L. M. Novak, M. C. Burl, and W. W. Irving. Optimal polarimetric processing for enhanced target detection. IEEE Trans. AES, 29:234--244, January 1993.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....for post processing the resulting (complex) SAR imagery, to produce classification of pixels into regions with specific properties of interest. One class of algorithms seeks to adaptively model the background clutter, and detect targets as deviations from the model predictions (e.g. 3] 4] [5], 6] Another class of algorithms uses prior knowledge to construct a target signature, and compares the image under test to the signature (e.g. 7] 8] 9] 10] 11] 12] The latter comparison may be accomplished in a variety of ways, ranging from optimized neural networks (e.g. 11] ....

....cuts through imaged mines, for horizontally and vertically polarized radiation. The magnitude of the complex image is plotted. a conclusive statement on this score. The standard algorithm for combining radar data collected with different polarizations is the polarimetric whitening filter (PWF) [5]. Polarimetric whitening is described below, and its performance is compared with that of the algorithm developed here. The important difference between PWF and the algorithm developed here is that PWF does not use information about the target signature. Hence, the PWF is optimal only if nothing ....

[Article contains additional citation context not shown here]

L. M. Novak, M. C. Burl, and W. W. Irving, "Optimal polarimetric processing for enhanced target detection," IEEE Trans. on Aero and Elect Syst., 29, 1, Jan. 1993.

....[2] y V V [2] 1 C A ; 0 B y HH [7] y HV [7] y V V [7] 1 C A (3.1) where each 3 tuple contains the pixel s multiple polarity values for a specific subaperture. To simplify the analysis of sub aperture trajectories, the polarimetric whitening filter (PWF) developed by Novak, et al. [9] is applied to each trajectory. The PWF uses the multiple polarity values in each 3 tuple to estimate the overall radar return intensity at each sub aperture in the trajectory. The effect of anisotropic radar returns on individual sub aperture trajectories is shown in Figure 3, which compares the ....

....for use 30 Target Clutter Test Cell Distort Clutter Estimation) Window Reference CFAR (So Target Does Not Guard Area M Cells M Cells Figure 12: Diagram showing how the CFAR detector uses a ring of reference pixels to estimate the local clutter mean and variance. on SAR imagery by Novak, et. al [18, 9]. Although it is not a MASAR ATD algorithm, the CFAR detector proved better than several other conventional SAR ATD algorithms studied by Whitehead [6] Thus, CFAR detection results provide a good comparison between MASAR detection performance and conventional SAR detection performance. As shown ....

L. M. Novak, M. C. Burl, and W. W. Irving, "Optimal polarimetric processing for enhanced target detection," IEEE Transact. on Aerospace & Electronic Sys., vol. 29, pp. 234--244, Jan. 1993.

....a good approximation for speckle [16] Our goal is to show that wavelet based noise reduction can be used for minimizing the effects of speckle when the observed image y is a digitized SAR image. Several algorithms, not based on wavelet theory, for reducing speckle in SAR images have been proposed [7, 34]. However, no single method has been found (possible with the exception of the polarimetric whitening filter (PWF) 34] which uses multiple observations to reduce speckle) that does a good job of removing speckle without a significant loss of image resolution or extensive knowledge of ground ....

....effects of speckle when the observed image y is a digitized SAR image. Several algorithms, not based on wavelet theory, for reducing speckle in SAR images have been proposed [7, 34] However, no single method has been found (possible with the exception of the polarimetric whitening filter (PWF) [34] which uses multiple observations to reduce speckle) that does a good job of removing speckle without a significant loss of image resolution or extensive knowledge of ground truth. The results as presented here are not new and have independently been reported by Moulin [31] and Guo et al. 21, ....

[Article contains additional citation context not shown here]

L. M. Novak, M. C. Burl, and W. W. Irving. Optimal polarimetric processing for enhanced target detection. IEEE Trans. AES, 29:234--244, January 1993.

....and recognition systems. It can be shown and simply verified by measurement that the WGN model is a good approximation for speckle [3] The goal is to minimize the effects of speckle when the observed image y is a digitized SAR image. There are several algorithms for removing speckle in SAR images [1, 6]. However, no single method has been found that does a good job of removing speckle without a significant loss of image resolution. A classic measure of speckle is the standard deviation to mean (s=m) ratio [3, 1] and log standard deviation [6] Table 1 shows those two values for original and ....

