In this thesis we present a method for detecting planar targets in real-time on power-constrained, or low-powered, hand-held devices such as mobile phones. We adopt the feature recognition (also referred to as feature matching) approach and employ fast-to-compute local feature descriptors to establish point correspondences. To obtain a satisfactory localization accuracy, most local feature descriptors seek a transformation of the input intensity patch that is invariant to various geometric and photometric deformations. Generally, such transformations are computationally intensive, hence are not ideal for real-time applications on limited hardware platforms. On the other hand, descriptors which are fast to compute are typically limited in their ability to provide invariance to a vast range of deformations. To address these shortcomings, we have developed a learning-based approach which can be applied to any local feature descriptor to increase the system’s robustness to both affine and perspective deformations. The motivation behind applying a learning-based approach is to transfer as much of the computational burden (as possible) onto an offline training phase, allowing a reduction in cost during online matching. The approach comprises of identifying keypoints which remain stable under artificially induced perspective transformations, extracting the corresponding feature vectors, and finally aggregating the feature vectors of coincident keypoints to obtain the final descriptors. We strictly focus on objects which are planar, thus allowing us to synthesize images of the object in order to capture the appearance of keypoint patches under several perspectives.

Title from first page of PDF document. Includes bibliographical references (p. 40).

Thesis (M.S. in Computer Science) Naval Postgraduate School, March 1994. / Thesis advisors, Sehung Kwak, Judith H. Lind. AD-A280 040. "March 1994." Includes bibliographical references (p. 113-114). Also available online.

In this thesis we present a method for detecting planar targets in real-time on power-constrained, or low-powered, hand-held devices such as mobile phones. We adopt the feature recognition (also referred to as feature matching) approach and employ fast-to-compute local feature descriptors to establish point correspondences. To obtain a satisfactory localization accuracy, most local feature descriptors seek a transformation of the input intensity patch that is invariant to various geometric and photometric deformations. Generally, such transformations are computationally intensive, hence are not ideal for real-time applications on limited hardware platforms. On the other hand, descriptors which are fast to compute are typically limited in their ability to provide invariance to a vast range of deformations. To address these shortcomings, we have developed a learning-based approach which can be applied to any local feature descriptor to increase the system’s robustness to both affine and perspective deformations. The motivation behind applying a learning-based approach is to transfer as much of the computational burden (as possible) onto an offline training phase, allowing a reduction in cost during online matching. The approach comprises of identifying keypoints which remain stable under artificially induced perspective transformations, extracting the corresponding feature vectors, and finally aggregating the feature vectors of coincident keypoints to obtain the final descriptors. We strictly focus on objects which are planar, thus allowing us to synthesize images of the object in order to capture the appearance of keypoint patches under several perspectives.

The detection of weak, ground-moving targets can be improved through effective utilization of additional target signal energy collected over an extended dwell time. The signal model used in conventional radar processing limits integration of signal energy over an extended dwell. Two solutions that consider the complexity of the extended-dwell signal model and effectively combine signal energy collected over a long dwell are presented. The first solution is a single-channel algorithm that provides an estimate of the optimal detector to maximize output signal-to-interference-plus-noise ratio for the extended dwell time signal. Rather than searching for the optimal detector in an intractably large filter bank that contains all combinations of phase components, the single-channel algorithm projects dictionary entries against the data to estimate the signal’s linear and nonlinear phase components sequentially with small, phase-specific dictionaries in a multistage process. When used as the detector, the signal model formed from the estimated phase components yields near optimal performance for a wide range of target parameters for dwell times up to four seconds. In comparison, conventional radar processing methods are limited to an integration time of approximately 100 milliseconds. The second solution is a multichannel, multistage algorithm based on element-space pre-Doppler space-time-adaptive processing with two modifications that make it suitable for detection of weak targets whose energy is collected over an extended dwell time. The multichannel solution detects targets with lower radial velocities at significantly lower signal-to-noise ratios (SNRs) than conventional radar processing methods. The decrease in required input SNR for the multichannel solution as compared to conventional methods nearly doubles the detection range for a typical target of interest. Future related research includes extension of these concepts to other radar applications and investigation of algorithm performance for the multiple-target scenario.

Emotional cues within the environment capture our attention and influence how we perceive our surroundings. Past research has shown that emotional cues presented before the detection of a perceptual gap can actually impair the perception of elementary visual features (e.g., the lack of detail creating a spatial gap) while simultaneously improving the perception of fast temporal features of vision (e.g., the rapid onset, offset, and re-emergence of a stimulus). This effect has been attributed to amygdalar enhancements of visual inputs conveying emotional features along magnocellular channels. The current study compared participants’ ability to detect spatial and temporal gaps in simple stimuli (a Landolt Circle) after first being exposed to a facial cue in the periphery. The study was an attempt to replicate past research using younger adult samples while also extending these findings to an older adult sample. Unlike younger adults, older adults generally display an attentional bias toward positive instead of negative emotional facial expressions. It is not clear if this positivity bias is strictly driven by cognitive control processes or if there is a change in the human visual system with age that reduces the amplification of negative emotive expressions by the amygdala. The current study used psychophysical data to determine if the rapid presentation of an emotional cue and subsequent perceptual target to older adults leads to the same benefit to temporal vision evinced by younger adults or if amygdalocortical enhancements to perception degrade with age. The current study was only able to partly replicate findings from past research. The negative facial cues that were presented in the periphery did not lead to an enhancement in temporal gap detection for the younger adult sample nor a reduction in spatial gap detection. In fact, the opposite was found. Younger adults’ spatial gap detection benefited from the negative emotional cues. The negative and neutral emotional cues had no effect on the older adult sample. The older adults’ performance on both gap detection tasks was not impacted by the emotional cues

