Security Analytics x Hadoop
Group F
Fergus Buckley / Laurens ten Cate / Cedric Viret / Karim el Shamma / Ryan Hayashino / Javier Contreras / Richard Rohla
Satellite Imagery Labeling
Storing and Accessing
Problem Statement
Knowledge Extraction
Consolidated Database
Decision Makers
FBI
CIA
NSA
Variety of formats
(JPG + EXIF/ TIF / HTML)
Data Sources
Processed through HIPI
Governmental Agencies
Data Ingestion
Hadoop Image Processing Interface (HIPI)
MapReduce
OpenCV - Computer Vision
Data Processing
Hipi Image Bundle (hib)
MapReduce
Hipi Image
Bundle
Cull
Images
1...k
...
Images
n-k...n
Map 1
...
Map i
Shuffle
Reduce 1
...
Reduce j
Result
Data Life Cycle
Data Access issues
Data Preservation
Data Architecture
Cull
Hipi Image Bundle
MapReduce
Consolidate
Extract
Analyse
JPEG + EXIF
TIF
