Home
Watson Vehicle Damage Analyzer
Create a custom classifier for visual recognition using images of automobile damage.
Cognitive
We'll create a Visual Recognition custom classifier to identify issues with automobiles, such as a flat tire, broken window, or vandalism, and a server that will use this classifier given an uploaded image. We'll also create a mobile app that takes a picture of an automobile problem to sent to a server for classification.
By Scott D'Angelo
https://github.com/IBM/watson-vehicle-damage-analyzer
N/A
In this developer code pattern, we will create a mobile app using Apache Cordova, Node.js and Watson Visual Recognition. This mobile app sends pictures of auto and motorcycle accidents and issues to be analyzed by a server app, using Watson Visual Recognition. The server application will use pictures of auto accidents and other incidents to train Watson Visual Recognition to identify various classes of issues, i.e. vandalism, broken windshield, motorcycle accidnet, or flat tire. A developer can leverage this to create their own custom Visual Recognition classifiers for their use cases.
When the reader has completed this Code Pattern, they will understand how to:
Create a Node.js server that can utilize the Watson Visual Recognition service for classifying images. Have a server initialize a Visual Recognition custom classifier at startup. Create a Visual Recognition custom classifier in an application. Create an Android mobile application that can send pictures to a server app for classification using Visual Recognition.
- User interacts with the mobile app and captures an image.
- The image on the mobile phone is passed to the server application running in the cloud.
- The server sends the image to Watson Visual Recognition Service for analysis.
- Visual Recognition service classifies the image and returns the information to the server.
- Watson Visual Recognition: Visual Recognition understands the contents of images - visual concepts tag the image, find human faces, approximate age and gender, and find similar images in a collection.
- Mobile: Systems of engagement are increasingly using mobile technology as the platform for delivery.
- Node.js: An asynchronous event driven JavaScript runtime, designed to build scalable applications
Classify images sent from your phone app on Watson's Visual Recognition.
Scott D'Angelo
The implementation of Machine Learing is continuing to transform the world of artificial intelligence. One area where we've seen great impact is Visual Recognition. By using a supervised learning method, we can train a Visual Recognition classifer to identify classes of images by providing positive and negative examples, and letting the machine learning algorithms create the classifiers. The IBM Cloud offers a variety of cognitive tools for implementing Artificial Intelligence solutions. One of these is Visual Recognition, which offers the ability to create custom classifiers by uploading sample images. We can make use of this to implement an interesting use case for the insurance industry: creating a custom classifier for analyzing vehicle damage. Imagine your auto insurance company wants a mobile phone app to distribute to it's customers. The app could take a picture of some Vehicle damage, send the info to the insurance company, and recieve a classification of the damage. For a production app, more logic could be added to provide the customer with info about deductables, coverage, and next steps to take. In our Code Pattern for the Watson Vehicle Damage Analyzer we demonstrate the simple case of build a custom classifier, deploying a server to recive uploaded images and send back classification info, and building an Android phone app to take pictures and send to the server. Check it out at https://developer.ibm.com/code/patterns .