workzone

REST api setup for industrial robot work zone estimation and control. Include face detection/recognition module, object detection module and zone estimation module.

Features

Setup

Install from source

Clone repo and recursively update submodules

Download weights

• Download weight from Google Drive
• Move weights to the 'weights' folder inside parent folder

Setup database with docker

MongoDB used to store face descriptors for face recognition subsystem

Download MongoDB docker image

$ docker run --name mongo -d mongo:tag

And then start it

$ docker run -d -p 27017:27017 mongo

Start server

To start server use Flask CLI

$ flask run

or use python

$ python run.py

Setup robot workzone estimation subsystem

First update submodules

$ git submodule update --init --recursive

Move to robot_work_zone_estimation directory

$ cd robot_work_zone_estimation

Follow instructions from submodule README.md

Setup face recognition subsystem

• Provide faces for face recognition subsystem

Use the following directory structure:

├── workzone                        # Parent directory
    ├── face_database               # Directory with all persons
    │   ├── person_name_1           # Each person should have it's own folder
    │   │    ├── img1.jpg       
    │   │    ├── img2.jpg
    │   │    ├── img3.jpg
    │   │    └── ...
    │   ├── person_name_2           
    │   │    ├── img1.jpg
    │   │    ├── img2.jpg
    │   │    ├── img3.jpg
    │   │    └── ...
    │   └── ...
    └── ...

• Start server
• Route to https://127.0.0.1:5000/fill_db
• Done

Available routes

Scenarios

Start scenario

Add new scenarios

License