We have an exciting new project on Otters from the working group - YaleNUS Otters, in the upcoming DataJam on Sept 11 at Traveloka!
What's not to like?
Would you like to help Assistant Professor Philip Johns from Yale-NUS to sort out a lot of videos and images of Otters?
There will be a Project Accelerator around this.
We would like to look at creating a tool that measures facial/head proportions of otters in order to estimate their age.
We could use something like Zooniverse to annotate pics or vids for training purposes.
We could facilitate uploading information so records can be shared via https://www.inaturalist.org.
That could lay the groundwork for potential facial recognition of otters, which would be very useful.
Whilst we are open to suggestions some work has been done with tensorflow and openCV with the recognition of the Raffles Banded Langur.
Waterpoint Data Exchange
We will also resume our work with Waterpoint Data Exchange (WPDX) on image classification. WPDX aims to automate the water pipe labeling when volunteers upload water point data to their website.
In line with this, we'll be helping them create models which classify the water pump images accordingly.
The following are the associated tasks:
1. From the WPDX dataset, extract relevant water pump images (photo_lnk) to a file
2. Associate the images (photo_lnk) with their labels (water_tech)
3. Use the extracted data to create a model
4. Create a sample app to demo the model to WPDX team.
Sample Image: https://tinyurl.com/water-pump-image
Sample Label: PHE 6 Hand pump
Sample data can be obtained from this link: https://tinyurl.com/wpdx-waterdata
Looking forward to seeing you at the DataJam!
DataKind SG Organizing Team