Open Access Open Access  Restricted Access Subscription or Fee Access

Image Processing based Wall Crack Detection

Ranpise Mahesh, Pawar Abhijit, Raskar Akshay

Abstract


Concrete spalls and crack detection of structures may be a labour intensive and daily task. But, it plays a crucial role in health monitoring of civil infrastructures and buildings. Automated inspection with automated models has been considered one among the simplest ways to eliminate both error and price . This paper presents an automated approach using Camera and towards a Concrete Structure Spalling and Crack database (CSSC)[A], which is far and away the primary released database for deep learning inspection. We aim locate the spalling and crack regions to help 3D registration and visualization. For deep inspection, we offer an entire procedure of knowledge searching, labeling, training, and post processing. We further present a visible Simultaneously Localization and Mapping(SLAM) approach for localization and reconstruction. From comparative experiments and field tests, it is possible to achieve upto 70% accuracy using CSSC database.

Keywords: image processing, crack detection, ccny, machine learning, technology

Cite this Article: Ranpise Mahesh, Pawar Abhijit, Raskar Akshay, Mahi K. Image Processing based Wall Crack Detection. Journal of Image Processing & Pattern Recognition Progress. 2020; 7(2): 27–32p.

Full Text:

PDF


DOI: http://dx.doi.org/10.37591%2Fjoipprp.v7i2.2527

Refbacks

  • There are currently no refbacks.