Volume 19 (2016): Issue 1 (May 2016)

This paper presents the software development meta-process (SD-MP) as a proposal to set up software projects. Within this proposal we offer conceptual elements that help solve the war of methodologies and processes in favor of an integrating viewpoint, where the main flaws associated with conventional and agile approaches are removed. Our newly developed software platform to support the meta-process is also presented together with three case studies involving projects currently in progress, where the framework proposed in SD-MP has been applied.

The approach called “Topological Functioning Model for Software Engineering” (TFM4SE) applies the Topological Functioning Model (TFM) for modelling the business system in the context of Model Driven Architecture. TFM is a mathematically formal computation independent model (CIM). TFM4SE is compared to an approach that uses BPMN as a CIM. The comparison focuses on CIM modelling and on transformation to UML Sequence diagram on the platform independent (PIM) level. The results show the advantages and drawbacks the formalism of TFM brings into the development.

This article presents an ETL (Extract, Transform, Load) prototype called Valery as alternative approach to migration process which includes a compiler for dynamic generation of SQL queries. Its main features involve: SQL dynamic generation, set of configuration commands and environment for file uploading. The tests use the Northwind academic database and an individual environment. The model implementation uses flat files and SQL as query language. Finally, there is an analysis of the results obtained.

Software plays a key role nowadays in all fields, from simple up to cutting-edge technologies and most of technology devices now work on software. Software development verification and validation have become very important to produce the high quality software according to business stakeholder requirements. Different software development methodologies have given a new dimension for software testing. In traditional waterfall software development software testing has approached the end point and begins with resource planning, a test plan is designed and test criteria are defined for acceptance testing. In this process most of test plan is well documented and it leads towards the time-consuming processes. For the modern software development methodology such as agile where long test processes and documentations are not followed strictly due to small iteration of software development and testing, lean canvas transformation models can be a solution. This paper provides a new dimension to find out the possibilities of adopting the lean transformation models and metrics in the software test plan to simplify the test process for further use of these test metrics on canvas.

System modelling using Unified Modelling Language (UML) is the task that should be solved for software development. The more complex software becomes the higher requirements are stated to demonstrate the system to be developed, especially in its dynamic aspect, which in UML is offered by a sequence diagram. To solve this task, the main attention is devoted to the graphical presentation of the system, where diagram layout plays the central role in information perception. The UML sequence diagram due to its specific structure is selected for a deeper analysis on the elements’ layout. The authors research represents the abilities of modern UML modelling tools to offer automatic layout of the UML sequence diagram and analyse them according to criteria required for the diagram perception.

Despite the fact that model composition paradigm becomes very important and most commonly used, the support for their cooperation has not reached its full strength, especially in managing composition conflicts, because it’s often divided between and confused with other model composition features. This makes handling and dealing with these conflicts a crucial activity in the composition process. Models need to be put under version control in order to manage the probable conflicts, facilitate collaboration and control change. Therefore, a solution for detecting and resolving conflicts is needed. In this paper, we present a composition conflict resolver presented in the form of a repository that helps manage composition conflicts and analyse the model and operations performed on it.

The paper focuses on the comparison of video steganography methods for the purpose of digital watermarking in the context of copyright protection. Four embedding methods that use Discrete Cosine and Discrete Wavelet Transforms have been researched and compared based on their embedding efficiency and fidelity. A video steganography program has been developed in the Java programming language with all of the researched methods implemented for experiments. The experiments used 3 video containers with different amounts of movement. The impact of the movement has been addressed in the paper as well as the ways of potential improvement of embedding efficiency using adaptive embedding based on the movement amount. Results of the research have been verified using a survey with 17 participants.

An optimization problem of classifying shifting-distorted objects is studied. The classifier is 2-layer perceptron, and the object model is monochrome 60 × 80 image. Based on the fact that previously the perceptron has successfully been attempted to classify shifted objects with a pixel-to-shift standard deviation ratio for training, the ratio is optimized. The optimization criterion is minimization of classification error percentage. A classifier trained under the found optimal ratio is optimized additionally. Then it effectively classifies shifting-distorted images, erring only in one case from eight takings at the maximal shift distortion. On average, classification error percentage appears less than 2.5 %. Thus, the optimized 2-layer perceptron outruns much slower neocognitron. And the found optimal ratio shall be nearly the same for other object classification problems, when the number of object features varies about 4800, and the number of classes is between two and three tens.