Keith Price

Relaxation Matching Techniques-A Comparison

Many different relaxation schemes have been proposed for image analysis tasks. We have developed a general matching procedure for comparing semantic network descriptions of images, and we have implemented a variety of relaxation techniques. An automatic segmentation and description system is used to produce the image representations so that the matching procedures must cope with variations in feature values, missing objects, and possible multiple matches. This environment is used to test different relaxation matching schemes under a variety of conditions. The best performance (of those we compared), in terms of the number of iterations and the number of errors, is for the gradient-based optimization approach of Faugeras and Price. The related optimization approach of Hummel and Zucker performed almost as well, with differences primarily in difficult matches (i.e., where much of the evidence is against the match, for instance, poor segmentations). The product combination rule proposed by Peleg was extremely fast, indeed, too fast to work when global context is needed. The classical Rosenfeld, Hummel, and Zucker method is included for historical comparisons and performed only adequately, producing fewer correct matches and taking more iterations.

Structural Analysis of Natural Textures

Many textures can be described structurally, in terms of the individual textural elements and their spatial relationships. This paper describes a system to generate useful descriptions of natural textures in these terms. The basic approach is to determine an initial, partial description of the elements using edge features. This description controls the extraction of the texture elements. The elements are grouped by type, and spatial relationships between elements are computed. The descriptions are shown to be useful for recognition of the textures, and for reconstruction of periodic textures.

Semantic Description of Aerial Images Using Stochastic Labeling

This paper discusses the application of stochastic labeling to a general symbolic image description problem. A method used to compute initial likelihoods and compatibilities is described. It was derived from an earlier symbolic matching procedure, but was modified to provide the data needed for application of the labeling method. This labeling procedure differs from simpler ones, in that it minimizes a global criterion at each iteration. This technique is compared to other matching methods, and results on two scenes are presented.

Matching Segments of Images

This correspondence describes research in the development of symbolic registration techniques directed toward the comparison of pairs of images of the same scene to ultimately generate descriptions of the changes in the scene. Unlike most earlier work in image registration, all the matching and analysis will be performed at a symbolic level rather than a signal level. We have applied this registration procedure on several different types of scenes and the system appears to work well both on pairs of images which may be analyzed in part by signal based systems and those which cannot be so analyzed.

Automatic and interactive modeling of buildings in urban environments from aerial images

Automatically extracting object models from images is a complex task. We describe research in extracting 3D models of buildings from aerial images. This work has resulted in several related systems including assisted extraction (minimal manual interaction to guide automatic processing), automatic extraction with limited imagery and limited building models, and automatic extraction with very good imagery and digital elevation models and more complex building models. Some results are provided for the assisted system and one of the automatic systems.

Anything you can do, I can do better (no you can't)

Abstract Computer vision suffers from an overload of written information but a dearth of good evaluations and comparisons. This paper discusses why some of the problems arise and offers some guidelines we should all follow.

Locating Structures in Aerial Images

A technique for locating desired structures utilizing user specified information about properties of these structures and their relationships with other more easily extracted objects is described. An edge-based and region-based technique is used for scene segmentation. Experimental results of the processing of aerial pictures are presented.

The image understanding environment program

The history of the image understanding environment (IUE) project, a five-year program to develop a common software environment for the development of algorithms and application systems, is reviewed. An overview of some of the data structures that are currently evolving as a specification for the IUE is provided. The ultimate goal of the project is to provide the basic data structures and algorithms that are required to carry state-of-the-art research in image understanding.<<ETX>>

Hierarchical matching using relaxation

Abstract In this paper we outline the application of a general relaxation-based matching procedure to the problem of matching pairs of images and discuss the extension of the basic technique to matching hierarchical descriptions of scenes.

Image Segmentation: A Comment on ``Studies in Global and Local Histogram-Guided Relaxation Algorithms''

Theorem: The maximum number of trees in a forest derived from a quadtree that represents a square of dimension 2k X 2k is 4k-1, i.e., one-fourth the area of the square. Given a forest of quadtrees F, we can easily show how to reconstruct a quadtree. The reconstructed quadtree R(F) consists of real nodes (nodes in the forest) and virtual nodes (nodes that correspond to BAD nodes deleted while creating the forest). Since virtual nodes require no storage they are located by giving their coordinates. We denote the virtual node with coordinates (L, K) by v(L, K). The root of R(F) is either a real node (if the forest has one tree) or the virtual node v(l, 0). In either case we know its location and color in R(F). The offspring of any real node are found by following links. If v(L, K) is any virtual node, then its children in directions D & {NW, NE, SW, SE} have coordinates (L + 1, 4K + D), respectively. It is a simple matter of a table lookup to see if the offspring are real nodes or virtual nodes. Also, we can easily determine the color of a virtual node. It is GRAY if it has a descendant(s) in the table and WHITE otherwise. The check to see if the node has a descendant in the table may be performed efficiently if the table is stored in the left-to-right order produced by FOREST so we can apply the numerical test at the end of Section I. For example, we can easily establish that the virtual nodes v(3, 10) and v(3, 11) of the forest of Fig. 4 are WHITE because they lie between (in left-to-right order) the successive elements (3, 9) and (4, 48) in the table. Since the elements of the table are linearly ordered by this left-to-right order, we may perform a binary search to check on the color of a virtual node. If T represents a picture in a 2n X 2n grid, then such a search requires time 0(1og(number of trees in forest)) = 0(1og 4nl1) (by the theorem) = 0(n). For a real number x, let floor(x) denote the greatest integer less than or equal to x. The father of a virtual node v(L, K) is the virtual node v(L-1, floor(K/4)). If P is a real node, then either its father is the real node given by …