Reduce Manual

Effort

The automated approach reduces the manual effort required in visually inspecting and holding the screen at different angles to capture the right light in order to manually visualize the cracks in the screens and to then manually classify the same according to a gradation scale published by the client. It also eliminated manual errors and provided for a far more robust data capture mechanism for the client.

Card image cap
Card image cap

Cross-device

Damage Detection

The model has been trained to detect and grade screen damage across various mobile phone models and screen types, thereby providing efficiencies of scale and higher throughput from the assessment stations. The model also provides an automatic assessment of the extent of the screen damage as a function of the screen size and phone type.

Phone Make &

Model Detection

As the model was trained on multiple different brands of smart phones, as an adjunct information, the algorithm also identifies the Make and Model of the phone from its image; this helps in categorizing the device for further workflows and also in assessing the repair costs for the damage detected.

Card image cap

Want To Develop Something Like This ?

Let`s Talk

Convolutional Neural Networks

A bespoke machine learning model based on Convolutional Neural Networks was deployed using TensorFlow and trained on 8000+ training images across 200+ cracked and normal devices.

Efficiencies Delivered

The deployed model was able to automatically process more than 80% of the incoming inventory, resulting in a 90% reduction in processing time.

Deployment

The resultant TensorFlow model was deployed at three different warehouses for the customers for processing mobile phones; training is underway to detect cracked tablet screens next.

Get In Touch