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.

Evaluation of a PVT Air Collector

Hybrid Photovoltaic Thermal (PVT) collectors are an emerging technology that combines PV and solar thermal systems in a single solar collector producing heat and electricity simultaneously. The focus of this thesis work is to evaluate the performance of unglazed open loop PVT air system integrated on a garage roof in Borlänge. As it is thought to have a significant potential for preheating ventilation of the building and improving the PV modules electrical efficiency. The performance evaluation is important to optimize the cooling strategy of the collector in order to enhance its electrical efficiency and maximize the production of thermal energy. The evaluation process involves monitoring the electrical and thermal energies for a certain period of time and investigating the cooling effect on the performance through controlling the air mass flow provided by a variable speed fan connected to the collector by an air distribution duct. The distribution duct transfers the heated outlet air from the collector to inside the building. The PVT air collector consists of 34 Solibro CIGS type PV modules (115 Wp for each module) which are roof integrated and have replaced the traditional roof material. The collector is oriented toward the south-west with a tilt of 29 ᵒ. The collector consists of 17 parallel air ducts formed between the PV modules and the insulated roof surface. Each air duct has a depth of 0.05 m, length of 2.38 m and width of 2.38 m. The air ducts are connected to each other through holes. The monitoring system is based on using T-type thermocouples to measure the relevant temperatures, air sensor to measure the air mass flow. These parameters are needed to calculate the thermal energy. The monitoring system contains also voltage dividers to measure the PV modules voltage and shunt resistance to measure the PV current, and AC energy meters which are needed to calculate the produced electrical energy. All signals recorded from the thermocouples, voltage dividers and shunt resistances are connected to data loggers. The strategy of cooling in this work was based on switching the fan on, only when the difference between the air duct temperature (under the middle of top of PV column) and the room temperature becomes higher than 5 °C. This strategy was effective in term of avoiding high electrical consumption by the fan, and it is recommended for further development. The temperature difference of 5 °C is the minimum value to compensate the heat losses in the collecting duct and distribution duct. The PVT air collector has an area of (Ac=32 m2), and air mass flow of 0.002 kg/s m2. The nominal output power of the collector is 4 kWppv (34 CIGS modules with 115 Wppvfor each module). The collector produces thermal output energy of 6.88 kWth/day (0.21 kWth/m2 day) and an electrical output energy of 13.46 kWhel/day (0.42 kWhel/m2 day) with cooling case. The PVT air collector has a daily thermal energy yield of 1.72 kWhth/kWppv, and a daily PV electrical energy yield of 3.36 kWhel /kWppv. The fan energy requirement in this case was 0.18 kWh/day which is very small compared to the electrical energy generated by the PV collector. The obtained thermal efficiency was 8 % which is small compared to the results reported in literature for PVT air collectors. The small thermal efficiency was due to small operating air mass flow. Therefore, the study suggests increasing the air mass flow by a factor of 25. The electrical efficiency was fluctuating around 14 %, which is higher than the theoretical efficiency of the PV modules, and this discrepancy was due to the poor method of recording the solar irradiance in the location. Due to shading effect, it was better to use more than one pyranometer.

Real time video segmentation for recognising paint marks on bad wooden railway sleepers

Wooden railway sleeper inspections in Sweden are currently performed manually by a human operator; such inspections are based on visual analysis. Machine vision based approach has been done to emulate the visual abilities of human operator to enable automation of the process. Through this process bad sleepers are identified, and a spot is marked on it with specific color (blue in the current case) on the rail so that the maintenance operators are able to identify the spot and replace the sleeper. The motive of this thesis is to help the operators to identify those sleepers which are marked by color (spots), using an “Intelligent Vehicle” which is capable of running on the track. Capturing video while running on the track and segmenting the object of interest (spot) through this vehicle; we can automate this work and minimize the human intuitions. The video acquisition process depends on camera position and source light to obtain fine brightness in acquisition, we have tested 4 different types of combinations (camera position and source light) here to record the video and test the validity of proposed method. A sequence of real time rail frames are extracted from these videos and further processing (depending upon the data acquisition process) is done to identify the spots. After identification of spot each frame is divided in to 9 regions to know the particular region where the spot lies to avoid overlapping with noise, and so on. The proposed method will generate the information regarding in which region the spot lies, based on nine regions in each frame. From the generated results we have made some classification regarding data collection techniques, efficiency, time and speed. In this report, extensive experiments using image sequences from particular camera are reported and the experiments were done using intelligent vehicle as well as test vehicle and the results shows that we have achieved 95% success in identifying the spots when we use video as it is, in other method were we can skip some frames in pre-processing to increase the speed of video but the segmentation results we reduced to 85% and the time was very less compared to previous one. This shows the validity of proposed method in identification of spots lying on wooden railway sleepers where we can compromise between time and efficiency to get the desired result.

