can improve situational awareness.

The primary purpose of this study was to investigate the feasibility of using simulated data from the United Kingdom Meteorological Office (UKMO) global climate mathematical model to serve as boundary values for a regional model RM3 which has been used by NASA to make predictions about climate dynamics in West Africa. In the past, historical data has been used successfully as boundary data but this approach limits outcomes to time periods in the past. The advantage of using the UKMO data is its potential to provide input boundary data for future time periods resulting in future regional predictions. This study has provided NASA scientists with graphical and statistical summaries including visual animations that provide qualitative and quantitative information necessary for evaluating whether the UKMO data can be used as a driving force for the RM3 model. One definite conclusion of this investigation is that both spatial and temporal interpolation of UKMO results will be necessary in order to make its results compatible with the RM3 model.

Nowadays application of webcam becomes more and more popular. Thus webcams are being developed to have better resolution but lower cost. This has motivated us to evaluate the suitability of using webcam for indoor air quality monitoring. This monitoring involved determining the concentration of particulate matter with diameter less than 10 micron (PM10). An algorithm was developed to convert multispectral image pixel values acquired from this camera into quantitative values of the concentrations of PM₁₀. This algorithm was developed based on the regression analysis of relationship between the measured reflectance and the reflected components from a surface material and the ambient air. The computed PM10 values were compared to other standard values measured by a DustTrak^TM meter. The correlation results showed that the newly develop algorithm produced a high degree of accuracy as indicated by high correlation coefficient (R²) and low root-mean-square-error (RMS). This has showed that Webcam can be used for indoor air quality monitoring.

Recently we proposed a wavelet-based dynamic range compression algorithm to improve the visual quality of digital images captured from high dynamic range scenes with non-uniform lighting conditions. The fast image enhancement algorithm that provides dynamic range compression, while preserving the local contrast and tonal rendition, is also a good candidate for real time video processing applications. Although the colors of the enhanced images produced by the proposed algorithm are consistent with the colors of the original image, the proposed algorithm fails to produce color constant results for some "pathological" scenes that have very strong spectral characteristics in a single band. The linear color restoration process is the main reason for this drawback. Hence, a different approach is required for the final color restoration process. In this paper the latest version of the proposed algorithm, which deals with this issue is presented. The results obtained by applying the algorithm to numerous natural images show strong robustness and high image quality.