VPR-Bench: An Open-Source Visual Place Recognition Evaluation Framework with Quantifiable Viewpoint and Appearance Change

International Journal of Computer Vision - Visual place recognition (VPR) is the process of recognising a previously visited place using visual information, often under varying appearance...     

Trajectory equilibrium state detection and avoidance algorithm for multi-autonomous potential field mobile robots

An exceptionally efficient algorithm for local multi-mobile robot autonomous navigation is presented. The algorithm guides each robot independently although the algorithmic principle is identical for each robot. This approach is a combination of a novel rule-based mathematical algorithm and the virtual force field (VFF) navigational method. The need for the combinational algorithm is due to VFF's inability to guide efficiently multi-autonomous robots in the same environment owing to a trajectory equilibrium state (TES). TES is a new state that dramatically degrades performance of the VFF and is firstly identified and defined within this study. Test results of the novel algorithm yield near-optimum trajectories for the mobile robots.     

Implementing associations: UML 2.0 to Java 5

A significant current software engineering problem is the conceptual mismatch between the abstract concept of an association as found in modelling languages such as UML and the lower level expressive facilities available in object-oriented languages such as Java. This paper introduces some code generation patterns that aid the production of Java based implementations from UML models. The work is motivated by a project to construct model driven development tools in support of the construction of embedded systems. This involves the specification and implementation of a number of meta-models (or models of languages). Many current UML oriented tools provide code generation facilities, in particular the generation of object-oriented code from class diagrams. However, many of the more complex aspects of class diagrams, such as qualified associations are not supported. In addition, several concepts introduced in UML version 2.0 are also not supported.The aim of the work presented in this paper is to develop a number of code generation patterns that allow us to support the automatic generation of Java code from UML class diagrams that support these new and complex association concepts. These patterns significantly improve the code generation abilities of UML tools, providing a useful automation facility that bridges the gap between the concept of an association and lower level object-oriented programming languages.     

A Novel Weightless Artificial Neural Based Multi-Classifier for Complex Classifications

Artificial neural systems in general and weightless systems in particular, have traditionally struggled in performance terms when confronted with problem domains such as possessing a large number of independent pattern classes and pattern classes with non-standard distributions. A multi-classifier is proposed which explores problem domains with a large number of independent pattern classes typically found in forensic and security databases. Specifically, the multi-classifier system is demonstrated on the exemplar of fingerprint identification problem typical to forensic, biometric, and security. Furthermore, the multi-classifier is able to provide a reasonable solution to benchmark problems from medicinal and physical (science) fields, which are determining the health, state of thyroid glands and determining whether or not there is a structure in the ionosphere, respectively.