Machine vision for the automatic classification of images acquired from Non-destructive tests

This project is based on Artificial Intelligence (A.I) and Digital Image processing (I.P) for automatic condition monitoring of sleepers in the railway track. Rail inspection is a very important task in railway maintenance for traffic safety issues and in preventing dangerous situations. Monitoring railway track infrastructure is an important aspect in which the periodical inspection of rail rolling plane is required.Up to the present days the inspection of the railroad is operated manually by trained personnel. A human operator walks along the railway track searching for sleeper anomalies. This monitoring way is not more acceptable for its slowness and subjectivity. Hence, it is desired to automate such intuitive human skills for the development of more robust and reliable testing methods. Images of wooden sleepers have been used as data for my project. The aim of this project is to present a vision based technique for inspecting railway sleepers (wooden planks under the railway track) by automatic interpretation of Non Destructive Test (NDT) data using A.I. techniques in determining the results of inspection.

Non-destructive determination of photosynthetic pigments in the leaves of castor oil plants

Objetivou-se com o presente trabalho, estabelecer a relação entre os pigmentos fotossintéticos extraídos em DMSO e as leituras obtidas no clorofilômetro portátil ClorofiLOG® 1030, gerando modelos matemáticos capazes de predizer os teores de clorofila e de carotenóides em folhas de mamoneira. O trabalho foi conduzido na Empresa Brasileira de Pesquisa Agropecuária (EMBRAPA) Algodão, situada em Campina Grande, Estado da Paraíba, em outubro de 2010. Para a análise indireta, foi utilizado um equipamento portátil, sendo realizada a leitura em discos foliares com diferentes tonalidades de verde, sendo feita, nesses mesmos discos, a determinação da clorofila pelo método clássico. Para a extração da clorofila, utilizaram-se 5 mL de dimetilsulfóxido (DMSO), a qual foi mantida em banho-maria a 70ºC, por 30 minutos, e retirou-se 3 mL da alíquota para leitura em espectrofotômetro nos comprimentos de onda de 470, 646 e 663 nm. Os dados foram submetidos à análise da variância e regressão polinomial. A leitura obtida no clorofilômetro portátil foi a variável dependente, e os pigmentos fotossintéticos determinados pelo método clássico foi a variável independente. Os resultados indicaram que o clorofilômetro portátil ClorofiLOG® 1030, associado a modelos matemáticos, permitiu estimar a concentração dos pigmentos fotossintéticos, exceto a clorofila b, com alta precisão, com economia de tempo e com reagentes normalmente utilizados nos procedimentos convencionais.

Composite materials obtained from textile fiber residue

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Non-destructive genetic sampling in fish. An improved method for DNA extraction from fish fins and scales

DNA-based studies have been one of the major interests in conservation biology of endangered species and in population genetics. As species and population genetic assessment requires a source of biological material, the sampling strategy can be overcome by non-destructive procedures for DNA isolation. An improved method for obtaining DNA from fish fins and scales with the use of an extraction buffer containing urea and further DNA purification with phenol-chloroform is described. The methodology combines the benefits of a non-destructive DNA sampling and its high efficiency. In addition, comparisons with other methodologies for isolating DNA from fish demonstrated that the present procedure also becomes a very attractive alternative to obtain large amounts of high-quality DNA for use in different molecular analyses. The DNA samples, isolated from different fish species, have been successfully used on random amplified polymorphic DNA (RAPD) experiments, as well as on amplification of specific ribosomal and mitochondrial DNA sequences. The present DNA extraction procedure represents an alternative for population approaches and genetic studies on rare or endangered taxa.

Characterization of mineral contents on androgen deficient and chronic ethanol consuming rats by biochemistry and non-destructive techniques

Ethanol can compromise the body mineral composition and affect bone, and when associated to hypogonadism is considered an important risk factor for osteoporosis in man. The aim of this study was to investigate the effect of androgen deficient and chronic ethanol consuming on mineral contents by biochemistry and non-destructive techniques. Wistar rat (n=54) were divided in orchiectomy (ORQ) or SHAM-operated and subdivided by diet. They were daily fed with a Lieber DeCarli diet model for 8 weeks long. The controls groups were free-diet and pair-fed. Ca and P were analyzed by biochemistry test in the blood and by nX-ray fluorescence and FT-Raman on the femur area. Serum analysis revealed hypocalcaemia and hypeiphosphataemia in ethanol groups more than pair-fed and free-diet. In similarity, spectroscopy indicated a decrease in bone Ca content in ORQ groups, mainly for ethanol groups. Phosphorus content and Ca/P molar ratio, otherwise, doesn't diverge in all 6 groups. Ethanol consumption impaired Ca and P homeostasis in ORQ rat more than SHAM. The relationships among ethanol consume and androgen deficit support the hypothesis that ethanol affects the mineral-regulating hormones and may mediate some effects on bone. These findings demonstrate that ethanol seemed to interfere with the normal compensatory response to these Ca and P levels and is more significant M androgen deficiency rats.

Electric and magnetic properties of polymer electrolyte/carbon black composites

Measurements of 1H Nuclear Magnetic Resonance (NMR) relaxation times, Electron Paramagnetic Resonance (EPR) and AC Impedance Spectroscopy (IS) are reported for composites based on PEO8:LiClO4 and carbon black (CB), prepared by two methods: solvent and fusion processing. Three nuclear relaxation processes were identified for 1H nuclei: (i) belonging to the polymer chains in the amorphous phase, loosely bound to the CB particles, whose dynamics is almost the same as for unfilled polymer, (ii) belonging to the polymer chains which are tightly attached to the CB particles, and (iii) belonging to the crystalline phase in the loose polymer chain. The paramagnetic electronic susceptibility of the composite samples, measured by EPR, was interpreted by assuming a contribution of localized spin states that follow a Curie law, and a Pauli-like contribution of delocalized spins. A significant change of the EPR linewidth was observed at 40 K, which is the temperature where the Curie and Pauli susceptibilities equally contribute to the paramagnetic electronic susceptibility. The electrical properties are very sensitive to the preparation methods of the composites, which conditions the interaction between carbon particle-carbon particle and carbon particle-polymer chain. Classical statistic models to describe the conductivity in these media were not satisfactory. © 1998 Published by Elsevier Science B.V. All rights reserved.

