No description available.

Cette thèse présente la méthode de détection de changements MIMOSA (Method for generalIzed Means Ordered Series Analysis). Cette nouvelle méthode permet de détecter automatiquement des changements entre couples ou séries temporelles d’images SAR. En effet, grâce aux moyennes temporelles, le nombre d’images en jeu n’importe plus puisque seulement deux moyennes différentes sont comparées de sorte à détecter les changements (par exemple moyenne géométrique et moyenne quadratique). De ce fait, les grand volumes de données disponibles de nos jours sont exploitables plus facilement puisque l’information utile est «résumée» dans les moyennes. Le seul paramètre de la détection est le taux de fausses alarmes obtenu dans le résultat, ce qui rend son analyse plus intuitive. Les cartes de changements fournies par MIMOSA sont de très bonne qualité en comparaison à celles fournies par d’autres méthodes. De nombreux tests ont été mis en place pour constater la robustesse de la méthode MIMOSA face aux problèmes les plus souvent rencontrés, comme une mauvaise calibration radiométrique, ou encore un mauvais recalage. Une interface graphique a de plus été développée autour de MIMOSA, incorporant de nombreux outils de préparation et traitement des données, ainsi que des outils d’analyse des résultats. / This PhD thesis presents the MIMOSA (Method for generalIzed Means Ordered Series Analysis) change detection methood. This new technique can automatically detect changes between SAR image pairs or within time series. Indeed, thanks to the temporeal means, the number of involved images doesn’t matters because only two different means are compared to detect the changes (for example, the geometric and quadratic means). Thus, large data volumes can be processed easily, since the useful information is condensed within the temporal means. The only change detection parameter is the false alarm rate that will be MIMOSA method are very good compared to other methods. Several tests have been performed in order to quantify the robustness of the method facing the most common problems, like image misregistration or radiometric calibration errors. A graphical user interface has also been developed for MIMOSA, including many useful tools to prepare and process SAR data, but also several analyse tools.

京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第18937号 / 工博第3979号 / 新制||工||1613(附属図書館) / 31888 / 京都大学大学院工学研究科都市社会工学専攻 / (主査)教授 松岡 俊文, 教授 田村 正行, 教授 小池 克明 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

No description available.

Deviations from a nominal, straight-line flight path of a synthetic aperture radar (SAR) lead to inaccurate and defocused radar images. This thesis is an investigation into the improvement of the motion compensation algorithm created for the BYU inteferometric synthetic aperture radar, YINSAR. The existing BYU SAR processing algorithm produces improved radar imagery but does not fully account for variations in attitude (roll, pitch, yaw) and does not function well with large position deviations. Results in this thesis demonstrate that a higher order motion compensation algorithm is not as effective as using a segmented reference track, coupled with the current lower-order motion compensation algorithm. Attitude variations cause a Doppler shift and are corrected by limiting the processed azimuth bandwidth or by reversing the frequency shift with a range-dependent filter. Another important area considered is the effects of motion compensation on interferometry. When performing interferometry with YINSAR, motion compensating both channels to a single track has two effects. First, the applied MOCO phase corrections remove the "flat-earth" differential phase from the interferogram. Second, range resampling coregisters the two images. All of these changes have helped to improve YINSAR imagery.

Brigham Young University has developed a low-cost, light-weight, and low power consumption SAR for flight on a small unmanned aerial vehicle (UAV) at low altitudes. This micro-SAR, or uSAR, consumes only 18 watts of power, ideal for application on a small UAV. To meet these constraints, a linear frequency modulation-continuous wave (LFM-CW) transmit signal is utilized. Use of an LFM-CW signal introduces some differences from the typical strip map SAR processing model that must be addressed in signal processing algorithms. This thesis presents a derivation of the LFM-CW signal model and the associated image processing algorithms used for the uSAR developed at BYU. A data simulator for the BYU LFM-CW SAR is detailed and results are provided for the case when the simulated data are processed using the uSAR algorithms. Data processing schemes are discussed, including compression, receive signal phase detection, interference filtering and auto-focusing. Finally, data collected from the instrument itself are processed and presented.