....algorithms for removing speckle in SAR images [1, 6] However, no single method has been found that does a good job of removing speckle without a significant loss of image resolution. A classic measure of speckle is the standard deviation to mean (s=m) ratio [3, 1] and log standard deviation [6]. Table 1 shows those two values for original and processed images for two typical regions. In Fig. 1 we have plotted the original and the processed version of a SAR image Table 1: Standard deviation to mean (std=m) and log standard deviation (log std) for SAR HH clutter data. The table givs a ....

L. M. Novak, M. C. Burl, and W. W. Irving. Optimal polarimetric processing for enhanced target detection. IEEE Trans. AES, 29:234--244, January 1993.

....return as a function of sub aperture angle. Figure 2 shows intensity trajectories for a target pixel, a tree pixel, and a grass pixel. Here, the overall intensity has been estimated from the complex HH, HV , and V V returns using a polarimetric whitening filter (PWF) developed by Novak, et al.[2] The target pixel trajectory shows two definite peaks and clearly has more anisotropic scattering than the tree and grass pixels. Since the two peaks in the target pixel trajectory are separated by only 60 ffi , they cannot both result from the entire target being broadside to the radar at that ....

....target pixel labeled according to which state is most likely to produce each sub aperture return. SAR two parameter CFAR detection. The two parameter CFAR detector was first developed by Goldstein for use on radar range profiles and was later extended for use on SAR imagery by Novak, et al.[6, 2] Although it is not a MASAR ATD algorithm, the CFAR detector proved better than several other conventional SAR ATD algorithms studied by Whitehead. 7] Thus, we include CFAR detection results to compare MASAR detection performance with conventional SAR detection performance. The CFAR detector uses ....

L. M. Novak, M. C. Burl, and W. W. Irving. Optimal polarimetric processing for enhanced target detection. IEEE Transact. on Aerospace & Electronic Sys., 29(1):234--244, January 1993.

....of the window (presumed clutter) If this is larger than the value of the test pixel by a certain amount, the pixel is selected as target, otherwise it is selected as clutter. Better results are obtained when the decision includes not only the mean c , but the standard deviation oe c as well[56, 55]: f(i; j) K oe c c = g(i; j) 1 (target) f(i; j) K oe c c = g(i; j) 0 (clutter) 2) Figure 2 illustrates this concept. Several variations of this idea have been proposed. Other approaches include the use of linear discriminant function of local features to select interest ....

L. Novak, M. Burl, and W. Irving, "Optimal Polarimetric Processing for Enhanced Target Detection", IEEE Transactions on Aerospace and Electronic Systems, Vol. 29, No. 1, pp. 234244, Jan. 1993.

....and ground clutter HMMs and N 0 is the total number of states in all three HMMs. B. Computations Required by Two parameter CFAR Detection The two parameter CFAR detector was first developed by Goldstein for use on radar range profiles and was later extended for use on SAR imagery by Novak, et. al [5, 6]. Although it is not a MASAR ATD algorithm, the CFAR detector proved better than several other conventional SAR ATD algorithms studied by Whitehead [7] Thus, CFAR detection results provide a good comparison between MASAR detection performance and conventional SAR detection performance. As shown ....

L. M. Novak, M. C. Burl, and W. W. Irving, "Optimal polarimetric processing for enhanced target detection," IEEE Transact. on Aerospace & Electronic Sys., vol. 29, pp. 234--244, Jan. 1993.

....we already have a digitized speckled image. Dewaele et al. 17] compared several speckle reduction techniques, including Lee s statistical filter, the sigma filter, and Crimmins geometric filter. These methods achieve moderate speckle reduction, but smooth out sharp features in the image. Novak [63] derived a polarimetric whitening filter (PWF) for fully polarimetric SAR data. However, this method does not utilize spatial correlation only the correlation across polarizations is used. We propose a novel speckle reduction method based on thresholding the wavelet coefficients of the ....

....performance: Standard deviation to mean ratio (s m) The quantity s=m(both in power) is a measure of image speckle in homogeneous region [30, 1, 17, 51] We computed the s=m ratio for each type of clutter region to quantify the speckle reduction capacity of our algorithm. Log standard deviation [63]: The standard deviation of the clutter data (in dB) This is an important quantity that directly affects the target detection performance of a standard two parameter constant false alarm rate detector (CFAR) algorithm. Target to clutter ratio(t c) The difference between the target and clutter ....

[Article contains additional citation context not shown here]

L. M. Novak, M. C. Burl, and W. W. Irving. Optimal polarimetric processing for enhanced target detection. IEEE Trans. AES, 29:234--244, January 1993.