Investigations into the use of the finite element method and artificial neural networks in the non-destructive analysis of metallic tubes

This work presents an investigation into the use of the finite element method and artificial neural networks in the identification of defects in industrial plants metallic tubes, due to the aggressive actions of the fluids contained by them, and/or atmospheric agents. The methodology used in this study consists of simulating a very large number of defects in a metallic tube, using the finite element method. Both variations in width and height of the defects are considered. Then, the obtained results are used to generate a set of vectors for the training of a perceptron multilayer artificial neural network. Finally, the obtained neural network is used to classify a group of new defects, simulated by the finite element method, but that do not belong to the original dataset. The reached results demonstrate the efficiency of the proposed approach, and encourage future works on this subject.

Background interference on the color of dental composite materials with different thickness by digital contrast

In this work the color dependence of resin composites with the background color was evaluated. The objective was to measure since what thickness the color of the sample stops being influenced by the color of the background over which the resin is placed and the methodology used in experiment was based in analyzing the contrast of digital images of the sample over a black background. The results shown that since 0.8 mm the images contrast becomes almost constant; it prove that since this thickness the color of resin composite depends on the optical resin properties only. The experiment was repeated under three conditions of luminosity to evaluate the influence of it on the image contrast and the results obtained were identical.

Sparse arrays and image compounding techniques for non-destructive testing using guided acoustic waves

Piezoelectric array transducers applications are becoming usual in the ultrasonic non-destructive testing area. However, the number of elements can increase the system complexity, due to the necessity of multichannel circuitry and to the large amount of data to be processed. Synthetic aperture techniques, where one or few transmission and reception channels are necessary, and the data are post-processed, can be used to reduce the system complexity. Another possibility is to use sparse arrays instead of a full-populated array. In sparse arrays, there is a smaller number of elements and the interelement spacing is larger than half wavelength. In this work, results of ultrasonic inspection of an aluminum plate with artificial defects using guided acoustic waves and sparse arrays are presented. Synthetic aperture techniques are used to obtain a set of images that are then processed with an image compounding technique, which was previously evaluated only with full-populated arrays, in order to increase the resolution and contrast of the images. The results with sparse arrays are equivalent to the ones obtained with full-populated arrays in terms of resolution. Although there is an 8 dB contrast reduction when using sparse arrays, defect detection is preserved and there is the advantage of a reduction in the number of transducer elements and data volume. © 2013 Brazilian Society for Automatics - SBA.

Lamb mode diversity imaging for non-destructive testing of plate-like structures

Several Lamb wave modes can be coupled to a particular structure, depending on its geometry and transducer used to generate the guided waves. Each Lamb mode interacts in a particular form with different types of defects, like notches, delamination, surface defects, resulting in different information which can be used to improve damage detection and characterization. An image compounding technique that uses the information obtained from different propagation modes of Lamb waves for non-destructive testing of plate-like structures is proposed. A linear array consisting of 16 piezoelectric elements is attached to a 1 mm thickness aluminum plate, coupling the fundamental A0 and S0 modes at the frequencies of 100 kHz and 360 kHz, respectively. For each mode two images are obtained from amplitude and phase information: one image using the Total Focusing Method (TFM) and one phase image obtained from the Sign Coherence Factor (SCF). Each TFM image is multiplied by the SCF image of the respective mode to improve contrast and reduce side and grating lobes effects. The high dispersive characteristic of the A0 mode is compensated for adequate defect detection. The information in the SCF images is used to select one of the TFM mode images, at each pixel, to obtain the compounded image. As a result, dead zone is reduced, resolution and contrast are improved, enhancing damage detection when compared to the use of only one mode. © 2013 Elsevier Ltd.

Influence of filler charge on gloss of composite materials before and after in vitro toothbrushing

Objectives: This study evaluated the gloss behaviour of experimental resin composites loaded with different filler percentages, immediately after polishing and after toothbrushing simulation. Methods: Sixteen disc-shaped specimens were fabricated for each different-charged composite (40%, 50%, 60%, 70% and 75%) and polished with SiC abrasive papers. Gloss measurements were made prior to simulated toothbrushing. The specimens were subjected to the simulation for 5, 15, 30 and 60 min using an electrical toothbrush with a standardized pressure while being immersed in a toothpaste/artificial saliva slurry. Results: Baseline composite gloss values ranged from 69.7 (40%) to 81.3 (75%) GU (gloss units) and from 18.1 (40%) to 32.3 (75%) GU after 1 h of brushing. Highest gloss values were obtained by 75%-charged resin, while the lowest values were obtained by the 40%-charged one. Conclusions: All tested materials showed a gloss decrease. However, the higher filler load a composite resin has, the higher gloss it can achieve. Clinical significance: Gloss of resin composite materials is an important factor in determining aesthetic success of anterior restorations, and this property may vary according to the filler charge of the restorative material. Higher filler load of a composite resin results in higher gloss values. © 2013 Elsevier Ltd. All rights reserved.

Non-destructive thermal wave method applied to study thermal properties of fast setting time endodontic cement

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Non-destructive equations to estimate the leaf area of Styrax pohlii and Styrax ferrugineus

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Tomografia por biosusceptometria de corrente alternada

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)