Abstract
This paper addresses the task of finding outliers within each class in the context of supervised classification problems. Class-based outliers are cases that deviate too much with respect to the cases of the same class. We introduce a novel method for outlier detection in labelled data based on Random Forests and compare it with existing methods both on artificial and real-world data. We show that it is competitive with the existing methods and sometimes gives more intuitive results. We also provide an overview for outlier detection in labelled data. The main contribution are two methods for class-based outlier description and interpretation.

Abstract
In many engineering and science areas, models are developed and validated using high-level programing languages and environments as is the case with MATLAB. In order to target the multicore heterogeneous architectures being used on embedded systems to provide high performance computing with acceptable energy/power envelops, developers manually migrate critical code sections to lower-level languages such as C and OpenCL, a time consuming and error prone process. Thus, automatic source-to-source approaches are highly desirable. We present an approach to compile MATLAB and output equivalent C/OpenCL code to target architectures, such as GPU based hardware accelerators. We evaluate our approach on an existing MATLAB compiler framework named MATISSE. The OpenCL generation relies on the manual insertion of directives to guide the compilation and is also capable of generating C wrapper code to interface and synchronize with the OpenCL code. We evaluated the compiler with a number of benchmarks from different domains and the results are very encouraging.

Abstract
Tire inspection is presently done by workers who have as their main problems, besides identifying the defects, the time available for defect identification and the inherent costs. Companies can become more sustainable by adopting automated methods to perform such type of processes, such as artificial vision, with advantages both in the processing time and in the incurred costs. This paper addresses the development of an artificial vision system that aims to be an asset in the field of tyre inspection, having as main characteristics its execution speed and its reliability. The conjugation of these criteria is a prerequisite for this system to be able to be integrated in inspection machines. The paper focusses on the study of three image processing methods to be used in the identification of marks (red dots) on tires. In this work was used the free Open Computer Vision artificial vision library to process the images acquired by a Basler matrix camera. Two different techniques, namely Background Subtraction and Hough Transform, were tested to implement the solution. After developing the artificial vision inspection application, tests were made to measure the performance of both methods and the results were promising: processing time was low and, simultaneous, the achieved accuracy is high.

Abstract
In this paper, we proffer an explicit representation of solutions for a specific class of linear repetitive processes with smoothing. This representation is used to obtain direct controllability and observability criteria of this same class of discrete time 2-D systems. Not only classical controllability properties are considered, where control of the system is obtained by choosing its inhomogeneity appropriately, but also controllability of the system by steering it through boundary data control. From the point of view of technical applications, for instance in high pressure gas network modelling (see Azevedo-PerdicoA(0)lis and Jank in Proceedings of n-DS, international workshop on multidimensional systems, Thessaloniki. 2009), it seems to be more reliable to consider boundary data controls. Therefore, in this paper we emphasise boundary data control properties of the system. A disturbed optimal boundary control problem with a quadratic criterion is also solved.

Abstract

Abstract
Cross-bite, as a malocclusion effect, is defined as a transversal changing of the upper dental arch, in relation to the lower arch, and may be classified as skeletal, dental or functional. As a consequence, the expansion of maxilla is an effective clinical treatment used to correct transversal maxillary discrepancy. The maxillary expansion is an ancient method used in orthodontics, for the correction of the maxillary athresia with posterior crossbite, through the opening of the midpalatal suture (disjunction), using orthodontic-orthopaedic devices. Same controversial discussion arises among the clinicians, about the effects of each orthodontic devices as also about the technique to be employed. The objective of this study was to compare the strain field induced by two different orthodontic devices, named disjunctor with and without a connecting bar, in an acrylic model jaw, using fiber Bragg grating sensors to measure the strain patterns. The orthodontic device disjunctor with the bar, in general, transmits higher forces and strain to teeth and maxillae, than with the disjunctor without bar. It was verified that the strain patterns were not symmetric between the left and the right sides as also between the posterior and anterior regions of the maxillae. For the two devices is also found that in addition a displacement in the horizontal plane, particularly in posterior teeth, also occurs a rotation corresponding to a vestibularization of the posterior teeth and their alveolar processes.