Notilyze open sources satellite image based refugee camp population estimation model
Together with Elva, Notilyze works on a project to estimate the number of refugees in refugee camps to more accurately estimate the number of necessary supplies in refugee camps. We made a model to extract information from satellite images to help estimating the number of refugees in refugee camps in Nigeria. This project helps to improve the position of refugees. Therefore we open sourced the code of Paul for this model through Github.
During his thesis, our colleague Paul combined three existing Github projects to come to a complete project that includes:
- image preprocessing
- training of a Faster R-CNN model (object detection model, read more)
- evaluation of the results using a mean Average Precision score.
Check it out: https://github.com/Notilyze/Faster-R-CNN-Object-Detection
