David W. Paglieroni

The position-orientation masking approach to parametric search for template matching

A new search method over (x,y,/spl theta/), called position-orientation masking is introduced. It is applied to vertices that are allowed to be separated into different bands of acuteness. Position-orientation masking yields exactly one /spl theta/ value for each (x,y) that it considers to be the location of a possible occurrence of an object. Detailed matching of edge segments is performed at only these candidate (x,y,/spl theta/) to determine if objects actually do occur there. Template matching is accelerated dramatically since the candidates comprise only a small fraction of all (x,y,/spl theta/). Position-orientation masking eliminates the need for exhaustive search when deriving the candidate (x,y,/spl theta/). Search is guided by correlations between template vertices and distance transforms of image vertices. When a poor correlation is encountered at a particular position and orientation, nearby positions at that orientation and nearby orientations at that position are masked out. Position and orientation traversal are by quadrant and binary decomposition. >

Design considerations for image segmentation quality assessment measures

Abstract Factors to consider when designing quality assessment measures for image segmentation are discussed. Quality assessment requires one manually generated segmentation (for reference) plus computer-generated segmentations corresponding to different image segmentation algorithms or algorithm parameter settings. Since true pixel class assignments are seldom available, one must typically rely on a trained human analyst to produce a reference by using a mouse to draw boundaries of perceived regions on a digital image background. Different algorithms and parameter settings can be compared by ranking computed disparities between maps of computer-generated region boundaries and region boundaries from a common reference. Proximity-based association between two boundary pixels is discussed in the context of association distance. Motivated by the concept of phase-modulated signals, a penalty factor on the degree of association is then introduced as some non-negative power (phase modulation order) of the cosine of disparity in phase (boundary direction) between two boundary pixels. Families of matching measures between maps of region boundaries are defined as functions of associations between many pairs of boundary pixels. The measures are characterized as one-way (reflecting relationships in one direction between region boundaries from two segmentations) vs. two-way (reflecting relationships in both directions). Measures of inconsistency between perceived and computed matches of computer and manually generated region boundaries are developed and exercised so that effects of association distance, phase modulation, and choice of matching measure on image segmentation quality assessment can be quantified. It is quantitatively established that consistency can be significantly improved by using two-way measures in conjunction with high-order phase modulation and moderate association distances.

K-means reclustering: algorithmic options with quantifiable performance comparisons

This paper presents various architectural options for implementing a K-Means Re-Clustering algorithm suitable for unsupervised segmentation of hyperspectral images. Performance metrics are developed based upon quantitative comparisons of convergence rates and segmentation quality. A methodology for making these comparisons is developed and used to establish K values that produce the best segmentations with minimal processing requirements. Convergence rates depend on the initial choice of cluster centers. Consequently, this same methodology may be used to evaluate the effectiveness of different initialization techniques.

Imaging Modes for Ground Penetrating Radar and Their Relation to Detection Performance

The focus of this paper is an empirical study conducted to determine how imaging modes for ground penetrating radar (GPR) affect buried object detection performance. GPR data were collected repeatedly over lanes whose buried objects were mostly nonmetallic. This data were collected and processed with a GPR antenna array, system hardware, and processing software developed by the authors and their colleagues. The system enables GPR data to be collected, imaged, and processed in real-time on a moving vehicle. The images are focused by applying multistatic and synthetic aperture imaging techniques either separately or jointly to signal scans acquired by the GPR antenna array. An image-based detection statistic derived from the ratio of buried object energy in the foreground to energy of soil in the background is proposed. Detection-false alarm performance improved significantly when the detection algorithm was applied to focused multistatic synthetic aperture radar (SAR) images rather than to unfocused GPR signal scans.

Matching Random Tree Models of Spatio-Temporal Patterns to Tables or Graphs

The problem of matching random tree models of multi-component patterns to tables or graphs containing components extracted from diverse data sources is considered. We focus on bi-level trees whose branches emanate from one root node and terminate on different leaf nodes. Node and branch attributes are treated as random variables. Tree nodes represent pattern components of specified types that occur in tables or graphs to be searched. For each item in the table or graph with a type match to the tree root, there is a set of components from the table or graph that are candidate leaves for optimal matches to the tree model. We adopt a view of optimal matches to random tree models as minimum cost assignments of candidate leaves to tree branches. Model-based formulas are derived for computing costs associated with assignments of specific candidate leaf components from tables or graphs to specific tree branches. We specify an ontology suitable for dynamic geo-spatial query problems in which (1) tree nodes represent physical objects or events on the ground (buildings, roads, communication transmissions...), and (2) branch attributes characterize, with uncertainty, distance or time separations between components, and angles between links connecting components. Our approach is used to search very large images for specific types of buildings in probabilistically constrained spatial arrangements, with the goal of ranking model matches for efficient inspection by human analysts.

