The up to date mapping data is of great importance in social services and disaster relief as well as in city planning. The vast amounts of data and the constant increase of geographical changes lead to large loads of continuous manual analysis. This thesis takes the process of updating maps and breaks it down to the problem of discovering buildings by comparing different machine learning methods to automate the finding of buildings. The chosen methods, YOLOv3 and Mask R-CNN, are based on Region Convolutional Neural Network(R-CNN) due to their capabilities of image analysis in both speed and accuracy. The image data supplied by Lantmäteriet makes up the training and testing data; this data is then used by the chosen machine learning methods. The methods are trained at different time limits, the generated models are tested and the results analysed. The results lay ground for whether the model is reasonable to use in a fully or partly automated system for updating mapping data from aerial imagery. The tested methods showed volatile results through their first hour of training, with YOLOv3 being more so than Mask R-CNN. After the first hour and until the eight hour YOLOv3 shows a higher level of accuracy compared to Mask R-CNN. For YOLOv3, it seems that with more training, the recall increases while precision decreases. For Mask R-CNN, however, there is some trade-off between the recall and precision throughout the eight hours of training. While there is a 90 % confidence interval that the accuracy of YOLOv3 is decreasing for each hour of training after the first hour, the Mask R-CNN method shows that its accuracy is increasing for every hour of training,however, with a low confidence and can therefore not be scientifically relied upon. Due to differences in setups the image size varies between the methods, even though they train and test on the same areas; this results in a fair evaluation where YOLOv3 analyses one square kilometre 1.5 times faster than the Mask R-CNN method does. Both methods show potential for automated generation of footprints, however, the YOLOv3 method solely generates bounding boxes, leaving the step of polygonization to manual work while the Mask R-CNN does, as the name implies, create a mask of which the object is encapsulated. This extra step is thought to further automate the manual process and with viable results speed up the updating of map data. / Uppdaterad kartdata är av stor betydelse för sociala tjänster och katastrofhjälp såväl som inom stadsplanering. De enorma mängderna data och den ständiga ökningen av geografiska förändringar leder till mycket arbete för kontinuerlig manuell analys. Denna avhandling kommer att behandla detta problem med att uppdatera kartor, bryta ned det till det specifika problemet att upptäcka byggnader och ur den synvinkelen jämföra olika maskininlärningsmetoder för automatisera detektering av byggnader. De valda metoderna, YOLOv3 och Mask R-CNN, är baserade på Region Convolutional Neural Network (R-CNN) på grund av dess förmåga av bildanalys i både hastighet och träffsäkerhet. Bildmaterial från Lantmäteriet utgör tränings- och testdatan, denna data används sedan av de utvalda maskininlärningmetoderna. Metoderna tränas med olika tidsgränser och de genererade modellerna testas och resultaten analyseras. Resultaten lägger grund för huruvida modellen är rimlig att använda i ett helt eller delvis automatiserat system för uppdatering av kartdata från flygbilder. De testade metoderna visade varierande resultat under sin första timmes träning, med YOLOv3 mer så än Mask R-CNN. Efter den första timmen fram till den åttonde timmen visar YOLOv3 en högre nivå av precision jämfört med Mask R-CNN. För YOLOv3 ser det ut som att mer träning ökar recall samtidigt som precision minskar. För Mask R-CNN är det emellertid en avvägning mellan recall och precision under de åtta timmarnas träning. Medan det finns en 90 % konfidens att accuracy minskar med YOLOv3 för varje timmes träning efter första timmen så visar Mask R-CNN-metoden att dess accuracy ökar för varje timmes träning, det är dock med låg konfidens och har därmed inte vetenskapligt stöd. På grund av skillnader i konfigurationer varierar bildstorleken mellan metoderna, de tränar och testar dock på samma områden för att ge en rättvis jämförelse. I dessa test analyserar YOLOv3 en kvadratkilometer 1.5 gånger snabbare än Mask R-CNN. Båda metoderna visar potential för en automatiserad generering av footprints. Dock så genererar YOLOv3-metoden endast en bounding box, vilket gör att polygoniseringen återstår för manuellt arbete medan Mask R-CNN, som namnet antyder, skapar en mask som objektet inkapslas i. Detta extrasteg är tänkt att automatisera den manuella processen och med rimliga resultat påskynda uppdateringen av kartdata.

