Cloud Foundry Summit Silicon Valley 2017

Artificial Intelligence (AI) can be the difference between applications that just perform tasks and applications that seem truly intelligent and intuitive to humans. With cloud platforms like IBM Bluemix and Amazon Web Services, AI is as accessible as ever to developers through provided and hosted APIs. These cloud platforms also allow simplify the process of using AI with other powerful cloud services such as IoT, big data, messaging, and serverless functions.

Hardware systems are unique in that they are able to collect real-time data from their environments with various sensors. These systems can become even more powerful and perceptive when this data can be rapidly analyzed with machine learning models. A basic detection system will send an alert when it detects any kind of motion near the sensor, but an intelligent system “knows” what the detected object is. Using image recognition, it can classify the moving object as human and even make a trained guess at what emotion that human was feeling at the moment it was detected.

In this talk, Linda Nichols will describe how a real-life use-case (pets jumping on the kitchen counters) evolved from a simple motion detector to an intelligent and inexpensive cat detection system. Using IBM Watson IoT Platform and Node.js, she was able to connect a Raspberry Pi with IBM Bluemix OpenWhisk functions and IBM Watson image recognition APIs. She will demonstrate how her system can quickly and reliably notify owners when it detects a cat, but ignore other categories of moving objects. She will also discuss how she was able to reduce the complexity of the system architecture and the amount of static code stored on the Raspberry Pi by using “serverless” IBM Bluemix OpenWhisk functions.