Directional distance transforms and height field preprocessing for efficient ray tracing

It is known that height field ray tracing efficiency can be improved if the empty space above the height field surface is first parameterized in terms of apex heights and opening angles of inverted cones of empty space whose vertical axes are regularly spaced. Once such a parameterization has been performed, rays can be traversed in steps across inverted cones of empty space rather than across successive height field grid cells. As the cone opening angles increase, ray tracing efficiency tends to improve because steps along rays across the inverted cones get longer. Circular horizontal cross-sections of an inverted cone can be divided into contiguous nonoverlapping sectors. Given that the inverted cones can contain nothing but empty space, the maximum possible opening angle within any such sector may significantly exceed the opening angle of the inverted cone. It is shown that ray tracing efficiency can be significantly improved by replacing the inverted cones of empty space with cones that have narrow sectors. It is also known that the parameters of the inverted cones can be derived from distance transforms (DTs) of successive horizontal cross-sections of the height field. Each cross-section can be represented as a 2D binary array, whose DT gives the distance from each element to the nearest element of value 1. DTs can be directionalized by requiring the element of value 1 closest to a given element to lie within a sector emanating from that given element. The parameters of inverted cones within specific sectors can be derived from such directional DTs. An efficient new algorithm for generating directional DTs is introduced.

Matching flexible polygons to fields of corners extracted from images

We propose a novel efficient method that finds partial and complete matches to models for families of polygons in fields of corners extracted from images. The polygon models assign specific values of acuteness to each corner in a fixed-length sequence along the boundary. The absolute and relative lengths of sides can be either constrained or left unconstrained by the model. Candidate matches are found by using the model as a guide in linking corners previously extracted from images. Geometrical similarity is computed by comparing corner acutenesses and side lengths for candidate polygons to the model. Photometric similarity is derived by comparing directions of sides in candidate polygons to pixel gradient directions in the image. The flexibility and efficiency of our method is demonstrated by searching for families of buildings in large overhead images.

Resolution analysis for Gradient Direction Matching of object model edges to overhead images

The problem of computer-assisted broad area search for specific objects of interest in overhead images is considered. To this end, we present a novel efficient Gradient Direction Matching (GDM) algorithm that matches gradient directions associated with object edges to pixel gradient directions (as opposed to image edges, which are less reliable). GDM seamlessly integrates information associated with pixel location and orientation in such a way that the FFT can be exploited for computational efficiency, and it inherently rejects background clutter. The effects of spatial resolution on GDM statistical performance are studied empirically with the goal of gaining insight into how far GDM computational cost can be reduced before matching performance becomes too severely compromised.

Phase Sensitive Cueing for 3D Objects in Overhead Images

A 3D solid model-aided object cueing method that matches phase angles of directional derivative vectors at image pixels to phase angles of vectors normal to projected model edges is described. It is intended for finding specific types of objects at arbitrary position and orientation in overhead images, independent of spatial resolution, obliqueness, acquisition conditions, and type of imaging sensor. It is shown that the phase similarity measure can be efficiently evaluated over all combinations of model position and orientation using the FFT. The highest degree of similarity over all model orientations is captured in a match surface of similarity values vs. model position. Unambiguous peaks in this surface are sorted in descending order of similarity value, and the small image thumbnails that contain them are presented to human analysts for inspection in sorted order.

Convergent coarseness regulation for segmented images

In segmentation of remotely sensed images, the number of pixel classes and their spectral representations are often unknown a priori. Even with prior knowledge, pixels with spectral components from multiple classes lead to classification errors and undesired small region artifacts. Coarseness regulation for segmented images is proposed as an efficient novel technique for handling these problems. Beginning with an over-segmented image, perceptually similar connected regions are iteratively merged using a method reminiscent of region growing, except the primitives are regions, not pixels. Interactive coarseness regulation is achieved by specifying the area alpha of the largest region eligible for merging. A region with area less than alpha is merged with the most spectrally similar connected region, unless the regions are perceived as spectrally dissimilar. In convergent coarseness regulation, which requires no user interaction, alpha is specified as the total number of pixels in the image, and the coarseness regulation output converges to a steady-state segmentation that remains unchanged as alpha is further increased. By applying convergent coarseness regulation to AVIRIS, IKONOS and DigitalGlobe images, and quantitatively comparing computer-generated segmentations to segmentations generated manually by a human analyst, it was found that the quality of the input segmentations was consistently and dramatically improved