The increasing demand of mankind by use of water quality and sufficient volume to meet their needs in their different uses, shows that water resources must be managed and preserved. A watershed is the physical unit that integrates the entire flow of water that makes up the hydrological cycle, but its use is being carried out in a disorderly way, without complying with the compliance of their characteristics and fragilities. Brazil has a very advanced legislation regarding the environment, especially related to water, and a growing awareness of the need to demarcate areas in order to safeguard the different ecosystems found in this vast territory. In Brazil, many of the areas selected to include environmental protection within its boundaries major resources stocks. However, deforestation and poor land use along these water resources may affect and jeopardize the supply and quality of water throughout their watersheds, affecting populations downstream. The purpose of this paper is to present an Integrated Information System, embodied in a Geographic Information System (GIS), which adds the elements required for management and conservation of water resources in protected areas. In the development of GIS will be used the new structure and Geographical Data Vector (EDGV) and the Geospatial Metadata Profile of Brazil (MGB), approved in 2008 by the National Cartography (Conc), aiming to standardize the structures spatial data, facilitating data sharing, interoperability and rationalization of resources among data producers and users of geospatial information. In the proposed system shows that the systematic organization into a Geographic Information System (GIS) of water resources and the various elements and factors about them impacting facilitates the analysis and subsequent management of water resources conservation units (CUs). Through these spatial analysis and comparative data were obtained as a result of the making of a final map containing an indication of inharmonious elements in the area. Indicated the best places to collect water in order to assess its quality and, eventually, we proposed a new limit for the conservation unit. The final map will serve as support for management to operate and monitor the areas according to their real possibilities and implement corrective measures and mitigation. Additionally, information based on the database associated with the GIS will enable the integration of various public and private entities that deal with specific issues, enabling interoperability and cooperation between them, as well as confirming that the use of Geotechnology is an appropriate option for the best management of the areas involved. The techniques and procedures used can be applied in the management of water resources from other areas. We used the União Biological Reserve - Rebio Union - located in the State of Rio de Janeiro, as a case study.

Submitted by LUCAS TELES DE FARIA null (lucas.teles.faria@gmail.com) on 2016-04-06T05:42:40Z No. of bitstreams: 1 TESE_DOUTORADO_ENG.ELETRICA_LUCAS-TELES-DE-FARIA.pdf: 4869615 bytes, checksum: 5cea8811bb7d053c5440e3d6fb5d55cd (MD5) / Approved for entry into archive by Juliano Benedito Ferreira (julianoferreira@reitoria.unesp.br) on 2016-04-07T19:51:32Z (GMT) No. of bitstreams: 1 faria_lt_dr_ilha.pdf: 4869615 bytes, checksum: 5cea8811bb7d053c5440e3d6fb5d55cd (MD5) / Made available in DSpace on 2016-04-07T19:51:32Z (GMT). No. of bitstreams: 1 faria_lt_dr_ilha.pdf: 4869615 bytes, checksum: 5cea8811bb7d053c5440e3d6fb5d55cd (MD5) Previous issue date: 2016-02-26 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Neste trabalho o espaço geográfico é incorporado ao estudo das perdas não técnicas. Os trabalhos avaliados em perdas comumente não consideram a localização espacial das mesmas de forma explícita. No entanto, o estudo das características do lugar onde elas ocorrem pode trazer informações imprescindíveis para melhor compreensão do problema. O espaço é incorporado via técnicas de análise espacial de dados geográficos. A saber: análise espacial de padrões de pontos e análise espacial de dados agregados por áreas. A localização das perdas é obtida através de dados de inspeções reais georreferenciados obtidos a partir de uma concessionária de energia elétrica. Os atributos socioeconômicos do censo demográfico e da rede de distribuição de energia do lugar onde ocorrem as perdas são considerados via técnicas de regressões espaciais. São elas: modelo aditivo generalizado (GAM) e regressão geograficamente ponderada (GWR). Esses atributos são as variáveis independentes das regressões espaciais e auxiliam na explicação da disposição das perdas no espaço geográfico do município em estudo. Essas regressões são combinadas com as cadeias de Markov para produção de mapas de probabilidades de perdas. Esses mapas indicam as subáreas do município que são mais vulneráveis às perdas em termos probabilísticos. Por meio deles, estima-se a evolução das perdas não técnicas no espaço geográfico do município ao longo do tempo. Os mapas de probabilidade de perdas são uma ferramenta gráfica, de fácil interpretação e que auxiliam no planejamento de uma série de ações de prevenção e combate às perdas. Este estudo foi realizado em um município de porte médio do interior paulista com aproximadamente 81 mil unidades consumidoras, sendo