Identifying the most relevant tablet regions in the image detection of counterfeit medicines.

This paper proposes a novel image-based approach to detect counterfeit medicines and identify the most relevant regions of the tablet in the task of classification. Images of medicine tablets undergo an initial pre-processing step which (i) removes the background to find the region of interest, (ii) clusters individual pixels into super-pixels, and (iii) extracts features containing color and texture information. The classification relying on Support Vector Machine (SVM) defines the class the respective image will be inserted into. The task of identifying the relevant regions of the tablets for counterfeiting detection is performed using the concept of support vectors, generating a heat map that indicates the regions that contribute the most to the classification purpose. Two datasets containing images of authentic and counterfeit tablets of Cialis and Viagra were used to validate our propositions, achieving correct classification rates of 100% on both datasets. Regarding the task of identifying the most relevant regions, our proposition outperformed the traditional LIME (Local Interpretable Model-agnostic Explanations) method by yielding more robust explanations.

3D scanner and cad software application in animal osteometry / Aplicação do scanner 3D e software cad em osteometria animal

The research of the structure of wild animal anatomy is important and can be used to help with species conservation. The objective of this study was to describe the methodology used to digitally scan bone structure of these animals. Thus, all the steps used to perform the scapula scan of an adult specimen of Cerdocyon thous was thoroughly described as a way to illustrate the process that is being developed for the study of skeletons. A scapula was chosen to be digitalized because it is a plane bone of the forelimb and is fundamental to unite the limb to the animal's trunk. Advanced morphometric analysis of osteometry was depicted by creating three-dimensional physical models of bone structures utilizing the 3D Model Uniscan, a computational engineering tool that enables accurate results of the bone structure in the model. The equipment works by reading the points of the piece through the reflection of the laser in the lenses and sensors. These points are computed and positioned by VXelements software which is responsible for the capture of the targets, which in turn, generates a triangular surface mesh in real time which reproduces the 3D surface of the bone instantaneously. Finally, these sets of acquired data can be converted into various file formats compatible with CAD software, such as Solidworks which allows the dimensions to be extracted at different points of the object. This way of characterization and modeling allows the analysis of the structures' dimensions to be more informative and real. The obtained results of such analyzes have application in the studies of ecomorphology and evolution.

O detalhamento da anatomia de animais silvestres é importante, uma vez que pode ser usado a favor da conservação. Objetivou-se, com este trabalho, descrever a metodologia para realizar a digitalização da estrutura óssea desses animais. Assim, foram descritas todas as etapas utilizadas para efetuar a digitalização da escápula de um espécime adulto de Cerdocyon thous como uma forma de exemplificar o processo que está sendo desenvolvido para o estudo de esqueletos. Além disso, escolheu-se digitalizar a escápula, pois se trata de um osso plano do membro torácico, fundamental para unir o membro ao tronco do animal. As análises morfométricas avançadas de osteometria foram realizadas por meio da criação de modelos tridimensionais de estruturas ósseas, através de modelos físicos reais, com o uso do equipamento SCANNER 3D Modelo Uniscan, uma ferramenta de engenharia computacional que possibilita obter resultados precisos da estrutura óssea do modelo. O equipamento funciona por meio da leitura dos pontos da peça pela reflexão do laser nas lentes e sensores. Estes pontos são computados e posicionados pelo software VXelements que é responsável pela captação de alvos que, por sua vez, geram uma malha de superfície triangular em tempo real, o qual reproduz a superfície 3D instantânea do osso. Ao final, estes conjuntos de dados adquiridos podem ser convertidos em diversos formatos de arquivo compatíveis com softwares CAD como, por exemplo, o Solidworks que permite que as dimensões sejam extraídas em diferentes pontos do objeto. Sendo assim, esta forma de caracterização e modelagem permite tornar a análise da dimensão mais elucidativa e real. Os resultados obtidos nessas análises têm aplicação em estudos de ecomorfologia e evolução.