<p>The main goal of this thesis is to show the strength of combining datasets from two different types of sensors to find anomalies in their data. The sensors used in this thesis are a hyperspectral camera and a scanning 3-D laser.</p><p>The report can be divided into two main parts. The first part discusses the properties of one of the datasets and how these are used to isolate anomalies. An issue to deal with here is not only what properties to look at, but how to make the process automatic. The information retained from the first dataset is then used to make intelligent choices in the second dataset. Again, one of the challenges is to make this process automatic and accurate. The second part of the project consists of presenting the results in a way that gives the most information to the user. This is done with a graphical user interface that allows the user to manipulate the way the result is presented.</p><p>The conclusion of this project is that the information from the combined sensor datasets gives better results than the sum of the information from the individual datasets. The key of success is to play to the strengths of the datasets in question. An important block of the work in this thesis, the calibration of the two sensors, was completed by Kevin Chan as his thesis work in Electrical Engineering at the University of Lund. His contribution gave access to calibrated data that supported the results presented in this report.</p>

The main goal of this thesis is to show the strength of combining datasets from two different types of sensors to find anomalies in their data. The sensors used in this thesis are a hyperspectral camera and a scanning 3-D laser. The report can be divided into two main parts. The first part discusses the properties of one of the datasets and how these are used to isolate anomalies. An issue to deal with here is not only what properties to look at, but how to make the process automatic. The information retained from the first dataset is then used to make intelligent choices in the second dataset. Again, one of the challenges is to make this process automatic and accurate. The second part of the project consists of presenting the results in a way that gives the most information to the user. This is done with a graphical user interface that allows the user to manipulate the way the result is presented. The conclusion of this project is that the information from the combined sensor datasets gives better results than the sum of the information from the individual datasets. The key of success is to play to the strengths of the datasets in question. An important block of the work in this thesis, the calibration of the two sensors, was completed by Kevin Chan as his thesis work in Electrical Engineering at the University of Lund. His contribution gave access to calibrated data that supported the results presented in this report.

The active surveillance of public and private sites is increasingly becoming a very important and critical issue. It is therefore, imperative to develop mobile surveillance systems to protect these sites. Modern surveillance systems encompass spatially distributed mobile and static sensors in order to provide effective monitoring of persistent and transient objects and events in a given Area Of Interest (AOI). The realization of the potential of mobile surveillance requires the solution of different challenging problems such as task allocation, mobile sensor deployment, multisensor management, cooperative object detection and tracking, decentralized data fusion, and interoperability and accessibility of system nodes. This thesis proposes a market-based framework that can be used to handle different problems of mobile surveillance systems. Task allocation and cooperative target-tracking are studied using the proposed framework as two challenging problems of mobile surveillance systems. These challenges are addressed individually and collectively.

Whenever a new sensor or system comes online, engineers and analysts responsible for processing the measured data turn first to methods that are tried and true on existing systems. This is a natural, if not wholly logical approach, and is exactly what has happened in the advent of hyperspectral imagery (HSI) exploitation. However, a closer look at the assumptions made by the approaches published in the literature has not been undertaken. This thesis analyzes three key aspects of HSI exploitation: statistical data modeling, covariance estimation from training data, and dimension reduction. These items are part of standard processing schemes, and it is worthwhile to understand and quantify the impact that various assumptions for these items have on target detectability and detection statistics. First, the accuracy and applicability of the standard Gaussian (i.e., Normal) model is evaluated, and it is shown that the elliptically contoured t-distribution (EC-t) sometimes offers a better statistical model for HSI data. A finite mixture approach for EC-t is developed in which all parameters are estimated simultaneously without a priori information. Then the effects of making a poor covariance estimate are shown by including target samples in the training data. Multiple test cases with ground targets are explored. They show that the magnitude of the deleterious effect of covariance contamination on detection statistics depends on algorithm type and target signal characteristics. Next, the two most widely used dimension reduction approaches are tested. It is demonstrated that, in many cases, significant dimension reduction can be achieved with only a minor loss in detection performance. In addition, a concise development of key HSI detection algorithms is presented, and the state-of-the-art in adaptive detectors is benchmarked for land mine targets. Methods for detection and identification of airborne gases using hyperspectral imagery are discussed, and this application is highlighted as an excellent opportunity for future work.