Project

Cam 1001 - low power modification

Short description

Digits Recognition is a low power device for digitizing various analogue instruments and sensors.

Digits Recognition Low Power Device

This product, based on the SONY Spresense Boards, operates with several types of analogue devices: mechanical counters, digital displays and pointer meters. There are two options: to go from spot to spot taking the data from the sensors or just let the camera capture all the data automatically. This allows automating the process of keeping the clients updated whether a situation is fine or whether it needs personal intervention sending alert.

Device features:

  • Low power consumption.
  • Transmitting data to the client’s server.
  • Reading data from digital displays.
  • Reading data from pointer meters with simple dials in the background.
  • Sending SMS with warnings when threshold values ​​are exceeded on devices.
  • Sending data to the server.

Technologies used

• SONY Spresense mainboard
• SONY Spresense extension board
• SONY Spresense camera board
• LTE modem based on SIM7600E-H produced by Waveshare Electronics

• Spresense Arduino Libraries
• Arduino JPEGDecoder library (https://github.com/Bodmer/JPEGDecoder)
• Sony Neural Network Console (https://dl.sony.com/)
• Neural networks based on the modified AlexNet architecture
• Our own dataset of seven-segment digits
• Dataset SVHN (Street View House Numbers) with CC0: Public Domain (https://www.kaggle.com/stanfordu/street-view-house-numbers)
• Classic image processing methods
• Our own function for reading pointer’s angle
• LTE modem control by AT-commands

Hire Us To Get Started!

Our team is waiting for you to discuss your vision for the perfect application.Help us know more about your ideas!