David Yoon

Spectral moving removal of non-isolated surface outlier clusters

In this paper, we propose a new algorithm for the fast removal of non-isolated surface outlier clusters. It consists of three basic components: (a) an intrinsic metric for detecting outliers on the basis of minimum variance principle; (b) bi-means clustering of a normalized histogram; (c) surface propagation for a geometric coherence check. The unique contributions of our approach include (a) a new idea of identifying non-isolated outlier clusters and linking the local spectral property to a global outlier removal process; (b) a modified data clustering scheme with a geometric coherence check. In comparison with existing algorithms, our algorithm is evaluated in terms of the quality and computational cost of outlier removal. Numerical experiments indicate the effectiveness of our approach in the aspects of convergence, accuracy, time and space efficiency.

Genetic search for optimally-constrained multiple-line fitting of discrete data points

Numerical computations were conducted under many conditions, including different chromosome lengths and search sizes.The good experimental performance of two schemes indicates that the genetic search proposed in this paper is a better way to locate exactly a constrained point at geometric discontinuity.Our algorithm is invariant of chromosome length and search space size to a certain degree. Advanced sensing technologies have produced a significant amount of discrete point data in the past decade. Measurement uncertainty frequently occurs at the geometric discontinuity of mechanical parts. In this paper, a genetic search algorithm is developed for optimally-constrained multiple-line fitting of discrete data points. It contains two important technical components: (a) constrained least-squares fitting of multiple lines, and (b) genetic search for optimal corner/edge points. The algorithm is designed for both two-dimensional and three-dimensional cases. Numerical experiments demonstrate the effectiveness of the proposed approach, compared to the conventional least-squares fitting method as well as exhaustive search method. A comparative study with a particle swarm method indicates that both the genetic search and particle swarm search produce similar results in terms of minimum fitting errors. It can be used for the effective determination of sharp edges or corners based on discrete data points measured for high-precision industrial inspection and manufacturing.

Application of Sweeping Techniques to Reverse Engineering

Abstract Reverse engineering refers to the process of developing a CAD model from a set of data collected from a physical part. Most reverse engineering applications assume that the parts are complete and the collected data provide complete information on them. In this paper, however, the authors consider the case where the parts are not complete due to breakage, and develop algorithms to infer the shapes of the original parts. Sweeping techniques are extensively used for the inference of shapes. Sweeping refers to the moving of a generator such as a curve, surface, or solid along a path . In this paper, the generator of sweep will be limited to a curve. There are three cases to consider: translational, rotational, and general. In each case, a part will be represented as a surface swept by a profile curve along a spine curve or a trajectory, both of which are modelled as NURBS curves. Once the original shape is inferred, the dimensions may be extracted from the original or mating part. The ultimate goal of this project is to develop a software system that infers the shape of the original part from the incomplete data so that the broken or worn-out piece may be designed and manufactured.

An Object-Oriented Approach to Computer Integrated Systems

In recent years computers have been incorporated into large scale systems such as nuclear plant, flight control, and manufacturing systems. Such Computer Integrated Systems (CIS) normally consist of heterogeneous subsystems. The integration of heterogeneous subsystems requires that the subsystems be portable, inter-operable, and integrable at both software and hardware levels so that the integrated system should function properly. Objects and nets are proposed as the atomic elements of CISs. An object is defined as a computational model of an arbitrary entity. Then three representation schemes of an object are introduced: algebraic, modular, and graphical. Two operations on objects,Composition andUnion, are introduced as means of combining two objects into a larger one. As an application of this approach, a Computer Integrated Manufacturing (CIM) system is represented as a network of objects.

Removal of surface artifacts of material volume data with defects

The three-dimensional defect distribution in material test specimens is a very important piece of information for us to understand the deformation and failure mechanism of materials. This distribution is sometimes complicated by the surface roughness of specimens in the defect detection of computed tomography data. In this paper, we proposed a new local differentiation algorithm to remove the surface artifacts caused by surface roughness in the defect detection of material specimens from computed tomography (CT) volume data. The accuracy of our method is compared with a traditional scan-line algorithm in terms of defect volume fraction measured in an independent scanning electron microscope (SEM) test. The experimental result indicates that our method is significantly better than the existing scan-line approach for predicting the defect volume fraction.

OBJ-NET: An Object Oriented System Design Tool

Computer-based Systems are used in a wide range of complex systems like automobile assembly lines, power plants, etc. These systems consist of heterogeneous subsystems having a multitude of interactions with other subsystems and environments. Designing software for such systems is a difficult task due to the inherent complexity both at conceptual and implementation levels. When developing a large system, the constituent subsystems may be developed separately. Those subsystems may work well on their own, but may fail to work together with other subsystems and fail to respond to interfacing events. This paper presents OBJ-NET, an object oriented system design tool, which provides a suitable framework for the specification and simulation of complex dynamic systems. The theoretical basis of OBJ-NET is from Category Theory and Petri Nets. OBJ-NET helps users to integrate complex systems by modeling a system as a hierarchy of components. It accepts the flow of control from one set of objects to another and simulates the flow of control among the various components of the system. This will allow a user to validate the correctedness of design and interactively change the design to refine the working of the system. This paper describes the various components of OBJ-NET in detail and illustrates the application of OBJ-NET using, as a working example, the specification and simulation of an Automotive Dashboard System.

A Set of Denoising Algorithms for 2-D Closed Curves

AbstractAccurate reduction of two-dimensional curve noises is crucial to metrology and reverse engineering. In this paper we propose a set of schemes for two-dimensional curve denoising. Our approach consists of three main steps: 1) feature-based pre-smoothing of noised polylines; 2) curve partitioning of polylines into different regions; 3) two hybrid denoising schemes of arbitrary polylines with noises. Numerical experiments indicate the effectiveness of our approach against existing methods.

Robust Plotting of Polar Algebraic Curves, Space Algebraic Curves, and Offsets of Planar Algebraic Curves

In this paper, based on modifications to the well-known subdivision scheme in combination with modified affine arithmetic method to guide the subdivision, we propose four robust and reliable algorithms for plotting polar algebraic curves, space algebraic curves and offsets of planar algebraic curves.

The Algebraic Framework for Object-Oriented Systems

Category Theory is introduced as the mathematical model for object-oriented systems which are distributed, heterogeneous, real-time, embedded, and open-ended. Each object can be represented as an algebra. A collection of algebras with morphisms form a category if they satisfy some conditions. After a brief introduction of categorical concepts which are needed to formulate the framework for object-oriented systems, they are explicated in terms of objects. Then some system design methodologies such as SADT, JSD, MASCOT 3, OOD, HOOD, MOON, ADM 3, and Petri nets are examined in the categorical framework and classified into four groups: functional, process-based, object-oriented, and net-based. Combining theoretical and practical results, the interactive system design tool OBJ-NET is briefly introduced.

Denoising of two-dimensional geometric discontinuities

Abstract Accurate reduction of two-dimensional curve noises at sharp edges, corners, or C0 discontinuous end points is crucial to the fast and accurate metrology, because more severe measurement errors of various sensors are frequently observed near sharp edges and corners of the objects to be measured. In this paper we explore three schemes (interior-point, polynomial extrapolation-and reflection-point methods) for handling C0 discontinuity. We also propose a new interior-point method for dealing with the accurate denoising of C1 discontinuity. Numerical experiments have been conducted for demonstrating the efficacy of our new approach in terms of denoising accuracy.