Kai H. Chang
Auburn University
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Kai H. Chang.
acm southeast regional conference | 2004
Yiming Ji; Kai H. Chang; Chi-Cheng Hung
Gabor filter is a widely used feature extraction method, especially in image texture analysis. The selection of optimal filter parameters is usually problematic and unclear. This study analyzes the filter design essentials and proposes two different methods to segment the Gabor filtered multi-channel images. The first method integrates Gabor filters with labeling algorithm for edge detection and object segmentation. The second method uses the K-means clustering with simulated annealing for image segmentation of a stack of Gabor filtered multi-channel images. Various experiments with real images demonstrate the effectiveness of these approaches.
international conference on quality software | 2007
Young Lee; Jeong Yang; Kai H. Chang
This paper presents an analysis of the evolution of an open source software system, JFreeChart, which is an open source charting library [1], based on its size, fan-in/out coupling, and cohesion metrics. We developed JamTool, a Java automated measurement tool [4] to obtain the metrics and to observe the quality change along the evolution of the twenty-two released versions of JFreeChart. The empirical study clearly indicates that there are positive relations between the number of classes and the fan-in/out coupling, and the added class group has better software quality than the removed class group. Moreover, the experimental results follow Lehmans 1st, 2nd, 6th laws 6th but against 7th law of software evolution.
acm southeast regional conference | 2000
Young Choon Lee; Kai H. Chang
In this paper, a quality model for object-oriented software and an automated metric tool, Reconfigurable Automated Metrics for Object-Oriented Software (RAMOOS) are proposed. The quality model is targeted at the maintainability and reusability aspects of software which can be effectively predicted from the source code. RAMOOS assists users in applying customized quality model during the development of software. In the beginning of adopting RAMOOS, a user may need to use his intuition to select or modify a system-recommended metric model to fit his specific software project needs. If the initial metrics do not meet the expectation, the user can retrive the saved intermediate results and perform further modification to the metric model. The verified model can then be applied to future similar projects.
acm southeast regional conference | 2000
Jonathan D. Fouss; Kai H. Chang
The definition of what groupware is can be a topic of great debate and is often very broad. This allows many types of software to earn the name groupware but makes it very difficult to compare applications and to do any kind of background research in the field. This paper attempts to bring some organization to the chaos and serve as introduction to groupware.
acm southeast regional conference | 2005
Jeong Yang; T. Dean Hendrix; Kai H. Chang
This paper presents the results of an empirical study carried out to investigate if the complexity profile graph (CPG) is related with one of the predicators of software complexity measures, such as response time and error rate, and thus could be used as a valid indicator of program comprehensibility. Lower and upper division computer science and software engineering students were asked to respond to questions regarding the execution of a source code module written in JAVA and understanding the fundamental goal of each part of the source code. The response time for each question and the correctness of each response were recorded. The statistical analysis of the experimental data reports that there is a positive linear correlation between the CPG and the time taken to respond correctly to each question. In addition, the paper discusses if there is any effect of the control structure diagram (CSD) on measuring the difficulty of program comprehensibility.
acm southeast regional conference | 1999
Chun-Yu Chen; Richard Chapman; Kai H. Chang
Regression testing is an important and costly activity in software maintenance. Not only the changed codes need to be tested, but it is also necessary to ensure that the changed code will not inadvertently a ect other functionalities. Because regression testing is expensive, automation has become an important issue. In this paper, we rst introduce a method for generating test scenarios, then we present a method for automatically generating usage-based regression test suites from an Object-Z formal speci cation. This method can be used in both integration and functional testing.
acm southeast regional conference | 1997
Kai H. Chang; Liam Murphy; Byong G. Lee; Timothy W. Dollar; Yifang Chang; Jonathan D. Fouss
For the past decade, numerous commercial and experimental systems have been designed for collaborative writing applications. From our experience working in the area of CSCW research and development, we have come to realize the possibility and the potential significance of CSCW systems in the engineering design process. The objective of the research is to introduce and develop a CSCW environment to support engineering design, specifically, in a distributed environment that combines computer aided design (CAD) and computer aided software engineering (CASE). In an effort to combine them into a seamless engineering package, we first target supporting engineering design (using CAD) with computer simulation (using CASE). This paper addresses issues involved in the design and development of the environment, including the graphical user interface, text and graphics editors, video and audio communication, multimedia database, document integration, and software simulation.
acm southeast regional conference | 1995
A. Wesley Wear; Yu Gong; Kai H. Chang
Traditional computer applications have been designed to be run by one user at a time who does some work in a single medium, such as ASCII text, and very little regard has been given to the fact that people often work together. With the recent development of computer networks and the widespread deployment of networked workstations, automating the group writing process for geographically distributed users has become feasible. In this paper, a software package which supports distributed, real-time, multimedia collaborative work, known as the Distributed Collaborative Writing Aid (DCWA), is described. The DCWA has five major parts but the emphasis of this paper is on the DCWAs distributed database, which is crucial to making the entire system work together both logically and consistently. The database is a specialized, distributed system based on the client-server model implemented in C++ running under 4.3BSD Unix and provides dynamic management capabilities unique for a multimedia collaborative working environment.
acm southeast regional conference | 2004
Yiming Ji; Kai H. Chang; Patrick O. Bobbie
Due to the increasing complexity of real world problems, it is costly and difficult to evaluate todays software-intensive systems. This paper explores component interaction pattern, and provides a set of methodologies for specifying, verifying, and validating (by simulation) system temporal behavior, with the aim of validating software design iteratively at design and development time. The methods combine such components and technologies as UML, XMI, database, model checking, and simulation.
acm southeast regional conference | 1997
Tim W. Dollar; Liam Murphy; Kai H. Chang; Byong G. Lee; Yifang Chang; Jonathan D. Fouss
Most traditional collaborative authoring tools offer some degree of document security and concurrency control. Document security is usually accomplished by allowing users to maintain private windows and to assign private ownership to files, perhaps through operating system mechanisms. Shared portions of a document can be manipulated through shared windows. Some form of concurrency control, such as a locking mechanism, is usually used to provide editors of a shared document with a consistent view of the document. However, few implementations focus on providing security for a document produced by a limited group of collaborators. Instead, these systems provide access to either the owner, or everyone. This paper discusses an organizational paradigm known the Collaboration. Collaborations facilitate group organization, document security, and concurrency control, not through major changes in the operating system or through the introduction of new network protocols, but instead, through features already present in the UNIX operating system and existing network protocols. This paper will also briefly discuss how the Distributed Collaborative Writing Aid (DCWA), a prototype groupware suite, uses the Collaboration paradigm to accomplish these goals.