Detecting Plants on Railway Embankments

This paper investigates problems concerning vegetation along railways and proposes automatic means of detecting ground vegetation. Digital images of railway embankments have been acquired and used for the purpose. The current work mainly proposes two algorithms to be able to achieve automation. Initially a vegetation detection algorithm has been investigated for the purpose of detecting vegetation. Further a rail detection algorithm that is capable of identifying the rails and eventually the valid sampling area has been investigated. Results achieved in the current work report satisfactory (qualitative) detection rates.

Implantação de tecnologia de monitoramento remoto para e-service: um estudo exploratório

The purpose of this study is to define measures to increase customer satisfaction and company competitiveness using a remote monitoring technology, in an exploratory study of Alpha Elevator Company (nick name chosen to the company by the actor of the dissertation). Regarding the competitive market, the service industry is striving to achieve productivity, following the example of the manufacturing industry. Nevertheless, these efforts are limited by the amount of hours worked per week, month or year, since the sector charges its services based on the hours spent working on the equipment of the client or based on the numbers of visits. This study is based in the overcoming of the traditional paradigm of selling number of hours by a system of selling results and performance. Employing a remote monitoring system, the elevators under the company service are monitored continually and defects are detected and transmitted to the customer care center, via phone line. The customers can access this data through the Internet and obtain information like availability rate of their elevators and call back response time rate, besides being able to buy products on the company¿s home page and to send feedback. The results were obtained by participating in conferences among experts of the company, in Japan and the United States. Through the analysis of the business environment and based on the bibliographic reference, a strategy was developed to implement e-service as a competitive differentiation.

Board of directors and top management team: a study on CEO relative power and financial return

Best corporate governance practices published in the primers of Brazilian Securities and Exchange Commission and the Brazilian Corporate Governance Institute promote board independence as much as possible, as a way to increase the effectiveness of governance mechanism (Sanzovo, 2010). Therefore, this paper aims at understanding if what the managerial literature portraits as being self-evident - stricter governance, better performance - can be observed in actual evidence. The question answered is: do companies with a stricter control and monitoring system perform better than others? The method applied in this paper consists on comparing 116 companies in respect to the their independence level between top management team and board directors– being that measured by four parameters, namely, the percentage of independent outsiders in the board, the separation of CEO and chairman, the adoption of contingent compensation and the percentage of institutional investors in the ownership structure – and their financial return measured in terms return on assets (ROA) from the latest Quarterly Earnings release of 2012. From the 534 companies listed in the Stock Exchange of Sao Paulo – Bovespa – 116 were selected due to their level of corporate governance. The title “Novo Mercado” refers to the superior level of governance level within companies listed in Bovespa, as they have to follow specific criteria to assure shareholders ´protection (BM&F, 2011). Regression analyses were conducted in order to reveal the correlation level between two selected variables. The results from the regression analysis were the following: the correlation between each parameter and ROA was 10.26%; the second regression analysis conducted measured the correlation between the independence level of top management team vis-à-vis board directors – namely, CEO relative power - and ROA, leading to a multiple R of 5.45%. Understanding that the scale is a simplification of the reality, the second part of the analysis transforms all the four parameters into dummy variables, excluding what could be called as an arbitrary scale. The ultimate result from this paper led to a multiple R of 28.44%, which implies that the combination of the variables are still not enough to translate the complex reality of organizations. Nonetheless, an important finding can be taken from this paper: two variables (percentage of outside directors and percentage of institutional investor ownership) are significant in the regression, with p-value lower than 10% and with negative coefficients. In other words, counter affirming what the literature very often portraits as being self-evident – stricter governance leads to higher performance – this paper has provided evidences to believe that the increase in the formal governance structure trough outside directors in the board and ownership by institutional investor might actually lead to worse performance. The section limitations and suggestions for future researches presents some reasons explaining why, although supported by strong theoretical background, this paper faced some challenging methodological assumptions, precluding categorical statements about the level of governance – measured by four selected parameters – and the financial return in terms of financial on assets.