que os resultados das simulações foram comparados com dados reais de inspeções em campo. A taxa de acerto para estimação das áreas vulneráveis às perdas via modelo aditivo generalizado (GAM) e cadeias e Markov foi superior a 80%. / In this work the geographic space is incorporated into the study of non-technical losses. Studies on non-technical losses do not often consider the spatial location of them explicitly. However, the study of the characteristics of the place where they occur can provide essential information to better understanding of the problem. The space is incorporated via spatial analysis techniques of geographical data; to know: spatial analysis of point patterns and spatial analysis of data aggregated by areas. The location of the losses is determined via georeferenced inspections data obtained from an electrical power utility. Socioeconomic attributes of the census and the distribution network of energy of the place where the losses occur are considered using the spatial regressions techniques; namely: generalized additive model (GAM) and geographically weighted regression (GWR). These attributes are the independent variables of spatial regressions and assist in the provision of the explanation of the losses in the geographical space of the city under study. These regressions are combined with Markov chains to produce the loss probability maps. These maps show the city subareas that are more vulnerable to losses in probabilistic terms. Through them, the evolution of non-technical losses in the geographical area of the city over the time is estimated. The loss probability maps are a graphical tool, easy to interpret and to assist in planning a series of actions to prevent and combat to losses. This study was conducted in a medium-sized city of São Paulo with about 81,000 consumer units, and the simulation results were compared with real data obtained in field inspections. The hit rate for the estimation of areas vulnerable to losses via generalized additive model (GAM) and Markov chains surpasses 80%.

The increasing demand of mankind by use of water quality and sufficient volume to meet their needs in their different uses, shows that water resources must be managed and preserved. A watershed is the physical unit that integrates the entire flow of water that makes up the hydrological cycle, but its use is being carried out in a disorderly way, without complying with the compliance of their characteristics and fragilities. Brazil has a very advanced legislation regarding the environment, especially related to water, and a growing awareness of the need to demarcate areas in order to safeguard the different ecosystems found in this vast territory. In Brazil, many of the areas selected to include environmental protection within its boundaries major resources stocks. However, deforestation and poor land use along these water resources may affect and jeopardize the supply and quality of water throughout their watersheds, affecting populations downstream. The purpose of this paper is to present an Integrated Information System, embodied in a Geographic Information System (GIS), which adds the elements required for management and conservation of water resources in protected areas. In the development of GIS will be used the new structure and Geographical Data Vector (EDGV) and the Geospatial Metadata Profile of Brazil (MGB), approved in 2008 by the National Cartography (Conc), aiming to standardize the structures spatial data, facilitating data sharing, interoperability and rationalization of resources among data producers and users of geospatial information. In the proposed system shows that the systematic organization into a Geographic Information System (GIS) of water resources and the various elements and factors about them impacting facilitates the analysis and subsequent management of water resources conservation units (CUs). Through these spatial analysis and comparative data were obtained as a result of the making of a final map containing an indication of inharmonious elements in the area. Indicated the best places to collect water in order to assess its quality and, eventually, we proposed a new limit for the conservation unit. The final map will serve as support for management to operate and monitor the areas according to their real possibilities and implement corrective measures and mitigation. Additionally, information based on the database associated with the GIS will enable the integration of various public and private entities that deal with specific issues, enabling interoperability and cooperation between them, as well as confirming that the use of Geotechnology is an appropriate option for the best management of the areas involved. The techniques and procedures used can be applied in the management of water resources from other areas. We used the União Biological Reserve - Rebio Union - located in the State of Rio de Janeiro, as a case study.

This master's project deals with location based systems and their application in information distribution. The work discuss kinds of getting actual location. Then it focuses on problems of developing applications and description of Microsoft technologies for mobile devices. Based on obtained theoretical information we present design, implementation and testing of our own location based system.