Structural Damage Detector

General purpose
Structural Damage Detector

Overview

Real world damage detection from videos/photographs of Infrastructure undergoing construction or structural strength survey or maintenance survey. Manual surveys are prone to corruption & human error which has resulted in huge losses. With much accurate & faster eyes of computer & advancements in devices help us get more detailed data & in a result generate better predictions with better accuracy.

Technology

  • Deep Learning
  • Neural Networks
  • Computer Vision
  • Image Processing

Language

  • Python

Key Technical Challenges:

  • Working with all types of damages like glass breakage, doors slammed, dents, scratches to name a few.
  • Creating clustered environment for training due to huge processing time taken by image kind of data
  • Real time inference in mobile as well as web application.
  • Should work with any kind of background.

Business + Technical Points:

  • Quick response from trained model
  • Scalable & adaptable model for heavy vehicles
  • Easy training pipeline & responsive dashboard.

Result:

  • We were able to identify & meet requirements accurately with 44ms inference time.