Creating image recognition software to find cracks on smartphones
A specialist insurance company was looking to build a solution which would enable the validation of a mobile phone’s condition, regardless of time and location. To answer this need, they built a smart image recognition tool that allows users to check their device’s state at any given moment.
Scroll down for more
With advanced testing methods, the company can regularly and thoroughly test their products - especially when updating - saving money and adapting their products faster.
Results by Numbers
Validation time from photo to policy
(Actual result from test set - 89.623%)
Image Analysis steps
The entire imaging process is executed in seconds
As an insurance organisation, the company specialises in offering protection for mobile devices, working with network providers to offer insurance for new smartphones. They already had existing hardware, but they wanted a mobile app to expand their offers to protect older phones.
In both cases, the organisation identified a need for powerful software that could quickly and accurately detect such faults. Because photographs were the only source of information available, this software would need advanced and sophisticated image recognition capabilities.
Visual tracking and positioning systems, actor-model processing, image processing
.Net, Akka.Net micro services, Python, C++, OpenCV, Odroid, iOS, Android
C# 7.3, VS 2017, EF 6.2, C++ 17, Git, BitBucket, Jira, .NET 4.7, .NET Core, OWIN, Kestrel
6 x Backend Developers, 2 x Testers, 1 x Business Analyst, 3 x Mobile Developers, 3 x Frontend Developers
About the Client
The company provides specialist insurance for mobile devices. Their services are available on over 3 continents and they partner with a range of network providers and operators.
Need a software solution?
Don’t hesitate to get in touch via the form below or at:
+48 71 798 2692
Let us know what you’re looking for.