....is thus 45 degrees. Since 3 of the targets are not far enough from the image edges for all their target pixels to be shown in all 7 aspect angle images, this rotation leaves us 8 targets for analysis. To analyze the full polarity images, we implemented a Polarimetric Whitening Filter (PWF) [1, 2, 3] to synthesize speckle reduced images from the full polarity returns. The PWF algorithm processes the complex valued HH, HV, and VV images and returns a full resolution backscattering intensity image at each aspect angle, or 7 images in all. According to a heuristic argument proposed by several ....

....speckle reduced images from the full polarity returns. The PWF algorithm processes the complex valued HH, HV, and VV images and returns a full resolution backscattering intensity image at each aspect angle, or 7 images in all. According to a heuristic argument proposed by several researchers [4, 1], the distribution of pixel intensities for SAR imagery which has been PWF processed resembles (for simplicity) a Log normal distribution. Thus, converting the PWF image intensity x at each pixel to the dB intensity y using the equation y = 20 log 10 x should transform the target pixel ....

[Article contains additional citation context not shown here]

L. M. Novak, M. C. Burl, and W. W. Irving, "Optimal polarimetric processing for enhanced target detection," IEEE Transact. on Aerospace & Electronic Sys., vol. 29, pp. 234--244, Jan. 1993.

....with the computation of the threshold. The Order Statistic CFAR (OSCFAR) technique works better in such situations. An ordered statistic of the reference cells is used to compute the threshold in an OSCFAR detector. Two parameter CFAR detectors for multi polarization data are considered in [6] and [7] CFAR techniques can be used as a first step in a feature extraction scheme for SAR imagery. In this paper, we demonstrate the application of CACFAR and OSCFAR techniques to detect targets in high resolution SAR terrain images. Other techniques like, GreatestOf CFAR and Censored CFAR, do ....

....and material properties of the target, target orientation with respect to the radar etc. For detecting point targets using CFAR, a single polarization channel seems to suffice. Both a solid reference window, around the cell under test, and a hollow window, suitably separated from the test cell [6], were used to estimate the clutter parameters. The results of applying the various CFAR techniques to two test images (Figure 1) are shown in Figures 2 5. Figure 1 (a) is an image of a road bridge, where the metal guard rail on the bridge and other strong reflectors serve as targets. Figure 1 (b) ....

L. M. Novak, M. C. Burl, W. W. Irving and G. J. Owirka, "Optimal polarimetric processing for enhanced target detection," IEEE National Telesystems Conf. Proceedings, pp. 69-75, 1991.

.... s 2 I N 1 N 2 h 1 h 1 T s 2 I N 1 N 2 h 2 h 2 T s 2 I N 1 N 2 h 1 h 2 T s 2 I N 1 N 2 h 2 = s 2 h 1 T h 1 s 2 h 1 T h 2 s 2 h 2 T h 1 s 2 h 2 T h 2 s 2 h 1 0 0 s 2 h page 19 of 27 with a polarimetric whitening filter (PWF) (Novak et al., 1993) and then logarithmically scaled to units of dBsm (dB square meters) prior to being used for our experiments. The data used was collected at a depression angle of 20 degrees, that is the radar antenna was directed 20 degrees down from the horizon. ISAR images were extracted in the range 5 to 85 ....

Novak, L. M., M. C. Burl, and W. W. Irving (1993); Optimal Polarimetric Processing for Enhanced Target Detection, IEEE Transactions on Aerospace and Electronic Systems, Vol. 29, p. 234.

....N 1 N 2 W 2 3 3 W 3 1 3 g ( 3.0 EXPERIMENTAL RESULTS For these experiments we used vehicle data from the TABILS 24 ISAR data set. The radar used for the data collection is a fully polarimetric, K a band radar. The ISAR imagery was processed with a polarimetric whitening filter (PWF) [8] and then logarithmically scaled to dBsm prior to being used for our experiments. The data used was collected at a depression angle of 20 degrees. ISAR images were extracted in the range 32 to 45 degrees azimuth in increments of 0.125 degrees. This resulted in 100 ISAR images (50 training, 50 ....

Novak L.M., et al; Optimal Polarimetric Processing for Enhanced Target Detection, IEEE Transactions on Aerospace and Electronic Systems, Vol. 29, p. 234, 1993.

