Satellite images are a reminder about our planet and the science of Earth observation. It’s not just about appreciating the beauty but also about finding meaningful insights from them. Let’s learn about some machine learning techniques and do some Land Cover Classification with help of Python!
Have you ever worked with geospatial data with python? Mckinsey estimates that economic value of big geospatial data could reach $700 billion/yr by 2020. Working with Satellite images is absolutely fantastic. It is a reminder about our planet and the science of Earth observation. It’s not just about appreciating the beauty in these images but also about finding meaningful insights from them. Are you excited about learning new stuff? This talk aims for beginners who are starting with Machine Learning and are ready to learn. We will have some live coding and do some Land Cover Classification.
In this workshop we will learn and use the following libraries and tools :
1)Scikit Learn 2)GDAL 3)Numpy 4)Jupyter 5)Pandas 6)Matplotlib.
In this workshop, I will explain about these items, how it’s useful and why we should use it. The audience will learn about basics of Satellite images, different types of classifications in satellite images, Land Cover classification and more using a few Machine Learning Algorithms in Python.
In this talk, we will learn and use Scikit Learn, GDAL, Numpy, Jupyter, Pandas, and Matplotlib to classify satellite images. This summer I worked at GIC, AIT Thailand (who work with JAXA, United Nations, and World Bank) under the guidance of a Ph.D. from University of Tokyo. With this workshop, I want to spread the knowledge about the things I learned and hope that this can be useful for them as well.
I don’t have a background in remote sensing or even Geography. I am an Engineer. So as a self-learner I will have a better understanding of the problems that people are having than a person with such a degree because while learning I might have faced the same problem whereas someone from this field might take such nuances for granted.