Creating meanings, changing contexts: contested sustainability in the Brazilian beef industry

O objetivo desta pesquisa é compreender de que forma as organizações moldam seu ambiente, analisando por que algumas práticas tornam-se reconhecidas como ‘sustentáveis’ na indústria de carne bovina brasileira. O estudo dialoga com a literatura de institucionalismo organizacional ao apontar a necessidade de considerar a política (i.e. as negociações entre atores) e significados, a fim de entender como a estabilidade e a mudança institucional ocorrem em um contexto situado (i.e. em um tempo e espaço específicos). A pesquisa conclui que os entendimentos sobre o que poderia ser reconhecido como ‘sustentabilidade’ são o resultado de atores moldando o seu ambiente por meio de ações e interações que produzem significados. Seguindo uma abordagem de hegemonia, essas disputas não são apenas entre os atores que procuram vantagens recursivas, mas também procuram defender ou atacar as lógicas sociais que apoiam a posição dominante dos atores. Além disso, os atores exercem sua agência sobre as condições no presente (i.e. contexto situado), com base em um passado herdado e com o objetivo de produzir um futuro que eles imaginam. Para analisar tais processos uma abordagem de hegemonia entre atores e lógicas sociais foi desenvolvida para destacar a ordem de negociação, uma arena em que os atores lutam pela hegemonia. Como resultado de tais negociações, uma questão focal emerge, influenciando o discurso e interesses dos atores, bem como justificando as iniciativas, programas e tecnologias sobre tal questão; construindo, portanto, o consenso. Baseando-se em Realismo Crítico e Análise Crítica do Discurso, a pesquisa desenvolveu um estudo de caso longitudinal suportado por documentos públicos e confidenciais e entrevistas com especialistas, para examinar o caminho da sustentabilidade na indústria de carne bovina brasileira. Identificou-se três contextos diferentes para agência em relação à sustentabilidade. Enquanto no primeiro verifica-se um silêncio sobre práticas de sustentabilidade, o segundo enfatiza a emergência do desmatamento da Amazônia como uma questão focal, devido à agência do Greenpeace e MPF que força a indústria a desenvolver um sistema de monitoramento que rastreie seus fornecedores de gado de modo a evitar compra de suprimentos associadas ao desmatamento da Amazônia, dentre outras atividades ilegais. Finalmente, durante o terceiro contexto, o sistema de monitoramento permite que indústria de carne bovina se aproprie da sustentabilidade, assim o setor da carne passa a construir a sua legitimidade para influenciar sobre os riscos e oportunidades associadas ao contexto da sustentabilidade. Em termos de lógicas sociais, o desmatamento na Amazônia foi denunciado como um problema ambiental, nesta indústria, ancorado em algumas características da lógica do capitalismo, como a gestão de riscos, inovação e aumento da produtividade, cadeia de fornecimento global e governança. Embora este ataque questione a racionalidade da maximização racional lucro, impondo restrições ambientais para o comportamento das empresas, a solução desenvolvida é também ancorada sobre as mesmas características do capitalismo empregadas para atacá-lo. Como consequência, uma mudança gradual é ilustrada por uma transformação na ‘eficiência quantitativa’ do capitalismo, o aumento da produtividade devido à mudança da proporção de recursos consumidos para produção e à preocupação em evitar o desmatamento da Amazônia. No entanto, a ‘eficiência qualitativa’ do capitalismo é preservada uma vez que os grupos dominantes no poder ainda estão controlando os meios de produção e os recursos a eles associados (i.e. dinheiro, poder e legitimidade). Uma vez que estes processos de negociações são mediados pela racionalidade de se evitar risco aos negócios, consequentemente, a maximização do lucro, o núcleo duro da lógica do capitalismo é preservado. Portanto, os grupos dominantes mantêm sua hegemonia.

