LiLz uses computer vision to read gauges and dials where humans prefer not to tread
- BySaksham Daga | January 5, 2022
I don’t want to be a maintenance worker who has to walk 0.5 miles in a damp corridor just to check the pressure gauge on a valve somewhere. LiLz allows you to remotely monitor such impractical physical interfaces by applying machine learning in a clever and practical way.
Japanese (especially Okinawan) startups have been around for some time-TCJP colleagues also wrote them down. However, while the value of the service seems obvious at first glance, it has not yet achieved a breakthrough. LiLz has joined CES as part of the national trade association, along with many other companies listed here. The LiLz device looks like a chubby tablet without a screen. It basically has a large battery packaged in the camera, lights, processing and communication chips, and is fully powered for up to 3 years.
Mount the device so that you can see the associated measurement device or dial. After reviewing the photo and signal, configure the app to interpret what it points to. You can read circular, semi-circular, linear displays, digital and scrolling or analog numeric displays, or colored warning lights. (The ML included here is not easy-I came across this interesting article while investigating it.)
Send measurements live or at intervals to the central dashboard as soon as they are set up. An API that can be queried or recorded elsewhere. Data is sent via LTE or Bluetooth.
Follow Startup Story
Related Posts
