Vision-based sintering parameters reading system

• Developed with Raspbian, OpenCV, scikit-learn, etc.
• Allow temperature reading & electric current reading.
• The SVM classifier trained by 12000 seven-segment digit images is available.

Overview

Installation

       

User flow

       

Pre-settings

Set up parameters for temperature reading & electric current reading.

Temperature reading

• Localize the meter & obtain meter image by a perspective cropping.

       

• Determine the size, quantity, and location of interested digits.

       

Electric current reading

• Localize the meter & obtain meter image by a perspective cropping.

       

• Localize the scales & determine the amperage between scales.

       

Vision-based sintering parameters reading

Temperature reading

• Read temperature value of the sintering furnace.

       

Electric current reading

• Read current value of the sintering furnace.

       

Demo video

• Demo影片(中文)

Usage of classifier

• Step1: Use following command to install required packages

    pip install scikit-learn==0.22 scikit-image==0.14.2 numpy==1.17.4
  

• Step2: Run the script with image path to get recognition result on your screen

    python recognize_single_digit.py digit5.jpg
  

Contact

• Contact me at jerming0515@gmail.com for complete source code.