The knowledge-directed approach to image interpretation, popular in the 1980's, sought to identify objects in unconstrained two-dimensional images and to determine the threedimensional relationships between these objects and the camera by applying large amounts of object- and domain-specific knowledge to the interpretation problem. Among the primary issues faced by these systems were variations among instances of an object class and differences in how object classes were defined in terms of shape, color, function, texture, size, and/or substructures. This paper

In 1983 Canny presented criteria for measuring the quality of edge detectors and derived an optimal FIRFilter for step edges by optimizing them. Four years later Deriche proposed an approach to edge detection based on Canny design utilizing IIR-Filters which can be implemented very efficiently recursively. Unfortunately, using the Deriche-Filter leads to a distortion of the amplitudes of the edges depending on their direction. In this paper, it will be shown that these distortions are systematic errors which can be eliminated by a simple modification of the edge detection procedure. Because of its obvious "kinship" to the Deriche-filter the Shen-filter has been included in the investigations.

The last decade has seen a dramatic expansion in the deployment of Unmanned Airborne Vehicles (UAVs) as witnessed by deployments to Bosnia, Afghanistan and Iraq. Many low-cost UAVs operate without redundant attitude sensors and are therefore highly vulnerable to the failure of such sensors. It is common for low-cost UAVs to carry a vision sensor as its primary payload. Given that a human pilot is trained to control an aircraft with respect to a visual horizon under Visual Meteorological Conditions (VMC), it is logical to suggest that a similar capability be developed for a UAV in the event of the failure of the primary attitude system. In addition, it potentially gives the capability to estimate the attitude of a gimballed camera, without specifically equipping the gimballed platform with an angular sensor. In this paper, we develop a method for estimating the flight critical parameters of pitch angle, roll angle and the three body rates using horizon detection and optical flow. We achieve this through the use of an image processing front-end to detect candidate horizon lines through the use of morphological image processing and the Hough transform. The optical flow of the image for each candidate line is calculated, and using these measurements, we are able to estimate the body rates of the aircraft. Using an Extended Kalman Filter (EFK), the candidate horizon lines are propagated and tracked through successive image frames, with statistically

In recent years, increasing e ort has gone into evaluating computer vision algorithms in general, and edge detection algorithms in particular. Most of the evaluation techniques use only a few test images, leaving open the question of how broadly their results can be interpreted. Our research tests the consistency of the receiver operating characteristic (ROC) curve, and demonstrates why consistent edge detector evaluation is difficult to a chieve. We show how easily the framework can be manipulated to rank any of three modern edge detectors in any order by making minor changes to the test imagery. We also note that at least some of the inconsistency is the result of the erratic nature of the algorithms themselves, suggesting that it is still possible to create better edge detectors.

This report describes a method to track moving non-rigid objects. The method is based on a comparison between a model and an image, the comparison function employed is the partial Hausdorff distance. The model and the image are binary elements extracted from a sequence of gray levels images. The target's motion in a two-dimensional representation (image) is decomposed into two parts: a two-dimensional motion, corresponding to the change in the target 's position in the image space and a two-dimensional shape change, corresponding to a new aspect of the target. 1 Introduction The target tracking problem has received a good deal of attention in the computer vision community over the last years. Several techniques have been reported in the literature, and a variety of features have been proposed to perform the tracking. Some methods attempt to track 3D objects by using multiple views [8]. Others use contours as features to perform the tracking [2], [7], [14]. Some contour metho...

Contents I. Introduction 1 II. The Need for Machine Learning in Vision 2 III.The Role of Machine Learning 3 IV.Machine Learning Techniques 5 A. Inductive Learning : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 7 V. Case study 10 A. System Background : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 11 B. System Analysis : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 13 C. Machine Learning Enhancements : : : : : : : : : : : : : : : : : : : : : : 18 D. Results : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 26 VI.Conclusion and Future Work 29 ii I. Introduction Both human vision and human learning are far from fully understood, yet machine imitation of these abilities has proven very useful. Research up to now has made much progress in tracing the desired result backward to t