Arlene Cole-Rhodes

Multiresolution registration of remote sensing imagery by optimization of mutual information using a stochastic gradient

Image registration is the process by which we determine a transformation that provides the most accurate match between two images. The search for the matching transformation can be automated with the use of a suitable metric, but it can be very time-consuming and tedious. We introduce a registration algorithm that combines a simple yet powerful search strategy based on a stochastic gradient with two similarity measures, correlation and mutual information, together with a wavelet-based multiresolution pyramid. We limit our study to pairs of images, which are misaligned by rotation and/or translation, and present two main results. First, we demonstrate that, in our application, mutual information may be better suited for sub-pixel registration as it produces consistently sharper optimum peaks than correlation. Then, we show that the stochastic gradient search combined with either measure produces accurate results when applied to synthetic data, as well as to multitemporal or multisensor collections of satellite data. Mutual information is generally found to optimize with one-third the number of iterations required by correlation. Results also show that a multiresolution implementation of the algorithm yields significant improvements in terms of both speed and robustness over a single-resolution implementation.

Mutual information as a similarity measure for remote sensing image registration

Feature-based matching is essential for attaining sub-pixel registration of remotely sensed imagery. In this work, we focus on two different similarity metrics which are used to match extracted features, correlation and mutual information. Although mutual information has been successfully applied to medical image registration, these metrics have not been systematically studied for remote sensing applications. This paper presents some first results in the comparison of correlation and mutual information, relative to their respective accuracy and response to noise. The study is performed using Landsat-TM data.

An adaptive blind equalizer with signal separation for a MIMO system transmitting QAM signals

In this work we propose a multiple-input multiple-output (MIMO) equalization scheme which uses the CMA+AMA equalizer in a modified multi-stage approach that simultaneously recovers all signal sources. The CMA+AMA equalizer minimizes a linear combination of the constant modulus algorithm (CMA) and alphabet-matched algorithm (AMA) cost functions, and this has been shown to be an effective equalizer for recovering QAM signals in the SISO and SIMO environment. The MIMO equalizer applies individual multiple-input single-output (MISO) equalizers to the received data stream, which is modified during the update process, to simultaneously remove the recovered sources using a channel estimator. By measuring the mean square error (MSE) and symbol error rates (SER) for the recovered sources, we show the effectiveness of this new equalizer in recovering all sources transmitted by the MIMO system.

SPC06-2: Blind Adaptive Equalization for QAM Signals Using an Alphabet-Matched Algorithm

This paper presents a new blind adaptive channel equalizer, which is based on two well-defined cost functions, the constant modulus algorithm (CMA) [1] and the alphabet- matched algorithm (AMA) [2]. The equalizer takes into account both the amplitude and the phase of the equalizer output and is adaptive based on the received channel data. The new equalization scheme presented here is compared to the multimodulus algorithm (MMA), which has been proposed in [3,4] as an improved equalizer for non-constant modulus signals, such as QAM. It is shown to provide better performance than the MMA in terms of both average mean- squared error (MSE) and average symbol error rate (SER) for the channels tested.

Performance of an adaptive blind equalizer for QAM signals

This paper presents a performance study of a blind adaptive channel equalization scheme. This scheme is based on a single cost function, which is created by combining two well-defined cost functions, the constant modulus algorithm (CMA) and the alphabet-matched algorithm (AMA). The combined cost function considers both the amplitude and the phase of the equalizer output, which allows for more efficient equalization of QAM signals. We present results that compare equalization of a 16-QAM signal using the new cost function with that obtained using the CMA cost function alone. The performance measures used for the comparison are the average mean-squared error (MSE) and the average symbol error rate (SER)