In this thesis, the Omega-k algorithm used for processing stripmap synthetic aperture radar (SAR) data is explored in detail. While the original Omega-k algorithm does not achieve the same SNR as a matched filter, a modification is presented which enables the algorithm to nearly achieve that SNR. It is shown that the focused point spread function obtained when the Omega-k algorithm is used differs in important ways from the output of a modified version of the matched filter. Spread out sidelobes and a stretched mainlobe are observed when the data is processed by the Omega-k algorithm. These differences may increase the potential interference between some nearby scatterers; however, the amplitude of the resulting sidelobes is lower than that observed for the matched filter, and the potential interference between other nearby scatterers is reduced. The details of a discrete implementation of the algorithm are also presented. Two methods for mixing the frequency domain signal to baseband are compared, and one is shown to potentially reduce the required accuracy of the interpolation kernel. Finally, the errors associated with the key approximation used by the algorithm are explored through simulation, and it is shown that the approximation is sufficiently accurate for a particularly demanding configuration.

Factorized backprojection is a processing algorithm for reconstructing images from data collected by synthetic aperture radar (SAR) systems. Factorized backprojection requires less computation than conventional time-domain backprojection with little loss in accuracy for straight-line motion. However, its implementation is not as straightforward as direct backprojection. Further, implementing an azimuth window has been difficult in previous versions of factorized backprojection. This thesis provides a new, easily parallelizable formulation of factorized backprojection designed for both pulsed and linearly frequency modulated continuous wave (LFM-CW) stripmap SAR data. A method of easily implementing an azimuth window as part of the factorized backprojection algorithm is introduced. The approximations made in factorized backprojection are investigated and a detailed analysis of the corresponding errors is provided. We compare the performance of windowed factorized backprojection to direct backprojection for simulated and actual SAR data.

Synthetic aperture radar (SAR) systems are versatile, high-resolution radar imagers useful for providing detailed intelligence, surveillance, and reconnaissance, especially when atmospheric conditions are non-ideal for optical imagers. However, moving targets in SAR images are smeared. Along-track interferometry is a commonly-used method for extracting the motion parameters of moving targets but requires a dual-aperture SAR system, which may be power- size- or cost-prohibitive. This thesis presents a method of estimating target motion parameters in single-channel SAR data given geometric target motion constraints. I test this method on both simulated and actual SAR data. This estimation method includes an initial estimate, computation of the SAR ambiguity function, and application of the target motion constraints to form a focused image of the moving target. The constraints are imposed by assuming that target motion is restricted to a road. Finally, I measure its performance by investigating the error introduced in the motion estimates using both simulated and actual data.

The Synthetic Aperture Radar (SAR) algorithm processes multiple radar returns from the target space to generate a single high-resolution image. Targets moving through the target space during the capture sequence will appear distorted on the final image. In addition, there is no velocity information that is calculated as part of the processing. The objective of this thesis is to develop techniques to determine the azimuth and range velocities of moving objects in the target space in the early stages of SAR processing. The typical SAR processing steps are Range Compressed, Range Doppler, and final image generation. The range velocity of a target can be determined after the Range Compression stage, and the azimuth velocity can be determined after the Range Doppler image is created. Calculating the velocity of a target without performing all the steps of the SAR process allows such information can be obtained quicker than the final image. This work is done as part of Cal Poly’s SAR Automatic Target Recognition (ATR) project, sponsored by Raytheon Space and Airborne Systems Division and headed by Professor John Saghri. The simulations performed as part of this thesis are done in a MATLAB simulation environment implementing a two-dimension SAR target space, first introduced in Brian Zaharris’ thesis. This work has expanded on this environment by introducing point target azimuth and range velocity detection.