Fred W. DePiero

3-D Computer Vision Using Structured Light: Design, Calibration, and Implementation Issues

Abstract Structured light (SL) sensing is a well-established method of range acquisition for computer vision. We provide thorough discussions of design issues, calibration methodologies, and implementation schemes for SL sensors. The challenges for SL sensor development are described and a range of approaches are surveyed. A novel SL sensor, PRIME, the PRofile Imaging ModulE recently has been developed and is used as a design example in the detailed discussions.

Graph matching using a direct classification of node attendance

Abstract An algorithm has been developed that finds isomorphisms between both graphs and subgraphs. The development is introduced in the object recognition problem domain. The method isolates matching subgraphs, finds a node-to-node mapping and reorders nodes, thus permitting a direct comparison to be made between the resultant graphs. The algorithm is of polynomial order. It yields approximate results, maintaining a performance level for subgraph isomorphisms at or above 95% under a wide variety of conditions and with varying levels of noise. The performance on the full size comparisons associated with graph isomorphisms has been found to be 100 100 , also under a variety of conditions. Performance metrics, methods of testing and results are presented.

Position and orientation tracking system

A position and orientation tracking system presents a laser scanning appaus having two measurement pods, a control station, and a detector array. The measurement pods can be mounted in the dome of a radioactive waste storage silo. Each measurement pod includes dual orthogonal laser scanner subsystems. The first laser scanner subsystem is oriented to emit a first line laser in the pan direction. The second laser scanner is oriented to emit a second line laser in the tilt direction. Both emitted line lasers scan planes across the radioactive waste surface to encounter the detector array mounted on a target robotic vehicle. The angles of incidence of the planes with the detector array are recorded by the control station. Combining measurements describing each of the four planes provides data for a closed form solution of the algebraic transform describing the position and orientation of the target robotic vehicle.

On the design and development of a human-robot synergistic system

Abstract A great variety of robotic systems have been developed over the past decades, varying from fully human-controlled teleoperated systems, to fully motion-programmed, industrial-type devices, to sensor-equipped ‘intelligent’ machines with autonomous perception and reasoning capabilities. All these systems, however, have very static roles in supporting humans, in the sense that their capabilities are limited to very narrow functions or tasks domains and they cannot really adapt to changing environments where entirely new tasks may be necessary. By considering these systems as part of overall human-machine systems, interesting adaptation capabilities can be envisioned for the machine through the sharing of control and experiences, and observation of the human actions. This paper proposes a general conceptual architecture for enhancing the synergy between humans and machines in such cooperative systems, and investigates an implementation of the concept to the synergistic integration of teleoperative and autonomous operation of a manipulator arm for the execution of sequential tasks. The proposed architecture reflects a unique blend of many disciplines of artificial or machine intelligence and the functioning of its major modules is discussed. The paper then focuses on initial implementations of the Job Planner, Task Allocator, and Execution Monitor modules to discuss simple proof-of-principle experiments illustrating the coalescence of these modules needed to achieve skill-based dynamic task planning, allocation, and execution.

NetExam: a Web-based assessment tool for ABET2000

NetExam is a Web-based testing engine. In addition to automated testing and grading capabilities, NetExam computes statistics that are tied directly to program outcomes, for ABET2000 assessment purposes. NetExam provides advantages over scantron-style testing as it also presents statistics on program outcomes on the web. This facilitates review and black board-style comments by program constituents. Also, all of the exam generation, grading, statistics and program assessment features are integrated into the web-based system. First usage is planned for the 2001-2002 academic year.

Fast landmark-based registration via deterministic and efficient processing, some preliminary results

Preliminary results of a method for range view registration are presented. The method incorporates the LeRP algorithm, which is a deterministic means to approximate subgraph isomorphisms. Graphs are formed that describe salient scene features. Graph matching then provides the scene-to-scene correspondence necessary for registration. A graphical representation is invariant with respect to sensor standoff. Test results from real and synthetic images indicate that a reasonable tradeoff between speed and accuracy is achievable. A mean rotational error of /spl sim/1 degree was found for a variety of test cases. Mean compute times were found to be better than 2 Hz, with image sizes varying from 128/spl times/200 to 240/spl times/320. These tests were run on a 900 MHz PC The greatest challenge to this approach is the stable localization and invariant characterization of image features via fast, deterministic techniques.

Deterministic Surface Registration at 10Hz Based on Landmark Graphs With Prediction

Landmark graphs provide a means for surface registration, based on determining subgraph isomorphism tofind scene-to-scene correspondences. Surface data used herein included both range and colour imagery. Images were acquired of a static scene from a moving sensor. The continuous motion allowed the sensor position to be predicted, which helped stabilize graph formation. Landmarks were determined using the KLT corner detector. Graph structure was established using nodes (landmarks) and edges that agreed well with predicted locations. Subgraph matching was approximated using the LeRP algorithm. A 6 DOF rigid transformation including translation and rotation was found via Horn’s method. Test results on real and synthetic images indicate that a substantial speed improvement is possible, with greater determinism than ICP, while maintaining accuracy. Tests incorporated relatively large image displacements, spanning up to 30% of the sensor FOV for the image stream. Mean absolute errors remained under 0.8% FOV. Mean compute rates were ≈ 10 Hz with standard deviation ranging 6-9%, for an image size of 200x200. Tests were run on a 900 MHz PC. 141 test trials are reported, with comparisons against a fast version of ICP.

Illuminating electronics problem solving with the Cal Poly MoHat technique

This paper describes a Cal Poly version of circuit analysis to assist both students and instructors. The primary focus of the technique is helping students perform hand analysis, particularly the type of hand analysis that clarifies the students understanding of circuit operation in a manner that can benefit subsequent design decisions. The technique nicely complements graphical solution techniques such as load-line analyses and computer aided circuit simulation. MoHAT, an abbreviation of Model-Hypothesize-Analyze-Test, provides students with a roadmap to use, when analyzing even relatively complex circuits containing diodes and transistors. This paper emphasizes the pedagogy of MoHAT and describes a web-based software tool developed to help students visualize circuit operation.

Real-time range image segmentation using adaptive kernels and Kalman filtering

Segmentation is a fundamental process affecting the overall quality and utility of a machine vision system. Range profile tracking (RPT) is a systematic approach for stable, accurate and high speed segmentation of range images that is based on Kalman filtering. Tests of RPT have produced stable decompositions of second order surfaces bounded by jump and crease discontinuities, having a volumetric error of a few percent, in under 6 sec. for a wide variety of conditions. Results from over 900 tests on synthetic scenes and 150 real range images are presented.

Structural graph matching with polynomial bounds on memory and on worst-case effort

A new method of structural graph matching is introduced and compared against an existing method and against the maximum common subgraph. The method is approximate with polynomial bounds on both memory and on the worst-case compute effort. Methods work on arbitrary types of graphs and tests with strongly regular graphs are included. No node or edge colors are needed in the methods; the common subgraph is extracted based on structural comparisons only. Monte Carlo trials are benchmarked with 100% additional (clutter) nodes. Results are shown to be typically within 1-2 nodes of the maximum common subgraph. Over 7500 test trials are reported with graphs up to 100 nodes.