Faixa de suficiência para a cultura do algodão no centro-oeste do Brasil: II. micronutrientes

For to establish the micronutrients sufficiency range for the cotton, was used the method of mathematical chance for nutritional monitory data of three cotton growth locations in the Midwest of the Brazil, evaluating 152 areas. The method of mathematical chance was adequate for to establish cotton micronutrients references values, can be to help parameters were the conventional research cannot to prove in the short time. The micronutrients sufficiency range for productivity of 4000kg ha(-1) were, in the mg kg(-1), 41-89; 4-14; 90-230; 23-100; 25-50, and for 4500kg ha(-1) was 53-83; 4-12; 110-440; 40-60 e 25-50 for B, Cu, Fe, Mn and Zn, respectively. The sufficiency ranges indicated was similar to the existing official recommendations, however, is indicated the approach for to supply the specifics of the systems. The limitations of the method can be minimized with the expansion of the nutritional monitoring system in the cotton crops.

Sistema de supervisão aérea baseado em navegação visual para detecção de anomalias em instalações de petróleo e gás

This work deals with the development of a prototype of a helicopter quadrotor for monitoring applications in oil facilities. Anomaly detection problems can be resolved through monitoringmissions performed by a suitably instrumented quadrotor, i.e. infrared thermosensors should be embedded. The proposed monitoring system aims to reduce accidents as well as to make possible the use of non-destructive techniques for detection and location of leaks caused by corrosion. To this end, the implementation of a prototype, its stabilization and a navigation strategy have been proposed. The control strategy is based on dividing the problem into two control hierarchical levels: the lower level stabilizes the angles and the altitude of the vehicle at the desired values, while the higher one provide appropriate references signals to the lower level in order the quadrotor performs the desired movements. The navigation strategy for helicopter quadrotor is made using information provided by a acquisition image system (monocular camera) embedded onto the helicopter. Considering that the low-level control has been solved, the proposed vision-based navigation technique treats the problem as high level control strategies, such as, relative position control, trajectory generation and trajectory tracking. For the position control we use a control technique for visual servoing based on image features. The trajectory generation is done in a offline step, which is a visual trajectory composed of a sequence of images. For the trajectory tracking problem is proposed a control strategy by continuous servovision, thus enabling a navigation strategy without metric maps. Simulation and experimental results are presented to validate the proposal

Implementação de um sistema embarcado aplicado ao controle de pressão arterial

Hypertension is a dangerous disease that can cause serious harm to a patient health. In some situations the necessity to control this pressure is even greater, as in surgical procedures and post-surgical patients. To decrease the chances of a complication, it is necessary to reduce blood pressure as soon as possible. Continuous infusion of vasodilators drugs, such as sodium nitroprusside (SNP), rapidly decreased blood pressure in most patients, avoiding major problems. Maintaining the desired blood pressure requires constant monitoring of arterial blood pressure and frequently adjusting the drug infusion rate. Manual control of arterial blood pressure by clinical personnel is very demanding, time consuming and, as a result, sometimes of poor quality. Thus, the aim of this work is the design and implementation of a database of tuned controllers based on patients models, in order to find a suitable PID to be embedded in a Programmable Integrated Circuit (PIC), which has a smaller cost, smaller size and lower power consumption. For best results in controlling the blood pressure and choosing the adequate controller, tuning algorithms, system identification techniques and Smith predictor are used. This work also introduces a monitoring system to assist in detecting anomalies and optimize the process of patient care.

Desenvolvimento de um Sistema de Automação e Supervisão de Máquinas Dispensadoras Através da Internet

This paper aims to design and develop a control and monitoring system of vending machines, based on a Central Processing Unit with peripheral Internet communication. Coupled with the condom vending machines, a data acquisition module will be connected to the original circuits in order to collect and send, via internet, the information to the healthy government agencies, in the form of charts and reports. In the face of this, such agencies may analyze these data and compare them with the rates of reduction, in medium or long term, of the STD/AIDS in their respective regions, after the implementation of these vending machines, together with the conventional preventing programs. Reading the methodology, this paper is about an explaining and bibliography research, with the aspect of a qualitative-quantitative methodology, presenting a deductive method of approach and an indirect documentation technique research. About the results of the tests and simulations, we concluded that the implementation of this system will have the same success in any other type of dispenser machine