scholarly journals ULTRA-HIGH RESOLUTION IMAGING OF FACADES AND VERTICAL INFRASTRUCTURE BY CARBORNE SAR AND AIRBORNE CSAR

2019 ◽  
Vol IV-2/W7 ◽  
pp. 129-136 ◽  
Author(s):  
S. Palm ◽  
R. Sommer ◽  
A. Tessmann ◽  
U. Stilla
Keyword(s):  
High Resolution ◽  
Single Channel ◽  
Theoretical Background ◽  
Data Sets ◽  
Radar Sensor ◽  
Data Set ◽  
High Bandwidth ◽  
Airborne Sar ◽  
High Level ◽  
Very High

<p><strong>Abstract.</strong> In this paper we propose a strategy to focus ultra-high resolution single channel carborne SAR and airborne circular SAR (CSAR) data to image facades and vertical infrastructure. We illustrate the related theoretical background and the design of an optimal focusing geometry for carborne SAR applications while using backprojection focusing techniques. Of particular interest is thereby the determination of the minimum distance and orientation of the facade to the radar sensor. Potential image distortions due to a wrong choice of these parameters are illustrated. Effects on the final resolution of the data due to the rotation of the focusing geometry compared to typical airborne SAR are discussed. We validated the strategy by driving on conventional roads illuminating facades with an experimental mobile radar mapping (MRM) sensor operating at 300 GHz. We further present an adapted version of the proposed strategy to focus vertical infrastructure in CSAR data sets. By extracting the center coordinate and the principal orientation of an object from GiS data, the focusing plane is designed arbitrarily in the 3D space. For the CSAR data set, a radar sensor particularly designed for circular flight trajectories operating at 94 GHz was evaluated. An electrical pylon was chosen as potential target. In both applications, the final images show a high level of detail. The combination of proposed strategy and radar sensor with very high bandwidth is capable of subcentimeter imaging of facades. The height, shape and dimensions of objects can be extracted directly from the image geometry at very high accuracy.</p>

Use of a Deep Belief Network for Small High-Level Abstraction Data Sets Using Artificial Intelligence with Rule Extraction

2018 ◽  
Vol 30 (12) ◽  
pp. 3309-3326 ◽  
Author(s):  
Yoichi Hayashi
Keyword(s):  
Rule Extraction ◽  
Deep Belief Network ◽  
Data Sets ◽  
Data Set ◽  
Simple Method ◽  
Learning Capability ◽  
Belief Network ◽  
Cancer Data ◽  
High Level ◽  
Very High

We describe a simple method to transfer from weights in deep neural networks (NNs) trained by a deep belief network (DBN) to weights in a backpropagation NN (BPNN) in the recursive-rule eXtraction (Re-RX) algorithm with J48graft (Re-RX with J48graft) and propose a new method to extract accurate and interpretable classification rules for rating category data sets. We apply this method to the Wisconsin Breast Cancer Data Set (WBCD), the Mammographic Mass Data Set, and the Dermatology Dataset, which are small, high-abstraction data sets with prior knowledge. After training these three data sets, our proposed rule extraction method was able to extract accurate and concise rules for deep NNs trained by a DBN. These results suggest that our proposed method could help fill the gap between the very high learning capability of DBNs and the very high interpretability of rule extraction algorithms such as Re-RX with J48graft.

scholarly journals Building Extraction in Very High Resolution Imagery by Dense-Attention Networks

2018 ◽  
Vol 10 (11) ◽  
pp. 1768 ◽  
Author(s):  
Hui Yang ◽  
Penghai Wu ◽  
Xuedong Yao ◽  
Yanlan Wu ◽  
Biao Wang ◽  
...  
Keyword(s):  
Deep Learning ◽  
High Resolution ◽  
Building Extraction ◽  
Learning Networks ◽  
Feature Maps ◽  
Low Level ◽  
High Resolution Imagery ◽  
Very High Resolution Imagery ◽  
High Level ◽  
Very High