....sensor, which is unlike optical or infrared sensors. The shortcoming is the lack of resolution due to the wavelength and the intrinsic noise produced by the image formation which is called speckle. A useful improvement in SAR processing was the invention of the polarimetric whitening filter or PWF [6,7]. A cell in fully polarimetric SAR is represented by a 8 component vector x formed by four (complex) polarized measurements (VV, HV, HH, VH) where VV stands for vertical send vertical receive, HV horizontal send vertical receive, etc. PWF is a linear projection of the fully polarimetric data x ....

Novak L.M., Burl M.C., Irving W.W., "Optimal polarimetric processing for enhanced target detection", IEEE Trans. Aero. Elect. Systems, vol 29, #1, pp 234-244, 1993.

....coefficients [2] The wavelet based algorithm promises to have great advantages over other known methods for speckle removal. x log j Deltaj DWT Soft Thresholding IDWT Figure 1. SAR speckle reduction via wavelet thresholding. While the well known polarimetric whitening filter (PWF) [7] is based on exploiting polarimetric correlation without loss of resolution, the wavelet based speckle removal algorithm exploits spatial correlation with essentially no loss of resolution. Qualitatively there are two features of the wavelet based method that are important: i) no spurious ....

L. M. Novak, M. C. Burl, and W. W. Irving. Optimal polarimetric processing for enhanced target detection. IEEE Trans. AES, 29:234--244, January 1993. Submitted to ATRSTC IV, Monterey, CA - '94

....17, 16, 8, 21, 3] Applied to SAR the new algorithm promises to be superior to the classical wavelet denoising algorithm. This research was supported in parts by ARPA, TI, BNR and the Alexander von Humboldt Foundation While the well known polarimetric whitening filter (PWF) due to Novak [22] is based on exploiting polarimetric correlation in fully polarimetric SAR data without loss of resolution, the wavelet based algorithm for speckle reduction exploits spatial correlation with essentially no loss of image resolution. Furthermore, when combining the wavelet based method with the PWF ....

....model is a good approximation for speckle[10] when considering the SAR intensity magnitude image (e.g. the dB image) The goal is then to minimize the effects of speckle when the observed complex image y is a digitized SAR image. There are several algorithms for reducing speckle in SAR images [5, 22]. However, until recently [20, 14] no method existed that could significantly reduce speckle for single channel SAR images without loss of image resolution (e.g. local averaging) The wavelet based algorithms as described in the previous section and illustrated in Fig. 2, have been studied ....

[Article contains additional citation context not shown here]

L. M. Novak, M. C. Burl, and W. W. Irving. Optimal polarimetric processing for enhanced target detection. IEEE Trans. AES, 29:234--244, January 1993.

No context found.

L. M. Novak, M. C. Burl, and W. W. Irving. Optimal polarimetric processing for enhanced target detection. IEEE Trans. AES, 29:234--244, January 1993.

No context found.

L. M. Novak, M. C. Burl, and W. W. Irving, "Optimal polarimetric processing for enhanced target detection," IEEE Trans. on Aero and Elect Syst,vol. 29, no. 1, pp. 234-244, Jan. 1993.

No context found.

L. M. Novak, M. C. Burl, and W. W. Irving. Optimal polarimetric processing for enhanced target detection. IEEE Trans. Aero. Elec. Sys., 29(1):234--244, January 1993.

No context found.

L. M. Novak, M. C. Burl, and W. W. Irving, "Optimal polarimetric processing for enhanced target detection," IEEE Trans. Aero. Elec. Sys. 29, pp. 234--244, 1993.

No context found.

Novak et al., 1993 171 Novak, L. M., M. C. Burl, and W. W. Irving (1993); "Optimal polarimetric processing for enhanced target detection", IEEE Transactions on Aerospace and Electronic Systems, 29 (1): 234-243.

No context found.

L. M. Novak, M. C. Burl, and W. W. Irving. Optimal polarimetric processing for enhanced target detection. IEEE Trans. AES, 29:234--244, January 1993.

No context found.

L. M. Novak, M. C. Burl, And W. W. Irving, "Optimal Polarimetric Processing For Enhanced Target Detection," IEEE Trans. Aerospace And Electronic Systems, Vol. 29, p. 234, 1993.

No context found.

L. M. Novak, M. C. Burl, And W. W. Irving, "Optimal Polarimetric Processing For Enhanced Target Detection," IEEE Trans. Aerospace And Electronic Systems, Vol. 29, p. 234, 1993. Fisher and Principe7

No context found.

L. M. Novak, M. C. Burl, and W. W. Irving. Optimal polarimetric processing for enhanced target detection. IEEE Trans. AES, 29:234--244, January 1993.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC