Luisrosario2604 / Speed-Detection

Speed detection of objects (Color and shape recognition)

Speed Detection

The goal of the project is to detect the speed of an object through three steps:

• Color detection (green or blue to black and white)
• Shape detection (via color)
• Detection of the center of the object (via the shape)
• Velocity judgment (v = d / t)

Goals

• Video support (work with photos / video and camera)
• Detection of a specific object (via the 3 steps: color detection, shape and finally the center of the object)
• Object speed detection
• Minimization of errors / cost function (deletion of extravagant data, if any)

Requirements

• Python 3.7+
• opencv_python ~= 4.2.0.34
• numpy ~= 1.16.4
• picamera ~= 1.13

How to install all the requirements :

pip3 install -r requirements.txt

If you have an error installing picamera try :

export READTHEDOCS=True

Usage

Color detection (Step 1)

You can test the color detection with three programs :

• Green detection from webcam
• Green detection from video (car example) -> needed for step 2
• Blue detection from video (bouncy ball example) -> need for step 2

Green detection from webcam

If you want to test the color detection thanks to your camera (works with flashy green) :

Exit with ctrl + C in terminal or "q" key

./cameraGreenDetection.py

Green detection from video (car example)

This will generate a black and white video (outpyColor.avi) from an other one. It will only generate white when there is green on the original video. If you want to test it :

Exit with ctrl + C in terminal or "q" key

./videoGreenDetection.py

Blue detection from video (bouncy ball example)

This will generate a black and white video (outpyColor.avi) from an other one. It will only generate white when there is blue on the original video. If you want to test it :

Exit with ctrl + C in terminal or "q" key

./videoBlueDetection.py

Shape and speed detection (Step 2)

For shape and speed detection you need to do one of the two steps that will generate a video. (This step is based on the generated video) :

• Blue detection (bouncy ball example)
• Green detection (car example)

Exit with ctrl + C in terminal or "q" key

./videoRoundDetection.py

With the car example :

With the bouncy ball example :

Authors

• Luis Rosario - Member 1 - Luisrosario
• Matthieu Daugas - Member 2 - MatthieuDaugas