Concrete-Crack-Image-Classification

This is a DPHI'S Datathon Sprint Challenge #22: Concrete Crack Image Classification (Rank #7) Link to the challenge: https://dphi.tech/practice/challenge/58#problem

Problem Statement:

Imagine you being a Civil Engineer with knowledge of Data Science and Machine Learning too. You are asked by your state/local government to find out all the cracked concrete and replace it with a new one. Now, with a good knowledge of Machine Learning / Deep Learning, you take a decision to build a system that would alert you when a cracked concrete is detected.

Objective:

Build a Machine Learning or Deep Learning model that would help you detect the cracked concrete.

Evaluation:

Submissions are evaluated using Accuracy Score.

This repository contains 4 files:

1. Training and testing image folders.
2. Solution File (.ipynb)
3. Testing set concrete crack (CSV File).

The Concrete cracks were detected and classified using pre-trained VGG-16 model with hyper-parameter tuning. Further, data augumentation technique is used to boost the accuracy score.

The overall accuracy on the leaderboard comes out to be 99.92%.