Doctoral Dissertations

Date of Award

8-1995

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Electrical Engineering

Major Professor

R. C. Gonzalez, D. W. Bouldin

Committee Members

R. V. Dubey, M. O. Pace, M. M. Trivedi

Abstract

Active sensing is the process of exploring the environment using multiple views of a scene captured by sensors from different points in space under different sensor settings. Active sensing can be used for the modeling of unknown objects or the recognition of objects in a scene. Applications of active sensing are numerous and can be found in the medical field (limb reconstruction), in archeology (bone mapping), in the movie and advertisement industry (computer simulations and graphics), in manufacturing (quality control), as well cis in the environmental industry (mapping of nuclear dump sites). In this work, the focus is on the use of a single vision sensor (camera) to per-form the volumetric modeling of an unknown object in an entirely autonomous fashion. The sensor's state parameters are purposefully changed according to the sensing strategy, using an intelligent scheduled choice. The camera moves to acquire the necessary information in two ways: (a) viewing closely each local feature of interest using 2-D data; and (b) acquiring global information about the environment via 3-D sensor locations and orientations. A single object is presented to the camera and an initial arbitrary image is acquired. A 2-D op-timization process is developed. It brings the object in the field of view of the camera, normalizes it by centering the data in the image plane, aligns the princi-pal axis with one of the camera's axes (arbitrarily chosen), and finally maximizes its resolution (according to a criterion that will be set later) for better feature extraction. The enhanced image at each step is projected along the correspond-ing viewing direction. The new projection is intersected with previously obtained projections for volume reconstruction. The 3-D model of the object is defined by an occupancy grid which specifies the points belonging to the object and those belonging to free space. During the global exploration of the scene, the current image as well as previous images are used to maximize the information in terms of shape irregularity as well as contrast variations. The scene on the borders of occlusion (contours) is modeled by partitioning the contour images and evaluating an entropy-based objective functional on each contour segment. This functional is optimized to determine the best next view, which is recovered by computing the pose of the camera. The camera moves to acquire an orthogonal view of the contour segment of maximum information content along the direction of its median and from a distance defined according to the current view. A criterion based on the minimization of the difference between consecutive volume updates is set for termination of the exploration procedure. These steps are integrated into the design of an off-line Autonomous Model Construction System AMCS, based on data-driven active sensing. The system operates autonomously with no human intervention and with no prior knowledge about the object. The results of this work are illustrated using computer simulation applied to intensity images rendered by ray-tracing software package.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS