Creating a Smart and Serverless Cat Detection System with IoT and Image Recognition APIs
summary
A battery‑powered Raspberry Pi with PIR camera detects motion, uploads a photo to IBM Bluemix IoT Platform, and triggers lightweight IBM Bluemix OpenWhisk functions that call Watson Visual Recognition and Twilio to SMS only when Watson labels include “cat.” Node.js plus johnny-five keeps on‑device code minimal. Demo reliably flags cats (examples show 71–83% confidence and the system treats any cat label in the top 10 as a hit), scales to many devices, and reduces Raspberry Pi complexity with serverless handlers.
AI-written summary · about
chapters (12)
- 0:00Introduction and speaker intro
- 1:26Why build a cat detector
- 3:14Hardware and Raspberry Pi setup
- 6:04IoT platform and MQTT basics
- 7:11Serverless with OpenWhisk
- 9:00Scalability and cost benefits
- 10:25OpenWhisk function example
- 11:09Watson Vision Recognition API
- 13:20Code structure overview
- 14:04Live demo setup
- 22:35Demo results and limitations
- 23:01Q&A session
related talks
26mFrom Software to Hardware: How Do I Track My Cat with JavaScript
39mHijacking IoT with IBM Watson
44mChallenging the Internet of Things
37mUnstructured Data Processing with a Raspberry Pi AI Kit and Python
09mConnect and Control an IOT Device in 5 Minutes!
01h 00mRun your own IoT device in a snap
18mAbout Life, Robots and Cats!
18mAn Introduction to IoT (Internet of Toilets); Or How I Built an IoT Kitty Litter Box Using JS