FaceRecoginition-Web-App: Project Overview

• To classify the gender by uploading a image and model will predict the gender whether Male or Female
• Object detection using Haar Casacade
• For the preprocess images, we will extract features from the images, ie. computing Eigen images using principal component analysis
• Developed web server gateway interphase in flask by rendering HTML CSS and bootstrap in the frontend and in the backend written in Python
• Optimized Linear SVM algorithm using GridsearchCV to reach the best model
• Integrating the machine learning model to Flask App.

Code and Resources Used

Python Version: 3.7
Packages: pandas, numpy, sklearn, matplotlib,scipy,opencv

Live Working project

Face Recognition Web App

Dataset

Get the data from : https://data.vision.ee.ethz.ch/cvl/rrothe/imdb-wiki/ Data files :

• Divided into two folders male and female
• Both folder contains 7,000 images of male and female

Data Cleaning

After understanding business requirements, I needed to clean it up so that it was usable for our model.

• Handle the unstructured data
• First convert into grayscale image
• Then crop the faces
• Preprocess the data such as normalize and resize image(100x100)
• Flatten the Image
• Implement feature selection
• Splitting the dataset into train and test

Machine model application flow

Upload Image -> Crop image -> Data prerocessing -> Feature extraction -> ML model -> Output

Model Building

I split the random data into train and tests sets with a test size of 30% and applied pca Implement Linear support vector machine and used hyperparameter tuning GridsearchCv and ROC AUC Score - 0.91

Model performance

• Linear SVM : accuracy = 0.8186

Productionization

In this step, I built a flask API endpoint that was hosted on a local webserver