Feeds:
Posts
Comments

Archive for April, 2010

Date and venue:           Wednesday 28th April, 16:15, Geophysics Library, 5 Merrion Square, Dublin

Kevin Fleming of the Department of Spatial Sciences, Curtin University of Technology, Perth, Australia will regail an audience with tales of ‘Earth Observation Satellites and Australian hydrology’

Abstract

Earth Observation Satellites are a very useful tool in assessing the environment of Australia, given its vast area, range of climates and large areas without ground truth data.  This seminar examines observations from two satellites: the Tropical Rainfall Measuring Mission, and the Gravity Recovery and Climate Experiment. These are being used within a research program at Curtin University to assess changes in the terrestrial water storage of Australia.

Advertisements

Read Full Post »

A very clever idea from NASA, combining data archives, computing resources and a social networking site to promote cooperation and research among earth scientists. The NASA earth exchange seems the sort of model that is scalable both ways. Certainly with lots of researchers operating on a sort barter level at the moment, with research funds drying up – this sort of informal collaboration could keep outputs up for the next few years. One could start simply with a data “exchange and mart”, here’s the imagery we posses, heres the field work, here are our projects for next few years, and what we’d like to achieve and then instead of “looking for partners” researchers instead start conversations, what half realised idea you have? what half finished papers? whats struck you interesting in the last year, what policy stuff is coming up. That way informal collaboration begin, ideas are sparked and crucially resources that are sitting unused are made available. According to the nex wiki non-nasa reseacrhers will be able to join in sept. Perhaps the EO community here should start earlier with our own social network site – that crucilaly involves EO data users, earth scientist, environmentalists etc and not just EO reseachers.

Read Full Post »

Remote Sensing Data Analyst for the British Antarctic Survey.
It’s a 3 month post, more details here

Read Full Post »

Laser Scanning – a CPD Event

An evening event to inform you of developments in airborne, ground-based and vehicle-mounted laser scanning

Monday 26th April

18:00 – 20:00

Room 236, DIT Bolton Street

The Department of Spatial Information Sciences at DIT is pleased to invite you to this event as part of its Developments Series of Programmes for Industry.

Programme:

18:00 Developments in Laser scanning systems

Eugene McGovern and Kevin Mooney, Department of Spatial Information Sciences, DIT

18:45 Break

19:00 Experience with airborne and ground-based laser scannning

The Discovery Programme

19:30 Experience with vehicle-mounted laser scannning

NUI Maynooth

20:00 Close

To reserve a place at this event, please contact the Department of Spatial Information Sciences in DIT Bolton Street, at

spatial.planning@dit.ie

Read Full Post »

Volcanoe update!

The latest predictions from the VAAC show the ash cloud will be over Dublin by early tomorrow, I’d suggest get up a bit early and look at the sunrise, could be spectacular reflected of the ash cloud (of course conventional cloud could make it all a bit dull)

Read Full Post »

In light of the disruption today from the Icelandic ash cloud, here are a few relevant links:

The Volcanic Ash Alert Centre for this region is the london met office, here are the latest maps
You can learn more about volcanoes with this interactive map
This is only an inconvenience not a disaster..hasnt yet shown up on the Global disaster alert and Coordination System
but if for some you are worried here’s a helpful booklet from the USGS on surviving an ash fall

Read Full Post »

In this category we aim to highlight high quality peer reviwed published work, as its published. To start a aper from Maynooth looking at the vexed question of accuracy of objects:

Segmentation performance evaluation for object-based remotely sensed image analysis
Padraig Corcoran a; Adam Winstanley a;Peter Mooney a
 a National Centre for Geocomputation, Department of Computer Science, National University of Ireland Maynooth, Co. Kildare, Ireland

DOI: 10.1080/01431160902894475
 International Journal of Remote Sensing, Volume 31, Issue 3 April 2010 , pages 617 – 645

Abstract:

The initial step in most object-based classification methodologies is the application of a segmentation algorithm to define objects. Modelling the human visual process of object segmentation is a challenging task. Many theories in cognitive psychology propose that the human visual system (HVS) initially segments scenes into areas of uniform visual properties or primitive objects. If an accurate primitive-object segmentation algorithm is ever to be realized, a procedure must be in place to evaluate potential solutions. The most commonly used strategy to evaluate segmentation quality is a comparison against ground truth captured by human interpretation. A cognitive experiment reveals that ground truth captured in such a manner is at a larger scale than the desired primitive-object scale. To overcome this difficulty we consider the possibility of evaluating segmentation quality in an unsupervised manner without ground truth. Two requirements for any method which attempts to perform segmentation evaluation in such a manner are proposed, and the importance of these is illustrated by the poor performance of a metric which fails to meet them both. A novel metric, known as the spatial unsupervised (SU) metric, which meets both the requirements is proposed. Results demonstrate the SU metric to be a more reliable metric of segmentation quality compared to existing methodsThe initial step in most object-based classification methodologies is the application of a segmentation algorithm to define objects. Modelling the human visual process of object segmentation is a challenging task. Many theories in cognitive psychology propose that the human visual system (HVS) initially segments scenes into areas of uniform visual properties or primitive objects. If an accurate primitive-object segmentation algorithm is ever to be realized, a procedure must be in place to evaluate potential solutions. The most commonly used strategy to evaluate segmentation quality is a comparison against ground truth captured by human interpretation. A cognitive experiment reveals that ground truth captured in such a manner is at a larger scale than the desired primitive-object scale. To overcome this difficulty we consider the possibility of evaluating segmentation quality in an unsupervised manner without ground truth. Two requirements for any method which attempts to perform segmentation evaluation in such a manner are proposed, and the importance of these is illustrated by the poor performance of a metric which fails to meet them both. A novel metric, known as the spatial unsupervised (SU) metric, which meets both the requirements is proposed. Results demonstrate the SU metric to be a more reliable metric of segmentation quality compared to existing methods.

See here for electronic access

Read Full Post »

Older Posts »

%d bloggers like this: