Eric Fegraus & Jorge Ahumada - Hp Big Data 2015 - theCUBE
summary
Conservation International describes a replicable analytics pipeline that processes millions of camera-trap images into species counts using a hybrid of machine learning and human review, a cloud/on‑prem deployable product that scales to national networks. Satellite processing moves from R to Distributed R on Vertica yielding a 10-fold speedup. Field use in Uganda detects declining cat populations and drives park management actions; a public repository called Wildlife Insights aggregates shared camera‑trap data.
AI-written summary · about
chapters (12)
- 0:00Introduction to HB Big Data 2015
- 0:43About Conservation International
- 2:09Analyzing camera trap images
- 3:12Technical challenges in data collection
- 5:39Distributed R for satellite images
- 7:47Big data for social good
- 9:54Impact on public policy
- 13:27Technology and funding barriers
- 15:34Wildlife Insights initiative
- 16:56Social media and community growth
- 17:39Goals for next 12-18 months
- 18:21Closing remarks
