A A A

Daniel FIsher - MSSL Imaging Group

dnf

Dan Fisher

MSSL Imaging Group

PhD Student

Tel: +53 3014962805

Internal: 2286

Email: dnf {at} mssl.ucl.ac.uk

Room: N102

Research Interests

  • The application of machine vision techniques and approaches to remote sensing tasks. In particular stereo image processing and feature detection and tracking.
  • Validation and error characterisation of datasets from both remote sensing systems and models.
  • The generation of observational climatologies of atmospheric parameters for improving the understanding of associated processes and characteristics, particularly in Polar Regions.
  • Image classification and image understanding, and the application of machine learning methodologies to remote sensing tasks.
  • The Earth’s climate system and the role that clouds play in its perturbation

Qualifications

B.Sc. Geography, University College London (2006)

M.Sc. Remote Sensing, University College London (2009)

Research Experience

PhD Project

"Application of Stereo-Photogrammetric Methods to the Advanced Along Track Scanning Radiometer for the Atmospheric Sciences."

SupervisorsProf. Jan-Peter Muller & Dr Caroline Poulsen

My PhD research project examined the potential for exploiting the unique multispectral and multi-look stereo capabilities of the British (A)ATSR(2) instruments for the derivation of a number of novel datasets for potential application within the atmospheric sciences.

A stereo-derived cloud fraction by altitude (CFBA) climatology has been generated over Greenland for the entire AATSR time series for the first time. This CFBA product, taking influence from the MISR product of the same name, provides height stratified estimates of cloud fraction. Due to the AATSR thermal imaging capabilities, important new insights into high cloud characteristics over Greenland appear to have been uncovered.

The CFBA product was generated using an implementation of the non-parametric census stereo matching algorithm with sub-pixel enhancements. In extensive inter-comparisons against the GMTED2010 digital elevation model, higher resolution stereo measurements from MISR, and lidar observations from CALIOP, then census stereo matcher has been shown to significantly outperform, in terms of accuracy, previous stereo matching algorithms applied to the ATSR instruments.

In order to obtain accurate stereo measurements from the ATSR instruments, the two instrument looks need to be accurately aligned, or co-registered. This co-registration was ensured, to pixel level accuracy or better, through the design and application of an automated image co-registration algorithm based on the SIFT feature detector.

A further study undertaken involved the assessment of a tandem satellite stereo systems for the accurate determination and height assignment of tropospheric winds through a cloud tracking approach. The study exploited the tandem operation of ATSR-2 and AATSR to evaluate this potential along with the Farneback optical flow algorithm. The results are extremely encouraging with winds observed to a potential accuracy of 0.5 m/s, at 1km horizontal and vertical resolution. This approach shows excellent potential for application to the upcoming SLSTR tandem operation on Sentinel-3A and -3B.

ALANIS Smoke Plumes Project

Consortium: UCL, NOVELTIS, WUR, JRC, LATMOS

The project aimed at advancing towards the development and validation of
novel EO-based multi-mission products and their integration into
suitable land-atmosphere coupled models (responding directly to the
specific scientific requirements of the iLEAPS community) for improving
the estimation of plume injection height of biomass burning events
occurred in boreal Eurasia and reducing current uncertainties in related greenhouse-gas and aerosol dispersion forecast. The outcomes of the project can be found here ALANIS.

UCL's main involvement was with the development and validation of a 'smoke-plume tuned' stereo matching algorithm and a smoke plume masking algorithm.

GlobAlbedo CCN

Consortium: UCL, Swansea University, Brockmann Consult, Freie Universität Berlin

The contract change notice of the GlobAlbedo project involved the determination of sea-ice albedo from ESA assets. This extension necessitated the application of AATSR data in order to effectively differentiate between cloud and sea-ice features.

My involvement in this project centred around the integration of AATSR data into the albedo retrieval, which, due to the poor geo-referencing of the AATSR instrument, was previously unattainable and in fact led to a degradation in the albedo retrievals.

To enable the integration of the AATSR data, my PhD research on image co-registration was extended to that of a multi-sensor task, in this instance, AATSR to MERIS. AATSR data was effectively co-registered to that of MERIS to better than pixel level accuracy. Enablind the effective discrimination between cloud and sea-ice features.

Stereo+ORAC

In this currently ongoing project, the census stereo algorithm with sub-pixel enhancements developed during my PhD research is being applied as an a priori input into the ORAC optimal estimation cloud retrieval algorithm. The aim being to improve the retrieval quality, particularly in terms of cloud height and cloud fraction estimates.

Publications

Journal Articles

Fisher, D., Muller, J. P., & Yershov, V. N. (2014). Automated Stereo Retrieval of Smoke Plume Injection Heights and Retrieval of Smoke Plume Masks from AATSR and their Assessment with CALIPSO and MISR. Geoscience and Remote Sensing, IEEE Transactions on, 52(2), pp. 1249-1258 doi:10.1109/TGRS.2013.2249073.

