Rufus H. Cofer

Extended Hough transform for linear feature detection

Improving the accuracy of line segment detection reduces the complexity of subsequent high-level processing common in cartographic feature detection. We developed a new extension to the Hough transform and reported on its application to building extraction. We expanded the Hough space by a third parameter, the horizontal or vertical coordinate of the image space, to provide incremental information as to the length of the lineal feature being sought. Using this extended HT transform allowed us to more accurately detect the true length of a line segment. In addition, we used a Bayesian probabilistic approach to process our extended Hough space that further increased the accuracy of our extended Hough transform.

Linear feature detection using multiresolution wavelet filters

We detected roads in aerial imagery based on multiresolution linear feature detection. Our method used the products of wavelet coefficients at several scales to identify and locate linear features. After detecting possible road pixels, we used a shortest-path algorithm to identify roads. The multiresolution approach effectively increased the size of the region we examined when looking for possible road pixels and reduced the effect of noise. We found that our approach leads to an effective method for detecting roads in aerial imagery.

Lineal Feature Detection Using Multi-resolution Wavelet Filters

We detected road pixels in aerial imagery using a multiresolution, wavelet-based approach. Our method involved the description of a differential geometry approach for road seed pixel detection in terms of wavelet transforms. Using this approach allowed us to extend the differential geometry approach to incorporate multiple scales. We found that using multiple scales significantly reduced the number of potential false positives. Our approach seemed to work well with a computer-assisted approach where the “seed” or potential pixels of interest should have a high confidence level of being correct. We found that our approach led to an effective method for detecting roads in aerial imagery. Our method is general, and in principle could be applied to other filtering techniques besides the one used here.

Optical estimation of fractal dimension for image assessment

We modeled an optical system for estimation of the fractal dimension to provide a measure of surface roughness for an entire image and for image segmentation. Although the simulated optical result was similar to that calculated by digital techniques, both suffered from problems known to occur with estimating fractal dimension. Furthermore, the optical estimation did not have as good a resolution as that obtained with digital estimates due primarily to the limited dynamic range of the detector.

Multiresolution template matching using an optical correlator

Infrared imagery of 512 x 512 pixels were processed with 128 x 128 arrays by computer simulation of an optical correlator using various correlation filters. Pyramidal processing using binary phase-only filters (BPOFs), synthetic discriminant function (SDF) filters, and feature-based filters was used to process an entire image in parallel at different resolutions. Results showed that both SDF and feature-based filters were more robust to the effects of thresholding input imagery than BPOFs. The feature-based filters offered a range of performance by setting a parameter to different values. As the value of the parameter was changed, correlation peaks within the training set became more consistent and broader. The feature-based filters were more useful than both the SDF and simple BPOFs for recognizing objects outside the training set. Furthermore, the feature-based filter was more easily calculated and trained than an SDF filter.

Imagery chain assessment for feature extraction

It is shown that the image chain has important effects upon the quality of feature extraction. Exact analytic ROC results are given for the case where arbitrary multivariate normal imagery is passed to a Bayesian feature detector designed for multivariate normal imagery with a diagonal covariance matrix. Plots are provided to allow direct visual inspection of many of the more readly apparent effects. Also shown is an analytic tradeoff that says doubling background contrast is equal to halving sensor to scene distance or sensor noise. It is also shown that the results provide a lower bound to the ROC of a Bayesian feature detector designed for arbitrary multivariate normal distributions.

Extended Hough methodology in geo-spatial image exploitation

Hough transform theory provides a heuristically appealing approach toward finding lineal features in imagery. Unfortunately direct algorithmic implementation of its theory results in many practical problems. We provide two interlocking theoretical extensions to greatly enhances the Hough transforms ability to handle finite lineal features and allow directed search for parallel lines within the scene while balancing memory and computational complexity. Both extensions involve expansion of the Hough space concept to allow easier access to processed data for both dedicated silicon and general-purpose computer implementations.

Pixel-registered image fusion

One of the highest potential uses of image fusion is that of recognition of critical targets. The continuing image fusion question then is how to make optimal use of the often disparate forms of encountered image detail during fusion. Toward this end, many techniques have been advanced for fusion to a single viewable image. Fewer techniques have been suggested toward fusion with the goal of directly improving target detection or recognition. Based upon emerging trends in pixel accurate registration of images, we show the theoretical foundations required to optimally fuse target imagery for recognition. Results obtained can be applied to both the cases of automatic target recognition and image analysis.

Obscured object detection via Bayesian target modeling techniques

Underground objects are by nature often severely obscured although the general character of the intervening random media may be reasonably understood. The task of detecting these underground objects also implies that their exact location and or orientation is not known. To partially counter these difficulties, one may; however, be given a model of the target of interest, e.g. a particular tank type, a water pipe, etc. To set up a quality framework for solution of the above problem, this paper utilizes the paradigm of Bayesian decision theory that promises minimum error detection given that certain probability density functions can be found. Within this framework, mathematical techniques are shown to handle the uncertainties of target location and orientation, many of the random obscuration problems, and how to make best use of the target model. The approach taken can also be applied to other synergistic cases such as seeing through obscuring vegetation.

Statistical design of ternary phase-only filters for distortion invariance

Abstract Ternary phase-only filters that identified objects outside a training set in the presence of unknown or nonrepeatable distortions are developed. In the experiments, the statistical filters recognized objects within the same class and in the presence of noise better than another popular binary distortion-invariant filter design.