Stephen C. Cain

Projection-based image registration in the presence of fixed-pattern noise

A computationally efficient method for image registration is investigated that can achieve an improved performance over the traditional two-dimensional (2-D) cross-correlation-based techniques in the presence of both fixed-pattern and temporal noise. The method relies on transforming each image in the sequence of frames into two vector projections formed by accumulating pixel values along the rows and columns of the image. The vector projections corresponding to successive frames are in turn used to estimate the individual horizontal and vertical components of the shift by means of a one-dimensional (1-D) cross-correlation-based estimator. While gradient-based shift estimation techniques are computationally efficient, they often exhibit degraded performance under noisy conditions in comparison to cross-correlators due to the fact that the gradient operation amplifies noise. The projection-based estimator, on the other hand, significantly reduces the computational complexity associated with the 2-D operations involved in traditional correlation-based shift estimators while improving the performance in the presence of temporal and spatial noise. To show the noise rejection capability of the projection-based shift estimator relative to the 2-D cross correlator, a figure-of-merit is developed and computed reflecting the signal-to-noise ratio (SNR) associated with each estimator. The two methods are also compared by means of computer simulation and tests using real image sequences.

Direct-Detection LADAR Systems

This text is designed to introduce engineers-in-training to the basic concepts and operation of 3D imaging LADAR systems. The book covers laser range equations; sources of noise in LADAR signals; LADAR waveforms; the effects of wavefront propagation on LADAR beams through optical systems and atmospheric turbulence; algorithms for detecting, ranging, and tracking targets; and comprehensive system simulation. Computer code for accomplishing the many examples appearing throughout the text is provided. Exercises appear at the end of each chapter, allowing students to apply concepts studied throughout the text to fundamental problems encountered by LADAR engineers. Also included is a CD-ROM with the MATLAB code from the examples.

Bound on range precision for shot-noise limited ladar systems

The precision of ladar range measurements is limited by noise. The fundamental source of noise in a laser signal is the random time between photon arrivals. This phenomenon, called shot noise, is modeled as a Poisson random process. Other noise sources in the system are also modeled as Poisson processes. Under the Poisson-noise assumption, the Cramer-Rao lower bound (CRLB) on range measurements is derived. This bound on the variance of any unbiased range estimate is greater than the CRLB derived by assuming Gaussian noise of equal variance. Finally, it is shown that, for a ladar capable of dividing a fixed amount of energy into multiple laser pulses, the range precision is maximized when all energy is transmitted in a single pulse.

Multichannel blind deconvolution of polarimetric imagery.

A maximum likelihood blind deconvolution algorithm is derived for incoherent polarimetric imagery using expectation maximization. In this approach, the unpolarized and fully polarized components of the scene are estimated along with the corresponding angles of polarization and channel point spread functions. The scene state of linear polarization is determined unambiguously using this parameterization. Results are demonstrated using laboratory data.

Flash light detection and ranging range accuracy limits for returns from single opaque surfaces via Cramer-Rao bounds

A Cramer-Rao lower bound on the range accuracy obtainable by a Flash light detection and ranging (LADAR) system receiving a return from a single surface in the instantaneous field of view of each detector is developed and verified with experimental data. The bound is compared to the performance of a new algorithm and that of a matched filter receiver by using both simulated and measured LADAR data. The simulated data are used to show that the estimator is nearly unbiased and efficient for systems that match the negative paraboloid model used in its derivation. It is found that the achievable range accuracy for the LADAR system and for the target geometry used to collect the measured data is of the order of 2.5 in. while the bound predicts a range accuracy limit of approximately 0.6 in.

Material Classification of an Unknown Object Using Turbulence-Degraded Polarimetric Imagery

In this paper, a material-classification technique using polarimetric imagery degraded by atmospheric turbulence is presented. The classification technique described here determines whether an object is composed of dielectric or metallic materials. The technique implements a modified version of the LeMaster and Cain polarimetric maximum-likelihood blind-deconvolution algorithm in order to remove atmospheric distortion and correctly classify the unknown object. The dielectric/metal classification decision is based on degree-of-linear-polarization (DOLP) maximum-likelihood estimates provided by two novel DOLP priors (one being representative of dielectric materials and the other being representative of metallic materials) developed in this paper. The DOLP estimate, which maximizes the log-likelihood function, determines the image pixels classification. Included in this paper is the review and modification of the LeMaster and Cain deconvolution algorithm. Also provided is the development of the novel DOLP priors, including their mathematical forms and the physical insight underlying their formulation. Lastly, the experimental results of two dielectric and metallic samples are provided to validate the proposed classification technique.

Maximum a posteriori image and seeing condition estimation from partially coherent two-dimensional light detection and ranging images

Recent developments in staring focal plane technology have spawned significant interest in the application of gated laser radar systems to battlefield remote sensing. Such environments are characterized by rapidly changing atmospheric seeing conditions and significant image distortion caused by long slant-range paths through the most dense regions of the atmosphere. Limited weight, space, and computational resources tend to prohibit the application of adaptive optic systems to mitigate atmospheric image distortion. We demonstrate and validate the use of a fast, iterative, maximum a posteriori (MAP) estimator to estimate both the original target scene and the ensemble-averaged atmospheric optical transfer function parameterized by Frieds seeing parameter. Wide-field-of-view sensor data is simulated to emulate images collected on an experimental test range. Simulated and experimental multiframe motion-compensated average images are deconvolved by the MAP estimator to produce most likely estimates of the truth image as well as the atmospheric seeing condition. For comparison, Frieds seeing parameter is estimated from experimentally collected images using a knife-edge response technique. The MAP estimator is found to yield seeing condition estimates within approximately 6% using simulated speckle images, and within approximately 8% of knife-edge derived truth for a limited set of experimentally collected image data.

Design of an image projection correlating wavefront sensor for adaptive optics

A new wavefront sensor data processing algorithm is described and analyzed. The wavefront sensing concept is similar to a Shack-Hartmann type wavefront sensor, but uses an image projection correlation algorithm as opposed to a centroiding approach to estimate optical tilt. This allows the wavefront sensor to estimate tilt parameters while guiding off of point sources and extended objects such as the surface granulation of the sun. The projection-based cross-correlating scheme differs from a 2-D correlation-based tilt estimation approach in that the images are vectorized on the focal plane array itself prior to readout. This on-chip preprocessing approach allows the wavefront sensor data to be compressed, which results in a large reduction in the amount of data read out of the focal plane array while maintaining the desired bandwidth of the adaptive optical system. An implementation of the projection-based wavefront sensor algorithm is described in detail, showing important signal processing steps on and off of the focal plane array of the sensor. The algorithm design is tested in simulation for speed and accuracy by processing simulated solar and astronomical datasets. Timing analysis is presented, which shows how the collection and processing of image projections is computationally efficient and lends itself to a wavefront sensor design that can produce both competitive speed and tilt estimation accuracy.

Sampling, radiometry, and image reconstruction for polar and geostationary meteorological remote sensing systems

In this paper, a Bayesian-based image reconstruction scheme is utilized for estimating a high resolution temperature map of the top of the earth’s atmosphere using the GOES-8 (Geostationary Operational Environmental Satellite) imager infrared channels. By simultaneously interpolating the image while estimating temperature, the proposed algorithm achieves a more accurate estimate of the sub-pixel temperatures than could be obtained by performing these operations independently of one another. The proposed algorithm differs from other Bayesian-based image interpolation schemes in that it estimates brightness temperature as opposed to image intensity and incorporates a detailed optical model of the GOES multi-channel imaging system. The temperature estimation scheme is compared to deconvolution via pseudo-inverse filtering using two metrics. One metric is the mean squared temperature error. This metric describes the radiometric accuracy of the image estimate. The second metric is the recovered Modulation Transfer Function (MTF) of the image estimate. This method has traditionally been used to evaluate the quality of image recovery techniques. It will be shown in this paper that there is an inconsistency between these two metrics in that an image with high spatial frequency content can be reconstructed with poor radiometric accuracy. The ramifications of this are discussed in order to evaluate the two metrics for use in quantifying the performance of image reconstruction algorithms.

IMPROVING THE SPACE SURVEILLANCE TELESCOPE'S PERFORMANCE USING MULTI-HYPOTHESIS TESTING*

The Space Surveillance Telescope (SST) is a Defense Advanced Research Projects Agency program designed to detect objects in space like near Earth asteroids and space debris in the geosynchronous Earth orbit (GEO) belt. Binary hypothesis test (BHT) methods have historically been used to facilitate the detection of new objects in space. In this paper a multi-hypothesis detection strategy is introduced to improve the detection performance of SST. In this context, the multi-hypothesis testing (MHT) determines if an unresolvable point source is in either the center, a corner, or a side of a pixel in contrast to BHT, which only tests whether an object is in the pixel or not. The images recorded by SST are undersampled such as to cause aliasing, which degrades the performance of traditional detection schemes. The equations for the MHT are derived in terms of signal-to-noise ratio (S/N), which is computed by subtracting the background light level around the pixel being tested and dividing by the standard deviation of the noise. A new method for determining the local noise statistics that rejects outliers is introduced in combination with the MHT. An experiment using observations of a known GEO satellite are used to demonstrate the improved detection performance of the new algorithm over algorithms previously reported in the literature. The results show a significant improvement in the probability of detection by as much as 50% over existing algorithms. In addition to detection, the S/N results prove to be linearly related to the least-squares estimates of point source irradiance, thus improving photometric accuracy.