Jan 20A brief introduction to satellite image segmentation with neural networksGet up to speed with this versatile technique for labelling pixel regions in an image Image segmentation is a very versatile technique which enables the identification of targets which range in size from a small cluster of pixels up to targets that fill an entire image. This post introduces segmentation and clarifies the different types of segmentation (semantic, instance and panoptic). It then discusses image…Deep Learning8 min readDeep Learning8 min read
Jan 20A brief introduction to satellite image classification with neural networksWhat is image classification, which models are used and how to approach your first project You may already be familiar with image classification from seeing the numerous cats vs dogs image classification tutorials on the internet. Image classification is therefore the task of assigning one (or more) labels to an entire image. Note however that the term ‘classification’ can mean different things to different people…Deep Learning6 min readDeep Learning6 min read
Jan 20Transformers in remote sensingIntroduction to Transformers and why you should care The Transformer is a neural network architecture first published in a 2017 paper titled Attention Is All You Need by researchers at Google. Transformers learn context by tracking relationships in sequential data using a technique called attention. Transformers are replacing CNN’s and RNN’s in many applications and have come to…Deep Learning5 min readDeep Learning5 min read
Jan 20Quick tour of the WorldStrat DatasetThis significant new dataset raises the bar for documentation quality A new dataset called the WorldStrat dataset was recently announced on Twitter by Julien Cornebise, Ivan Orsolic, and Freddie Kalaitzis. The dataset raises the bar in documentation quality of a dataset and the release includes a paper on ArXiv, code on Github, and the dataset itself on both Zenodo and…Deep Learning6 min readDeep Learning6 min read
Jan 20Understanding the success and significance of the satellite-image-deep-learning repositoryWhy has this repository been a success, why is it significant, and top 10 best practices for making your Github repositories stand out I maintain a popular repository on Github called the satellite-image-deep-learning, which has passed the 5k stars threshold, making it the first repository in the remote-sensing category to do so. In this post I discuss what I think makes this particular repository so successful, and ponder the significance of this success…Github8 min readGithub8 min read
Jan 20SOS! Sustainable open source softwareWhat are the strategies to make open source software projects sustainable? There is heightened awareness about the topic of sustainability: for example people are increasingly aware that burning fossil fuels and releasing carbon into the atmosphere is unsustainable and is creating big challenges for society. In the digital world, open source software is increasingly critical to the operations of many businesses…Github6 min readGithub6 min read
Jan 20A brief history of the satellite-image-deep-learning Github repositoryA brief history of my popular Github repository and why you should create something similar I maintain the popular Github repository satellite-image-deep-learning. This post provides the brief history of this repository, how I find material, and why you should create something similar. History When and why did I start this repository? I started this repository in April 2018 whilst I was working at Surrey Satellites. The…Satellite Imagery4 min readSatellite Imagery4 min read
