Skip to content


svm satellite

Support Vector Machines – on recognizing pixel clusters in satellite data

  • blog

Scikit-learn, the machine learning library built for Python over 10 years ago is an excellent resource for estimating data and can integrate into geospatial workflows. Helpfully, when choosing an estimator, scikit-learn supplies an interactive diagram to choose the best estimator for the job. One of the big things about machine learning is the need to… Read More »Support Vector Machines – on recognizing pixel clusters in satellite data

Space logo 2017

UK Space Conference 2017

By 2030 the UK would like to have a £40 billion or 10% slice of the global space economy. In 2015 the market was estimated at £13.7 billion. Ambitious targets? At the UK Space Conference this week I heard a speaker who noted that the Space economy was second in size only to Oil and… Read More »UK Space Conference 2017

Space is open

  • blog

This week I attended an excellent talk given by Dr Jon Blower, CTO at IEA in Reading, entitled “Satellites, sensors and statistics – from environmental data to decisions” which was a meetup of the Society of Data Miners at the Royal Statistical Society in London. It is always good to attend events like this when… Read More »Space is open

Earth Observation: Big Data

  • blog

What is Big Data? Big data is a term for data sets that are so large or complex that traditional data processing application softwares are inadequate to deal with them. In an event this week in central London I heard big data described as “non-spreadsheet” data. I think Earth Observation (EO) data could fit… Read More »Earth Observation: Big Data

Extracting values from satellite imagery

  • blog

Last week I wrote about using Earth Observation data without ever looking at a satellite image; extracting the values from an image and then presenting the data in an informative way. Just how easy is it to extract the values from images? Sentinel 2a is operating with 12 bands; that means every location that… Read More »Extracting values from satellite imagery

Using Earth Observation data without ever looking at a satellite image

  • blog

To enable a user to use space derived data without ever actually seeing a satellite image. To move from pixels to analytics. To break through the ‘that’s a nice image – but so what?’ barrier? To add value. I wrote about my thoughts on this last week; the need to move towards more than just… Read More »Using Earth Observation data without ever looking at a satellite image

Keep up to date

  • blog

How to consume news/information? If you are not on the first page of a Google search you are effectively nowhere; perhaps if you are not in the first three search results you don’t stand much chance of getting viewed. This presents a problem of trust; we have to trust that the Google search algorithms are… Read More »Keep up to date

Machine learning Landsat / Sentinel data

  • blog

What changes can be measured using Landsat and/or Sentinel-2 data? In large areas change detection (land use for example) is commonly used for these data sets. If companies like Orbital Insights are counting cars, using shadows from floating oil tanks to determine capacity and measuring levels of construction, what smaller objects and data analytics can be… Read More »Machine learning Landsat / Sentinel data