Krol, M., Peters, W., Hooghiemstra, P., George, M., Clerbaux, C., Hurtmans, D., McInerney, D., Sedano, F., Bergamaschi, P., El Hajj, M., Kaiser, J., van der Werf, G., Fisher, D., Yershov, V., Muller, J. P. (2013). How much CO was emitted by the 2010 fires around Moscow? Atmospheric Chemistry and Physics, 13(9), pp. 4737-4747.

Fisher, D., & Muller, J. P. (2013). Global warping coefficients for improving ATSR co-registration. Remote Sensing Letters, 4(2), pp. 151-160. doi: 10.1080/2150704X.2012.713138.

In preparation:

Fisher, D. and J-P. Muller, Macrophysical Cloud Characteristics over Greenland from AATSR Stereo. In Prep. Planned submission to Atmospheric Measurement Techniques.

Fisher, D. and J-P. Muller, Cloud Tracked Winds from AATSR-ATSR2 Tandem. In Prep. Planned submission to Atmospheric Science Letters.

Conference Proceedings

Fisher, D., J-P. Muller, (2013). “Improving AATSR geolocation to allow application in a multi-sensor science task.” ESA Living Planet Symposium. Edinburgh, U.K. 9-13 Sept. (ORAL)

Fisher, D., J-P. Muller, 2012. “Stereo derived cloud top height climatology over Greenland from 20 years of the Along Track Scanning Radiometer (ATSR) Instruments”. The XII Congress of the International Society for Photogrammetry and Remote Sensing. Melbourne, Australia. 25 August – 1 September. (ORAL)

Fisher, D., & Muller, J. P. (2012). “Global analysis of the improvements in AATSR nadir-forward co-registration following the application of an automated registration algorithm.” In Geoscience and Remote Sensing Symposium (IGARSS), 2012 IEEE International (pp. 1753-1756). IEEE. Munich, Germany 22-27th July. (ORAL)

Fisher, D., Muller, J. P., & Yershov, V. (2012). “The validation and analysis of novel stereo-derived smoke plume products from AATSR and their application to fire events from the 2008 Russian fire season.” EGU General Assembly. Vienna, Austria 22-27 April. (ORAL)

Muller, J-P., V. Yershov, D. Fisher, (2012). “Stereo retrievals of cloud and smoke winds and heights from EO platforms: past, present and future prospects”. 11th International Winds Workshop. Auckland, New Zealand. 20-24 February.

Fisher, D., J-P Muller, V. Yershov., (2011). “The Development of Novel Stereo Derived Smoke Plume Products for AATSR and their Application to the 2010 Russian Fire Season.” EGU General Assembly. Vienna, Austria 3-8 April. (POSTER)

Fisher, D., J-P Muller, (2010). “Improved Stereo Retrieval of Cloud-Top Heights and Winds from ATSR2 and AATSR: Techniques and Results.” Remote Sensing and Photogrammetry Society Annual Conference. Cork, Ireland 1-3 September. (ORAL)

Fisher, D., J-P. Muller, (2010). “Stereo Motion Vectors from ATSR-2 and AATSR.” ESA Living Planet Symposium. Bergen, Norway 28 June-2 July. (POSTER)

Muller, J-P., D. Walton, D. Fisher (2010). “AMVs from ATSR-2-AATSR and the proposed uncooled thermal IR pushbroom MISRLITE constellation”. 10th International Winds Workshop. Tokyo, Japan. 22-26 February.

Outreach

  • Mapping Science - Demonstrating to new M.Sc. Remote Sensing students methods in mapping science.

Activities

2013

  • ESA Living Planet Symposium, Edinburgh, U.K. (September)

2012

  • ALANIS Final Workshop, Potsdam, Germany (February)
  • RSPSoc Annual Student Meeting (ASM), Falmouth, UK (March)
  • EGU, Vienna, Austria (April)
  • IGARSS, Munich, Germany (July)
  • ISPRS, Melbourne, Australia (August)


2011

  • RSPSoc Annual Student Meeting (ASM), Buxton, UK (March)
  • NCEO Student Conference, Leicester, Swansea (April)
  • EGU, Vienna, Austria (April)


2010

  • RSPSoc Annual Student Meeting (ASM), Plymouth, UK (March)
  • GEWEX Cloud Assessment Meeting, Berlin, Germany (June)
  • ESA Living Planet Symposium, Bergen, Norway (June/July)
  • ESA Earth Observation Summer School, Frascati, Italy (August)
  • NCEO Annual Meeting, Leicester, UK (September)
  • RSPSoc 2010 Annual Conference, Cork, Ireland (September)

One fact about me...

I can identify over 20 varieties of apple... 

Page last modified on 13 jan 14 12:27