We're sorry but this page doesn't work properly without JavaScript enabled. Please enable it to continue.
Feedback

IOTA2: large scale land cover mapping operational chain

00:00

Formal Metadata

Title
IOTA2: large scale land cover mapping operational chain
Title of Series
Number of Parts
351
Author
Contributors
License
CC Attribution 3.0 Unported:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
Identifiers
Publisher
Release Date
Language
Production Year2022

Content Metadata

Subject Area
Genre
Abstract
The use of remote sensing data operating in different observation domains is an undeniable asset for the realization of quality land cover products. Indeed, satellites allow to cover large areas of interest in a regular way with a durable quality. Satellite data can be of different but often complementary natures, which makes it possible to broaden the possible fields of application (water management, snow cover, crop yield, urbanization, etc.). In addition to these new data, there are recent technological developments (or old but now usable due to the evolution of computing capacities, such as the use of neural networks), and means of service provision and dissemination that allow these applications to be carried out over a longer period of time (long time series that are computed more rapidly) and in a larger space at different scales, sometimes simultaneously (stationary, local, national, continental, global scale). iota2, developed by CESBIO and CNES with the support of CS GROUP, is a response to the growing demand for the creation of an Open Source tool, allowing the production of land cover maps at a national scale that is sufficiently generic to be adapted to the different objectives of users. In addition, this project ensures the production of an annual land use map of metropolitan France [REF doi.org/10.3390/rs9010095], with a satisfactory level of quality, thus proving its operational capacities. iota2 integrates several families of supervised algorithms used for the production of land use maps. Supervised algorithms (e.g., Random Forests or Support Vector Machine) that process pixels that can be parameterised by the users through a simple configuration file. iota2 also offers the user the option of using a deep learning model. In addition to the pixel approaches, contextual approaches are also proposed, with Autocontext [1] and OBIA (Object Based Image Analysis). Autocontext, based on RF, takes into account the context of a pixel in a window around its position. The OBIA approach exploits an input segmentation to classify objects directly. In addition to the supervised classification approaches, iota2 is also able to produce indicator maps (biophysical variables) either by supervised regression or by using user-provided processors, diversifying the possibilities of using iota2. One major interest in iota2 is it's ablility to deal with a huge amount a data, for instance the OSO product (theia.cnes.fr/atdistrib/rocket/#/collections/OSO/2327b748-a82c-5933-afb0-087bbfeff4cd) is generated using a stack of all available Sentinel-2 data over the France without any landscape discontinuity due to the Sentinel-2 grid. Another point of interest is its capability to produce a landcover map everywhere a Sentinel-2 data and a groundtruth are available (ie : agritrop.cirad.fr/597991/1/Rapport_Intercomparaison_iota2Moringa.pdf). 1. Derksen, D., Inglada, J., & Michel, J. (2020). Geometry aware evaluation of handcrafted superpixel-based features and convolutional neural networks for land cover mapping using satellite imagery. Remote Sensing, 12(3), 513.
Keywords
202
Thumbnail
1:16:05
226
242
Scale (map)Local GroupTexture mappingGroup actionSoftware engineeringPresentation of a groupGoodness of fitComputer animation
Local GroupPresentation of a groupStack (abstract data type)Open sourceProduct (business)Series (mathematics)OpticsAreaProcess (computing)SupercomputerLocal ringMultiplicationSocial classIntegrated development environmentData managementInterpolationAxonometric projectionUnified threat managementPoint cloudPixelContext awarenessComputer-generated imageryObject (grammar)Mathematical analysisSupport vector machineoutputRandom numberForestCovering spaceAreaMultiplicationOpticsOpen sourceNeuroinformatikAxiom of choiceMehrprozessorsystemVirtual machineInterpolationoutputPoint cloudProduct (business)Time seriesScaling (geometry)Social classBilderkennungForestMedical imagingProcess (computing)Software frameworkElectric generatorClosed setGoodness of fitReference dataWindowTesselationAddress spaceThread (computing)Projective planeCompilation albumCuboidTurm von HanoiDifferent (Kate Ryan album)Context awarenessObject (grammar)Support vector machineSoftware development kitComputer animation
Transcript: English(auto-generated)
Good afternoon, everyone. I am Julien Osman. I work as a software engineer at CS group company in France. Artur Vincent was not able to make it today, so he asked me to do the presentation for him. And I will present YoTattoo.
If you want to generate land cover maps on the large scale using machine learning, YoTattoo is a tool you need. It can take as input some optical images or SAR images.
It works with time series on very large areas and it's able to work with different geological areas. It's it relies on the Orpho toolbox,
multiprocessing and multithreading capabilities, and it can be distributed on multiple computing nodes through Dask. It's already used on production to produce the Ozo project, which is a land cover map of France
that's produced every year. It contains 23 classes. And also, it's an open source project, so you can check it out at this address. What do I mean when I say large scale? Well, first it works with
heterogeneous data, so you can work with multiple kind of sensors on multiple dates, and it will deal with the presence of clouds. It's able to work with multiple geological areas, as I said before, so for example in France,
you don't have the same climate close to the earth land or in the middle of France, so you can specify it as an input, and it will take it into account. Also, it will do some temporal interpolation to smooth the data, remove the clouds, and
also to homogenize the dates since you can put images of any date. So the temporal interpolation will deal with different dates. As I said, it's a framework for machine learning, so you will want to do some classification with it, and
you have multiple choice of classifier. I said earlier that it relies on the Orpho toolbox, so all the classifier in the Orpho toolbox are available, like the random forest or the SVM, but you can also use a classifier from the
Cyk-Hitteram or PyTorch if you want to do some deep learning. You can also do some object-based image analysis using some segmentation as input, and it also has a capability of auto-context to work with
SuperPixels and do some recurrent processing. So to conclude, Yota 2 will help you to do some long cover maps at large scale, so don't use it if you just have one tile. I mean, it's really large scale over countries, and
you will need to give as input some good quality reference data for your learning step. It doesn't work with Windows, only with Linux, but it's easy to install because we provide
a Conda package. You will find the documentation at this address, and you are welcome to contribute. Thank you.