Adding Image Recognition to the App

improve this page | report issue

Watson Visual Recognition

Image recognition capability is powered by Watson Visual recognition service on IBM Cloud. Create a Watson Visual Recognition instance on IBM Cloud. For more information, see here.

Once configured, you can now create a new Mobel and add classes to it. You can drag and drop images in to the Builder and then train your Model on those images. Once the training is complete, you can either download the CoreML model or use the Model in a AI control in your app.

To enable a Visual Recognition in your app, perform the following steps:

  1. Click Watson and then click Image Recognition. This displays the Work with Watson Visual Recognition screen.

    Watson Visual Recognition

  2. Click Connect to your Watson Visual Recognition instance.

    Watson Visual Recognition Instance

  3. Enter the API key details and specify the URL of your Watson Visual Recognition instance.
  4. Provide a Name to your Image Recognition instance in the app and click Connect. This displays the dashboard for your model.

    Watson VR new model

  5. Click Add new model to create a new model. This will display Create a new Model popup.

    Watson VR model name

  6. Enter the Model name and click Create. This will display the classes for that model and a Negative class.

    Watson VR Model Class

  7. Click Add new class. This will display a popup to specify a name for the new class.

    Watson VR Model Class Name

  8. Enter the Class name for the new class and click Create. This will display the workspace to add your images for training the model.

    Watson VR Model Class training

  9. Add the images to the model either by drag and drop them into the workspace or use Browse to access the images.

  10. You can go back to your workspace after adding the images and test by clicking Test Model.

    Watson VR Model Class testing

  11. In the Try your model section, add an image and then result is displayed.
Inclusive terminology note: The Mobile First Platform team is making changes to support the IBM® initiative to replace racially biased and other discriminatory language in our code and content with more inclusive language. While IBM values the use of inclusive language, terms that are outside of IBM's direct influence are sometimes required for the sake of maintaining user understanding. As other industry leaders join IBM in embracing the use of inclusive language, IBM will continue to update the documentation to reflect those changes.
Last modified on December 20, 2019