Development of an Outline Detection Tool (ODT) in QGIS environment for risk mitigation applications

Vision techniques are presently developed, within a GIS environment, to detect any type of structural and infrastructure damage caused by natural catastrophic events. The aim of this paper is to report on the implementation of a software tool which is able to identify the border of any system that could be damaged by a negative event. The potential of an open source tool named “Magic Wand” is investigated in order to create an innovative procedure which allows to quickly select buildings and artefacts in disaster areas. The pixels of satellite images are the input that the tool requires. Some examples are presented in order to provide the main features of the proposed function.

Download Full-text

Z-MERT: A Fully Configurable Open Source Tool for Minimum Error Rate Training of Machine Translation Systems

Prague Bulletin of Mathematical Linguistics ◽

10.2478/v10108-009-0018-2 ◽

2009 ◽

Vol 91 (1) ◽

pp. 79-88 ◽

Cited By ~ 23

Author(s):

Omar Zaidan

Keyword(s):

Open Source ◽

Machine Translation ◽

Error Rate ◽

Software Tool ◽

Minimum Error ◽

Open Source Tool ◽

Series Of Experiments ◽

Evaluation Metric ◽

Minimum Error Rate Training ◽

Translation Systems

Z-MERT: A Fully Configurable Open Source Tool for Minimum Error Rate Training of Machine Translation Systems We introduce Z-MERT, a software tool for minimum error rate training of machine translation systems (Och, 2003). In addition to being an open source tool that is extremely easy to compile and run, Z-MERT is also agnostic regarding the evaluation metric, fully configurable, and requires no modification to work with any decoder. We describe Z-MERT and review its features, and report the results of a series of experiments that examine the tool's runtime. We establish that Z-MERT is extremely efficient, making it well-suited for time-sensitive pipelines. The experiments also provide an insight into the tool's runtime in terms of several variables (size of the development set, size of produced N-best lists, etc).

Download Full-text

DEVELOPMENT OF AN OPEN SOURCE, MACHINE LEARNING BASED TOOLSET FOR THE IDENTIFICATION OF DIKES IN SATELLITE IMAGES THROUGH SEMANTIC SEGMENTATION

10.1130/abs/2020am-357672 ◽

2020 ◽

Author(s):

Ryan Gray ◽

◽

Tushar Mittal

Keyword(s):

Machine Learning ◽

Open Source ◽

Satellite Images ◽

Semantic Segmentation

Download Full-text

Dynamic simulation of CIG in large unbalanced distribution systems using an open source tool

IET Generation Transmission & Distribution ◽

10.1049/iet-gtd.2018.6089 ◽

2019 ◽

Vol 13 (9) ◽

pp. 1638-1645 ◽

Cited By ~ 1

Author(s):

Ziwei Yu ◽

Yingying Tang ◽

Tong Yao ◽

Raja Ayyanar

Keyword(s):

Open Source ◽

Dynamic Simulation ◽

Distribution Systems ◽

Open Source Tool

Download Full-text

Efficient onboard image motion compensation for orbital inclination and eccentricity of geostationary weather satellites

Proceedings of the Institution of Mechanical Engineers Part G Journal of Aerospace Engineering ◽

10.1177/0954410019900706 ◽

2020 ◽

Vol 234 (7) ◽

pp. 1311-1325

Author(s):

MP Ramachandran ◽

MK Agarwal ◽

DA Daniel

Keyword(s):

Motion Compensation ◽

Satellite Images ◽

Software Tool ◽

Image Motion ◽

Wind Vector ◽

Orbital Inclination ◽

Scanning Mirror ◽

The Earth ◽

Detail Image ◽

Geographical Locations

Image registration is important in geostationary weather satellites. Achieving consistent registration of the images with respect to the geographical locations on the Earth is here of interest. The consistency in the registration between the images is affected whenever the orbital inclination and eccentricity are not zero. The imaging payload has a two-axis scanning mirror to capture the Earth image. The above orbital effects together with scan mirror pointing direction are the factors that cause the misregistration. This paper presents an onboard algorithm that provides the scan compensation angles due to the above factors and achieves consistent registration. The compensation varies every second, which is the time taken for each scan. Hence it is preferred to have computations onboard than to have ground based bulk uplinks for the scan compensation. The paper presents an algorithm that is useful, say, when (i) the onboard computing capabilities are limited, (ii) the navigation accuracies are coarse and (iii) the image resampling is not preferred on the ground and the payload data are directly used for weather applications. The paper also discusses the tests that were carried on the onboard software in order to validate its performance in achieving the consistent registration before launch. This is done by using another independent software tool which is also described in detail. Image motion algorithm was invoked for a couple of days in INSAT 3DR. The atmospheric wind vector deduced directly from the satellite images is given at the end.

Download Full-text