Our work in

Tech

Tech

Case Study

Automate industrial asset management

Leverage computer vision algorithms for object detection and image classification to extract industrial asset information

Challenge

Our client manages industrial utility assets and regularly captures equipment images. A robust and scalable solution was needed to automatically read and interpret meter values with high confidence.

Solution

We trained deep learning object detection models that automatically extracted the digits from each meter. Since the records are used for customer billing, high accuracy was needed. A second algorithm assessed image quality and prediction confidence to decide whether to accept the model prediction or require manual review. Models were deployed to the customer's cloud using docker containers.

BackBack to tech case studies
View related case studies