2

I am working on a project where I need to detect objects from IP camera on PC using TensorFlow and the data of the detected object to raspberry pi and control its GPIO pins accordingly. I am using python on both ends. I read somewhere to form a client-server network between the two. I need some ideas on how to approach this problem. Thank you. I hope my question was clear.

Extra Information: I am currently using Raspberry pi on PC via ethernet cable. Also, both raspberry pi and my PC are connected to the same Wifi router. I hope this information helps.

vonludi
  • 419
  • 2
  • 20
Prashast Yadav
  • 31
  • 3

2 Answers2

1

You can use sockets to make the Raspberry Pi and your PC communicate. You can use socket.accept(conn, address) to accept a connection and conn.recv(buffer) to receive data.

Wouterr
  • 516
  • 7
  • 16
0

I'm a big fan of using Redis for this type of thing. It has a small footprint, is simple to set up and is lightning fast. It is a "networked data structure server". "Networked" means that its clients can be on the same machine or anywhere on the network. The data structures it serves include:

  • strings
  • integers
  • atomic integers
  • lists
  • queues
  • sets
  • sorted sets
  • hashes

There are bindings for Python, C, C++, PHP, Ruby and rather brilliantly, the shell. That means you can inject values into queues, lists or sets directly from any Terminal on your network whilst your app is running, in order to test it or monitor it.

Example here.

Mark Setchell
  • 191,897
  • 31
  • 273
  • 432