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.

Nonstationary feature extraction techniques for automatic classification of impact acoustic signals

Condition monitoring of wooden railway sleepers applications are generallycarried out by visual inspection and if necessary some impact acoustic examination iscarried out intuitively by skilled personnel. In this work, a pattern recognition solutionhas been proposed to automate the process for the achievement of robust results. Thestudy presents a comparison of several pattern recognition techniques together withvarious nonstationary feature extraction techniques for classification of impactacoustic emissions. Pattern classifiers such as multilayer perceptron, learning cectorquantization and gaussian mixture models, are combined with nonstationary featureextraction techniques such as Short Time Fourier Transform, Continuous WaveletTransform, Discrete Wavelet Transform and Wigner-Ville Distribution. Due to thepresence of several different feature extraction and classification technqies, datafusion has been investigated. Data fusion in the current case has mainly beeninvestigated on two levels, feature level and classifier level respectively. Fusion at thefeature level demonstrated best results with an overall accuracy of 82% whencompared to the human operator.

Classification of Parkinson’s Disease using MultiPass Lvq,Logistic Model Tree,K-Star for Audio Data set : Classification of Parkinson Disease using Audio Dataset

Parkinson's disease (PD) is a degenerative illness whose cardinal symptoms include rigidity, tremor, and slowness of movement. In addition to its widely recognized effects PD can have a profound effect on speech and voice.The speech symptoms most commonly demonstrated by patients with PD are reduced vocal loudness, monopitch, disruptions of voice quality, and abnormally fast rate of speech. This cluster of speech symptoms is often termed Hypokinetic Dysarthria.The disease can be difficult to diagnose accurately, especially in its early stages, due to this reason, automatic techniques based on Artificial Intelligence should increase the diagnosing accuracy and to help the doctors make better decisions. The aim of the thesis work is to predict the PD based on the audio files collected from various patients.Audio files are preprocessed in order to attain the features.The preprocessed data contains 23 attributes and 195 instances. On an average there are six voice recordings per person, By using data compression technique such as Discrete Cosine Transform (DCT) number of instances can be minimized, after data compression, attribute selection is done using several WEKA build in methods such as ChiSquared, GainRatio, Infogain after identifying the important attributes, we evaluate attributes one by one by using stepwise regression.Based on the selected attributes we process in WEKA by using cost sensitive classifier with various algorithms like MultiPass LVQ, Logistic Model Tree(LMT), K-Star.The classified results shows on an average 80%.By using this features 95% approximate classification of PD is acheived.This shows that using the audio dataset, PD could be predicted with a higher level of accuracy.

Speech Assessment for the Classification of Hypokinetic Dysthria in Parkinson Disease

The aim of this thesis is to investigate computerized voice assessment methods to classify between the normal and Dysarthric speech signals. In this proposed system, computerized assessment methods equipped with signal processing and artificial intelligence techniques have been introduced. The sentences used for the measurement of inter-stress intervals (ISI) were read by each subject. These sentences were computed for comparisons between normal and impaired voice. Band pass filter has been used for the preprocessing of speech samples. Speech segmentation is performed using signal energy and spectral centroid to separate voiced and unvoiced areas in speech signal. Acoustic features are extracted from the LPC model and speech segments from each audio signal to find the anomalies. The speech features which have been assessed for classification are Energy Entropy, Zero crossing rate (ZCR), Spectral-Centroid, Mean Fundamental-Frequency (Meanf0), Jitter (RAP), Jitter (PPQ), and Shimmer (APQ). Naïve Bayes (NB) has been used for speech classification. For speech test-1 and test-2, 72% and 80% accuracies of classification between healthy and impaired speech samples have been achieved respectively using the NB. For speech test-3, 64% correct classification is achieved using the NB. The results direct the possibility of speech impairment classification in PD patients based on the clinical rating scale.

Benchmark Evaluation of HOG Descriptors as Features for Classification of Traffic Signs

The purpose of this paper is to analyze the performance of the Histograms of Oriented Gradients (HOG) as descriptors for traffic signs recognition. The test dataset consists of speed limit traffic signs because of their high inter-class similarities.   HOG features of speed limit signs, which were extracted from different traffic scenes, were computed and a Gentle AdaBoost classifier was invoked to evaluate the different features. The performance of HOG was tested with a dataset consisting of 1727 Swedish speed signs images. Different numbers of HOG features per descriptor, ranging from 36 features up 396 features, were computed for each traffic sign in the benchmark testing. The results show that HOG features perform high classification rate as the Gentle AdaBoost classification rate was 99.42%, and they are suitable to real time traffic sign recognition. However, it is found that changing the number of orientation bins has insignificant effect on the classification rate. In addition to this, HOG descriptors are not robust with respect to sign orientation.

Corporate social responsibility and environmental sustainability: classification of the initiatives of the FMCG companies operating in Brazil

Este trabalho tem como objetivo fornecer uma análise detalhada do cenário da sustentabilidade ambiental e iniciativas de responsabilidade social corporativa nas empresas que operam no mercado de bens de consumo brasileiro. Para alcançar este objetivo as dez maiores empresas do mercado-alvo presente no Brasil foram analisadas por meio da classificação das suas iniciativas em três perspectivas amplas. Com esta classificação o cenário do mercado pode ser visto. As perspectivas utilizadas para a elaboração do trabalho são: (1) iniciativa ambiental ou social; (2) o foco interno ou externo e (3) a marca ou o custo como motivador. Depois de classificar todas as iniciativas, foi possível ver que as empresas similares, que operam em mercados semelhantes, têm estratégias que são muito parecidos entre si. Além disso, ficou claro que a estratégia de negócios da empresa influencia as suas políticas ambientais e sociais, em particular os objetivos que estas políticas procuram obter.Embora este trabalho apresente um panorama abrangente do setor de bens de consumo em relação a políticas de comportamento responsável das empresas, ele tem algumas limitações. A limitação mais significativa diz respeito a metodologia. As iniciativas foram avaliadas pela quantidade e a abrangência dos benefícios do impacto positivo não foram avaliados, impossibilitando assim a comparação do tamanho do impacto de cada empresa. Uma vez que pode haver um projeto de uma empresa que tem maior impacto do que vários outros feitos por alguma outra empresa. A metodologia foi baseada em clusters de categorias, no entanto, as iniciativas não são completamente uma coisa ou outra, ou seja, uma iniciativa pode ter diferentes impactos, drivers ou foco, nesses casos, os aspectos mais relevantes foram a escolhidos para classificá-los.

Growing research where economies grow: essays on strategy and international expansion of companies from emerging countries

In recent years, emerging countries have assumed an increasingly prominent position in the world economy, as growth has picked up in these countries and slowed in developed economies. Two related phenomena, among others, can be associated with this growth: emerging countries were less affected by the 2008-2009 global economic recession; and they increased their participation in foreign direct investment, both inflows and outflows. This doctoral dissertation contributes to research on firms from emerging countries through four independent papers. The first group of two papers examines firm strategy in recessionary moments and uses Brazil, one of the largest emerging countries, as setting for the investigation. Data were collected through a survey on Brazilian firms referring to the 2008-2009 global recession, and 17 hypotheses were tested using structural equation modeling based on partial least squares. Paper 1 offered an integrative model linking RBV to literatures on entrepreneurship, improvisation, and flexibility to indicate the characteristics and capabilities that allow a firm to have superior performance in recessions. We found that firms that pre-recession have a propensity to recognize opportunities and improvisation capabilities for fast and creative actions have superior performance in recessions. We also found that entrepreneurial orientation and flexibility have indirect effects. Paper 2 built on business cycle literature to study which strategies - pro-cyclical or counter-cyclical – enable superior performance in recessions. We found that while most firms pro-cyclically reduce costs and investments during recessions, a counter-cyclical strategy of investing in opportunities created by changes in the environment enables superior performance. Most successful are firms with a propensity to recognize opportunities, entrepreneurial orientation to invest, and flexibility to efficiently implement these investments. The second group of two papers investigated international expansion of multinational enterprises, particularly the use of distance for their location decisions. Paper 3 proposed a conceptual framework to examine circumstances under which distance is less important for international location decisions, taking the new perspective of economic institutional distance as theoretical foundation. The framework indicated that the general preference for low-distance countries is lower: (1) when the company is state owned, rather than private owned; (2) when its internationalization motives are asset, resource, or efficiency seeking, as opposed to market seeking; and (3) when internationalization occurred after globalization and the advent of new technologies. Paper 4 compared five concurrent perspectives of distance and indicated their suitability to the study of various issues based on industry, ownership, and type, motive, and timing of internationalization. The paper also proposed that distance represents the disadvantages of host countries for international location decisions; as such, it should be used in conjunction with factors that represent host country attractiveness, or advantages as international locations. In conjunction, papers 3 and 4 provided additional, alternative explanations for the mixed empirical results of current research on distance. Moreover, the studies shed light into the discussion of differences between multinational enterprises from emerging countries versus those from advanced countries.

Beyond ‘best practices': the international regulation of capital markets

It is widely acknowledged that there is considerable international pressure for international ‘best practices’ to be adopted via national legislation. This would occur either by means of model laws or through the passing of country specific legislation that closely replicates foreign legal formats, administrative rules, and or regulation. These attempts to spread the implementation of ‘best practices’ have gained importance in the international debate due to the liberalization of international capital flows. The oversight, country reports, and technical assistance carried out by international organizations along with the growing internationalization of investors have also contributed to this growing pressure. In this respect, due to the constant evolution of transactions and the end objective of making sure that capital markets are developed with just rules, structures, and methods, this article looks to analyze the adoption of standardized models of capital market regulation. Furthermore it looks to examine the motivation and interest of states and other ‘stakeholders’ at the international level.

Classification of Boal, Malvazia, Sercial and Verdelho wines based on terpenoid patterns

Thirty-six Madeira wine samples from Boal, Malvazia, Sercial and Verdelho white grape varieties were analyzed in order to estimate the free fraction of monoterpenols and C13 norisoprenoids (terpenoid compounds) using dynamic headspace solid phase micro-extraction (HS-SPME) technique coupled with gas chromatography–mass spectrometry (GC–MS). The average values from three vintages (1998–2000) show that these wines have characteristic profiles of terpenoid compounds. Malvazia wines exhibits the highest values of total free monoterpenols, contrary to Verdelho wines which had the lowest levels of terpenoids but produced the highest concentration of farnesol. The use of multivariate analysis techniques allows establishing relations between the compounds and the varieties under investigation. Principal component analysis (PCA) and linear discriminant analysis (LDA) were applied to the obtained matrix data. A good separation and classification power between the four groups as a function of their varietal origin was observed.

Environmental Fate of Neonicotinoids and Classification of Their Potential Risks to Hypogean, Epygean, and Surface Water Ecosystems in Brazil

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

A classification of Pepper yellow mosaic virus isolates into pathotypes

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

Self-Organizing Maps as a New Tool for Classification of Plants at Lower Hierarchical Levels

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