Recently, I've been working on a faceπ, eyeπ and bodyπΆ detection it is finally working!πΊ The source codes basically uses a image, video or the camera as user input and displays the type of detection desired by the user.
π¨βπ» The basic layout of detection through images is to create a classifier, open and fix the image colors and then using .detectMultiScale function is cascade classifiers to detect the face, eye or full body. π½ As far the as the input video is concerned, a while loop is preferred which captures the frames from the video and delivers the result on it. Image processing still requires some adjustments
πI'm very new to all this and so, I took a project offered by Ilias Papachristos (Data Analyst - Scientist). I'm sharing the the source code on my GitHub. Feel free to alter and hit me up!!!
π¨βπ« Also, I'd like to clearly state it: This is not something new, it has already been implemented years ago and even open sourced. I just stitched the pieces of several codes together and altered it to what I wanted to detect.
This section lists all the technologies that I used to built this project.
- Clone the repo
$ git clone https://github.com/yuvrajverma01/OpenCV-DetectionProject.git
- Install Numpy
$ pip install numpy
- Install Matplotlib
$ pip install matplotlib
- Install OpenCV
$ pip install opencv-python
- Install Jupyter Notebook
$ pip install notebook
- Initialise Jupyter Notebook
$ jupyter-notebook
- Run
Codeinjupyter
file
The file structure of the current project is structured as shown below:
OpenCV-DetectionProject
βββ Images
βββ Video
βββ Code.py
βββ Codeinjupyter.ipynb
βββ cascade_eye.xml
βββ cascade_frontalface_default.xml
βββ cascade_fullbody.xml
βββ haarcascade_car.xml
Made with β€ by Yuvraj Verma.