Building extraction from very high resolution (VHR) imagery plays an important role in urban planning, disaster management, navigation, updating geographic databases, and several other geospatial applications. Compared with the traditional building extraction approaches, deep learning networks have recently shown outstanding performance in this task by using both high-level and low-level feature maps. However, it is difficult to utilize different level features rationally with the present deep learning networks. To tackle this problem, a novel network based on DenseNets and the attention mechanism was proposed, called the dense-attention network (DAN). The DAN contains an encoder part and a decoder part which are separately composed of lightweight DenseNets and a spatial attention fusion module. The proposed encoder–decoder architecture can strengthen feature propagation and effectively bring higher-level feature information to suppress the low-level feature and noises. Experimental results based on public international society for photogrammetry and remote sensing (ISPRS) datasets with only red–green–blue (RGB) images demonstrated that the proposed DAN achieved a higher score (96.16% overall accuracy (OA), 92.56% F1 score, 90.56% mean intersection over union (MIOU), less training and response time and higher-quality value) when compared with other deep learning methods.

scholarly journals EFFECTIVENESS OF GUIDANCE AND COUNSELING SERVICES BY USING THE APPLICATIONS INSTRUMENTATION AND ACTIVITY DATA SET IN SMA NEGERI 1 METRO LESSON 2009/2010

2014 ◽  
Vol 4 (1) ◽  
pp. 35
Author(s):  
Agus Wibowo
Keyword(s):  
Qualitative Research ◽  
Counseling Services ◽  
Guidance And Counseling ◽  
Data Sets ◽  
Research Subjects ◽  
Activity Data ◽  
Data Set ◽  
Research Level ◽  
Data Collection Technique ◽  
High Level

Abstract: Implementation of guidance and counseling services should be based on the needs and problems of students, so the effectiveness of the service will be achieved to the fullest. But the reality is a lot of implementation of guidance and counseling services in schools, do not notice it. So that the completion of the problems experienced by students sama.Berangkat always use the services of this, the research level of effectiveness of guidance and counseling that implementation has been using the application activity instrumentation and data sets as the basis for an implementation of the service. The method used is a qualitative research subjects that teachers BK and Students at SMA Negeri 1 Metro. Data collection technique through interview, observation and documentation. Research results show that by utilizing activity instrumentation applications and data sets, the counseling services have a high level of effectiveness. In carrying out the service, BK teachers can identify problems and needs experienced by students, so that the efforts of the assistance provided to be more precise, and problem students can terentaskan optimally.Keyword: Guidance and Counseling, Instrumentation Applications, Data Association

Collecting training data to map forest management at global scale

2021 ◽  
Author(s):  
Myroslava Lesiv ◽  
Dmitry Schepaschenko ◽  
Martina Dürauer ◽  
Marcel Buchhorn ◽  
Ivelina Georgieva ◽  
...  
Keyword(s):  
Forest Management ◽  
High Resolution ◽  
Remotely Sensed ◽  
Global Scale ◽  
Google Earth ◽  
Training Data ◽  
Tree Cover ◽  
Data Set ◽  
The World ◽  
Very High

&lt;p&gt;Spatially explicit information on forest management at a global scale is critical for understanding the current status of forests for sustainable forest management and restoration. Whereas remotely sensed based datasets, developed by applying ML and AI algorithms, can successfully depict tree cover and other land cover types, it has not yet been used to depict untouched forest and different degrees of forest management. We show for the first time that with sufficient training data derived from very high-resolution imagery a differentiation within the tree cover class of various levels of forest management is possible.&lt;/p&gt;&lt;p&gt;In this session, we would like to present our approach for labeling forest related training data by using Geo-Wiki application (https://www.geo-wiki.org/). Moreover, we would like to share a new open global training data set on forest management we collected from a series of Geo-Wiki campaigns. In February 2019, we organized an expert workshop to (1) discuss the variety of forest management practices that take place in different parts of the world; (2) generalize the definitions for the application at global scale; (3) finalize the Geo-Wiki interface for the crowdsourcing campaigns; and (4) build a data set of control points (or the expert data set), which we used later to monitor the quality of the crowdsourced contributions by the volunteers. We involved forest experts from different regions around the world to explore what types of forest management information could be collected from visual interpretation of very high-resolution images from Google Maps and Microsoft Bing, in combination with Sentinel time series and Normalized Difference Vegetation Index (NDVI) profiles derived from Google Earth Engine (GEE). Based on the results of this analysis, we expanded these campaigns by involving a broader group of participants, mainly people recruited from remote sensing, geography and forest research institutes and universities.&lt;/p&gt;&lt;p&gt;In total, we collected forest data for approximately 230 000 locations globally. These data are of sufficient density and quality and therefore could be used in many ML and AI applications for forests at regional and local scale.&amp;#160; We also provide an example of ML application, a remotely sensed based global forest management map at a 100 m resolution (PROBA-V) for the year 2015. It includes such classes as intact forests, forests with signs of human impact, including clear cuts and logging, replanted forest, woody plantations with a rotation period up to 15 years, oil palms and agroforestry. The results of independent statistical validation show that the map&amp;#8217;s overall accuracy is 81%.&lt;/p&gt;

