Skip to content

Latest commit

 

History

History
34 lines (24 loc) · 1.08 KB

README.md

File metadata and controls

34 lines (24 loc) · 1.08 KB

icn_demo

The repository contains an edge computing demo for ICN'18

OCR

We implement and evaluate a simple Optical Character Recognition (OCR) algorithm to illustrate the execution of pure programs. Our simple algorithm is initialized in the enclave with a model of each letter of the alphabet; it can then be fed with input images to detect the list of embedded letters. We implemented this example to show how our system can be used to outsource execution of pure programs.

Requirements

The system requires libsfml. On Ubuntu type: sudo apt-get install libsfml-dev

Compilation

  cd ocr
  make
  ./app ./data/input_5_OK.png

Dnsmasq config

Just add this line to the file /etc/dnsmasq.conf with the name of the script that automatically creates a face towards the Android client in the AP.

dhcp-script=/home/pi/icn_demo/dhcp_log.sh

Android app

The NDN OCR Android app for the demo can be found in the following repository:

https://github.com/srene/icn_demo_Android

but also in Google Play:

https://play.google.com/store/apps/details?id=uk.ac.ucl.ndnocr