In this work, we describe the compression of an image restoration neural network using principal component analysis (PCA). We compress the SRN-Deblur network that was developed by Tao et al.1 and we evaluate the deblurring performance at various levels of compression quantitatively and qualitatively. A baseline network is obtained by training the network using the GOPRO training dataset9. The performance of the compressed network is then evaluated when deblurring images from the Kohler8, Kernel Fusion13 and GOPRO datasets, as well as from a customized evaluation dataset. We note that after a short retraining step, the compressed network behaves as expected, i.e. deblurring performance slowly decreases as the level of compression increases. We show that the SRN-Deblur network can be compressed by up to 40% without significant reduction in deblurring capabilities and without significant reduction of quality in the recovered image.
KEYWORDS: Modulation, Receivers, Error analysis, Signal to noise ratio, Signal processing, Computer simulations, Data processing, Data communications, Transmitters, Data acquisition
This paper presents the results of using a constant modulus algorithm (CMA) to recover shaped offset quadrature-phase
shift keying (SOQPSK)-TG modulated data, which has been transmitted using the iNET data packet structure. This
standard is defined and used for aeronautical telemetry. Based on the iNET-packet structure, the adaptive block
processing CMA equalizer can be initialized using the minimum mean square error (MMSE) equalizer [3]. This CMA
equalizer is being evaluated for use on iNET structured data, with initial tests being conducted on measured data which
has been received in a controlled laboratory environment. Thus the CMA equalizer is applied at the receiver to data
packets which have been experimentally generated in order to determine the feasibility of our equalization approach, and
its performance is compared to that of the MMSE equalizer. Performance evaluation is based on computed bit error rate
(BER) counts for these equalizers.
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.
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).
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.
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.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.