Multiresolution registration of remote sensing images using stochastic gradient

In image registration, we determine the most accurate match between two images, which may have been taken at the same or different times by different or identical sensors. In the past, correlation and mutual information have been used as similarity measures for determining the best match for remote sensing images. Mutual information or relative entropy is a concept from information theory that measures the statistical dependence between two random variables, or equivalently it measures the amount of information that one variable contains about another. This concept has been successfully applied to automatically register remote sensing images based on the assumption that the mutual information of the image intensity pairs is maximized when the images are geometrically aligned. The transformation which maximizes a given similarity measure has been previously determined using exhaustive search, but this has been found to be inefficient and computationally expensive. In this paper we utilize a new simple, yet powerful technique based on stochastic gradient, for the maximization of both similarity measures with remote-sensing images, and we compare its performance to that of the exhaustive search. We initially consider images, which are misaligned by a rotation and/or translation only, and we compare the accuracy and efficiency of a registration scheme based on optimization for this data. In addition, the effect of wavelet pre-processing on the efficiency of a multi- resolution registration scheme is determined, using Daubechies wavelets. Finally we evaluate this optimization scheme for the registration of satellite images obtained at different times, and from different sensors. It is noted that once a correct optimization result is obtained at one of the coarser levels in the multi-resolution scheme, then the registration process is much faster in achieving subpixel accuracy, and is more robust when compared to a single level optimization. Mutual information was generally found to optimize in about one third the time required by correlation.

Multiresolution image registration of remotely sensed imagery using mutual information

Wavelet-based image registration has previously been proposed by the authors. In previous work, maxima obtained from orthogonal Daubechies filters as well as from Simoncelli steerable filters were utilized and compared to register images with a multi-resolution correlation technique. Previous comparative studies between both types of filters have shown that the accuracy obtained with orthogonal filters seemed to degrade very quickly for large rotations and large amounts of noise, while results obtained with steerable filters appeared much more stable under these conditions. In other studies based on the use of mutual information for image registration, several authors have shown that maximizing mutual information enables one to reach sub-pixel registration accuracy. In this work, we are utilizing Simoncelli steerable filters to provide the basic data from which mutual information is maximized and we are applying this method to remotely sensed imagery.

Image Registration and Fusion Studies for the Integration of Multiple Remote Sensing Data

The future of remote sensing will see the development of spacecraft formations, and with this development will come a number of complex challenges such as maintaining precise relative position and specified attitudes. At the same time, there will be increasing needs to understand planetary system processes and build accurate prediction models. One essential technology to accomplish these goals is the integration of multiple source data. For this integration, image registration and fusion represent the first steps and need to be performed with very high accuracy. In this paper, we describe studies performed in both image registration and fusion, including a modular framework that was built to describe registration algorithms, a Web-based image registration toolbox, and the comparison of several image fusion techniques using data from the EO-1/ALI and Hyperion sensors

Multi-Sensor Registration of Earth Remotely Sensed Imagery

Assuming that approximate registration is given within a few pixels by a systematic correction system, we develop automatic image registration methods for multi-sensor data with the goal of achieving sub-pixel accuracy. Automatic image registration is usually defined by three steps; feature extraction, feature matching, and data resampling or fusion. Our previous work focused on image correlation methods based on the use of different features. In this paper, we study different feature matching techniques and present five algorithms where the features are either original gray levels or wavelet-like features, and the feature matching is based on gradient descent optimization, statistical robust matching, and mutual information. These algorithms are tested and compared on several multi-sensor datasets covering one of the EOS Core Sites, the Konza Prairie in Kansas, from four different sensors: IKONOS (4m), Landsat-7/ETM+ (30 m), MODIS (500 m), and SeaWIFS (1000m).

On the Performance of Equalization Techniques for Aeronautical Telemetry

This paper compares the performance of the zero forcing (ZF), minimum mean-squared error (MMSE), and MMSE-initialized CMA+AMA equalizers using SOQPSK-TG over multipath channels captured during channel sounding experiments at Edwards AFB. The ZF and MMSE equalizers are data-aided and are based on estimates of the channel conditions derived from the preamble and ASM bits included in each iNET packet. The simulated bit error rate results show that MMSE and CMA+AMA achieve similar performance and are both about 1 -- 3 dB better than the ZF equalizer. Comparison with the optimum AWGN detector shows that 4 -- 9 dB of link margin is required to account for the multipath interference.