Yu Chin Cheng

The distinctiveness of a curve in a parameterized neighborhood: extraction and applications

A new feature of curves pertaining to the acceptance/rejection decision in curve detection is proposed. The feature measures a curves distinctiveness in its neighborhood, which is modeled by a one-parameter family of curves. A computational framework based on the Hough transform for extracting the distinctiveness feature is elaborated and examples of feature extractors for the circle and the ellipse are given. It is shown that the proposed feature can be extracted efficiently and is effective in separating signals from false positives. Experimental results with circle and ellipse testing that strongly support the efficiency and effectiveness claims are obtained. The results further demonstrate that the proposed feature exhibits good noise resiliency

Exception handling refactorings: Directed by goals and driven by bug fixing

Exception handling design can improve robustness, which is an important quality attribute of software. However, exception handling design remains one of the less understood and considered parts in software development. In addition, like most software design problems, even if developers are requested to design with exception handling beforehand, it is very difficult to get the right design at the first shot. Therefore, improving exception handling design after software is constructed is necessary. This paper applies refactoring to incrementally improve exception handling design. We first establish four exception handling goals to stage the refactoring actions. Next, we introduce exception handling smells that hinder the achievement of the goals and propose exception handling refactorings to eliminate the smells. We suggest exception handling refactoring is best driven by bug fixing because it provides measurable quality improvement results that explicitly reveal the benefits of refactoring. We conduct a case study with the proposed refactorings on a real world banking application and provide a cost-effectiveness analysis. The result shows that our approach can effectively improve exception handling design, enhance software robustness, and save maintenance cost. Our approach simplifies the process of applying big exception handling refactoring by dividing the process into clearly defined intermediate milestones that are easily exercised and verified. The approach can be applied in general software development and in legacy system maintenance.

Change the face of software engineering education: A field report from Taiwan

Context: In Taiwan, the supply of software engineers provided by universities has suffered from both a quantity problem and a quality problem. An effort to change the software engineering education is in need. Objective: The Software Engineering Consortium (SEC) of Taiwan sets its objective to increase the number of college graduates that are better prepared for filling software development and maintenance jobs. Method: Four dysfunctions: avoidance of process, inattention to modeling, lack of awareness to software quality, and chasm between application domains and software engineering, of the current situation are identified. The effort to correct the dysfunctions involves design of a module-oriented software engineering curriculum, and organization of people, resource, and activities. Results: In the academic years from 2003 to 2008, both the number of software engineering courses offered and the enrollment size increased significantly by a space of some 250 courses and 5000 enrollments, respectively. Conclusion: The SEC effort to establishing software engineering modules has been received with enthusiasm by faculty members and students of the participating institutes. Inspired by the important foundational work such as SWEBOK and SE2004, we believe that the adopted strategy of identifying dysfunctions and then designing remedies to address these dysfunctions contributed significantly to the success of the SEC effort.

Polling an image for circles by random lines

A new random sampling strategy, designed for retrieving subsets consisting of two edge pixels from an input image, is proposed as the sampling process for RANSAC circle detection using coaxal transform. The proposed strategy is shown to have the following advantages over the conventional random sampling strategy. First, a poll size can be planned in a principled manner. Second, once a poll size is set, the probability that a circle is missed by the sampling process is kept relatively constant regardless of noise. Third, the actual number of subsets taken is automatically adjusted for different image complexities. Experimental results in agreement with the claimed advantages are presented.

Pseudo software: A mediating instrument for modeling software requirements

A new synthesis of software requirements models called pseudo software is proposed with the aim to cut requirements-related errors. Pseudo software achieves this aim by serving as a mediating instrument to empower stakeholders to participate in requirements elicitation and validation through model construction and manipulation, and to provide guidance to the development team to correctly interpret the requirements in the downstream development activities. Pseudo software obtains its traits as a mediating instrument through the choice of requirements information bits and the use of multimodal representations with tool support to integrate the requirements. Using historical data of fifty projects in the enterprise computing domain, pseudo software is shown to effectively cut the requirements-related errors committed by both the customer and the development team.

The Probabilistic Hough Transform with Localized Search Guided by Evidence Clusters

Two enhancements to the probabilistic Hough transform have been proposed, including the use of a new distinctiveness measure for hypothesis testing and a localized parameter vector search guided by evidence clusters. Preliminary experimental results show that large computational saving is achieved by employing the probabilistic Hough transform with distinctiveness measure when compared with the standard Hough transform. Furthermore, the use of evidence cluster can lead to further saving in computation time, especially when the image contains a large number of image points

Test-duo: a framework for generating and executing automated acceptance tests from use cases

The Test-Duo framework for generating and executing acceptance tests from use cases is presented. In Test-Duo, annotations are added to use cases to explicate system behaviors. ROBOT framework compatible test cases are then generated and applied to test the system under a search regime. Tool support is discussed.

Performance Improvement for an SOA Based Enterprise Application Using Service Redirection

A method using message redirection mechanism to reduce communication overhead of an SOA (service oriented application) based enterprise application is proposed. First, an SOA enterprise application design is modeled as a service invocation graph. Secondly, a refactoring method based on message redirection mechanism is introduced. It transfers the service invocation graph into a simplified service redirection graph. The SOA enterprise application designs behind the two graphs perform exactly the same functionality. However, the refactored new design can reduce the communication overhead up to 50% when comparing with the original one.

Contract Specification in Java: Classification, Characterization, and a New Marker Method

Design by Contract (DBC), originated in the Eiffel programming language, is generally accepted as a practical method for building reliable software. Currently, however, few languages have built-in support for it. In recent years, several methods have been proposed to support DBC in Java. We compare eleven DBC tools for Java by analyzing their impact on the developers programming activities, which are characterized by seven quality attributes identified in this paper. It is shown that each of the existing tools fails to achieve some of the quality attributes. This motivates us to develop ezContract, an open source DBC tool for Java that achieves all of the seven quality attributes. ezContract achieves streamlined integration with the working environment. Notably, standard Java language is used and advanced IDE features that work for standard Java programs can also work for the contract-enabled programs. Such features include incremental compilation, automatic refactoring, and code assist.

ezContract: Using Marker Library and Bytecode Instrumentation to Support Design by Contract in Java

Several approaches have been proposed to support Design by Contract in Java. In this paper, through the use of markers which are predefined dummy methods and attributes, a new approach to annotate contracts is presented. The annotated programs can be directly compiled by standard Java compilers. A bytecode instrumentor is developed to manipulate the bytecode to inject contract evaluation instructions and make the contracts executable at runtime. The marker approach avoids two primary problems found in the existing practices: source compatibility that depends on language extension and symbolic barrier that leaves contracts and their targets unrelated. It also facilitates streamlined integration with IDEs and improves readability as well as writability of the contract- annotated programs.