scholarly journals Using kites for 3-D mapping of gullies at decimetre-resolution over several square kilometres: a case study on the Kamech catchment, Tunisia

2018 ◽  
Vol 18 (6) ◽  
pp. 1567-1582 ◽  
Author(s):  
Denis Feurer ◽  
Olivier Planchon ◽  
Mohamed Amine El Maaoui ◽  
Abir Ben Slimane ◽  
Mohamed Rached Boussema ◽  
...  
Keyword(s):  
High Resolution ◽  
Aerial Photography ◽  
Airborne Lidar ◽  
Data Set ◽  
Topographic Data ◽  
Independent Field ◽  
Wide Range ◽  
Laser Systems ◽  
High Level

Abstract. Monitoring agricultural areas threatened by soil erosion often requires decimetre topographic information over areas of several square kilometres. Airborne lidar and remotely piloted aircraft system (RPAS) imagery have the ability to provide repeated decimetre-resolution and -accuracy digital elevation models (DEMs) covering these extents, which is unrealistic with ground surveys. However, various factors hamper the dissemination of these technologies in a wide range of situations, including local regulations for RPAS and the cost for airborne laser systems and medium-format RPAS imagery. The goal of this study is to investigate the ability of low-tech kite aerial photography to obtain DEMs with decimetre resolution and accuracy that permit 3-D descriptions of active gullying in cultivated areas of several square kilometres. To this end, we developed and assessed a two-step workflow. First, we used both heuristic experimental approaches in field and numerical simulations to determine the conditions that make a photogrammetric flight possible and effective over several square kilometres with a kite and a consumer-grade camera. Second, we mapped and characterised the entire gully system of a test catchment in 3-D. We showed numerically and experimentally that using a thin and light line for the kite is key for a complete 3-D coverage over several square kilometres. We thus obtained a decimetre-resolution DEM covering 3.18 km2 with a mean error and standard deviation of the error of +7 and 22 cm respectively, hence achieving decimetre accuracy. With this data set, we showed that high-resolution topographic data permit both the detection and characterisation of an entire gully system with a high level of detail and an overall accuracy of 74 % compared to an independent field survey. Kite aerial photography with simple but appropriate equipment is hence an alternative tool that has been proven to be valuable for surveying gullies with sub-metric details in a square-kilometre-scale catchment. This case study suggests that access to high-resolution topographic data on these scales can be given to the community, which may help facilitate a better understanding of gullying processes within a broader spectrum of conditions.

scholarly journals Reconstruction of Axial Tomographic High Resolution Data from Confocal Fluorescence Microscopy: A Method for Improving 3D FISH Images

2000 ◽  
Vol 20 (1) ◽  
pp. 7-15 ◽  
Author(s):  
R. Heintzmann ◽  
G. Kreth ◽  
C. Cremer
Keyword(s):  
High Resolution ◽  
Laser Scanning ◽  
Axial Direction ◽  
Laser Scanning Microscopy ◽  
Three Dimensions ◽  
Confocal Laser ◽  
Data Sets ◽  
Data Set ◽  
High Resolution Data ◽  
3D Data

Fluorescent confocal laser scanning microscopy allows an improved imaging of microscopic objects in three dimensions. However, the resolution along the axial direction is three times worse than the resolution in lateral directions. A method to overcome this axial limitation is tilting the object under the microscope, in a way that the direction of the optical axis points into different directions relative to the sample. A new technique for a simultaneous reconstruction from a number of such axial tomographic confocal data sets was developed and used for high resolution reconstruction of 3D‐data both from experimental and virtual microscopic data sets. The reconstructed images have a highly improved 3D resolution, which is comparable to the lateral resolution of a single deconvolved data set. Axial tomographic imaging in combination with simultaneous data reconstruction also opens the possibility for a more precise quantification of 3D data. The color images of this publication can be accessed from http://www.esacp.org/acp/2000/20‐1/heintzmann.htm. At this web address an interactive 3D viewer is additionally provided for browsing the 3D data. This java applet displays three orthogonal slices of the data set which are dynamically updated by user mouse clicks or keystrokes.

Bayesian networks: Theory, applications and sensitivity issues

2017 ◽  
Vol 01 (01) ◽  
pp. 1630014 ◽  
Author(s):  
Ron S. Kenett
Keyword(s):  
Bayesian Networks ◽  
Expert Opinion ◽  
Data Science ◽  
Information Quality ◽  
Theoretical Background ◽  
Hospital Environment ◽  
Statistical Tool ◽  
Data Set ◽  
Operational Variables ◽  
High Level

This chapter is about an important tool in the data science workbench, Bayesian networks (BNs). Data science is about generating information from a given data set using applications of statistical methods. The quality of the information derived from data analysis is dependent on various dimensions, including the communication of results, the ability to translate results into actionable tasks and the capability to integrate various data sources [R. S. Kenett and G. Shmueli, On information quality, J. R. Stat. Soc. A 177(1), 3 (2014).] This paper demonstrates, with three examples, how the application of BNs provides a high level of information quality. It expands the treatment of BNs as a statistical tool and provides a wider scope of statistical analysis that matches current trends in data science. For more examples on deriving high information quality with BNs see [R. S. Kenett and G. Shmueli, Information Quality: The Potential of Data and Analytics to Generate Knowledge (John Wiley and Sons, 2016), www.wiley.com/go/information_quality.] The three examples used in the chapter are complementary in scope. The first example is based on expert opinion assessments of risks in the operation of health care monitoring systems in a hospital environment. The second example is from the monitoring of an open source community and is a data rich application that combines expert opinion, social network analysis and continuous operational variables. The third example is totally data driven and is based on an extensive customer satisfaction survey of airline customers. The first section is an introduction to BNs, Sec. 2 provides a theoretical background on BN. Examples are provided in Sec. 3. Section 4 discusses sensitivity analysis of BNs, Sec. 5 lists a range of software applications implementing BNs. Section 6 concludes the chapter.

scholarly journals Fennec dust forecast intercomparison over the Sahara in June 2011

2016 ◽  
Vol 16 (11) ◽  
pp. 6977-6995 ◽  
Author(s):  
Jean-Pierre Chaboureau ◽  
Cyrille Flamant ◽  
Thibaut Dauhut ◽  
Cécile Kocha ◽  
Jean-Philippe Lafore ◽  
...  
Keyword(s):  
High Resolution ◽  
Surface Wind ◽  
Dust Emission ◽  
Airborne Lidar ◽  
Data Sets ◽  
Density Currents ◽  
Data Set ◽  
Near Surface ◽  
Western Sahara ◽  
Deep Well

Abstract. In the framework of the Fennec international programme, a field campaign was conducted in June 2011 over the western Sahara. It led to the first observational data set ever obtained that documents the dynamics, thermodynamics and composition of the Saharan atmospheric boundary layer (SABL) under the influence of the heat low. In support to the aircraft operation, four dust forecasts were run daily at low and high resolutions with convection-parameterizing and convection-permitting models, respectively. The unique airborne and ground-based data sets allowed the first ever intercomparison of dust forecasts over the western Sahara. At monthly scale, large aerosol optical depths (AODs) were forecast over the Sahara, a feature observed by satellite retrievals but with different magnitudes. The AOD intensity was correctly predicted by the high-resolution models, while it was underestimated by the low-resolution models. This was partly because of the generation of strong near-surface wind associated with thunderstorm-related density currents that could only be reproduced by models representing convection explicitly. Such models yield emissions mainly in the afternoon that dominate the total emission over the western fringes of the Adrar des Iforas and the Aïr Mountains in the high-resolution forecasts. Over the western Sahara, where the harmattan contributes up to 80 % of dust emission, all the models were successful in forecasting the deep well-mixed SABL. Some of them, however, missed the large near-surface dust concentration generated by density currents and low-level winds. This feature, observed repeatedly by the airborne lidar, was partly forecast by one high-resolution model only.

Sign in / Sign up

Export Citation Format

Share Document