Paper Title
Real Time Inspection of Concrete Structures using Computer Vision and Deep Learning

Abstract
In this paper, a novel method for crack detection is proposed which involves training a classifier with raw images of concrete surfaces with a broad range of image variations. The images are taken from a complex engineering building using a decent camera. Once the cracks are captured via a camera, several pre-processing techniques are applied for highlighting the cracks. The prepared training image set is fed into an object detection model for separating cracked from intact concrete images in the validation set. Once the crack has been detected, they can be classified into different types based on their severity, structural impact, and size. Such data can form the basis of a decision support system which can be used for the structural analysis of a building. Keywords - Concrete Cracks, Structural Analysis, Deep Learning, Computer Vision, Image Pre-processing