LUNA and the Google Cloud Vision API

Our October 2016 release introduces the integration of the Google Cloud Vision API to make use of Google’s image analysis capabilities in LUNA. This technology analyzes the content of an image, turning it into information that you can use to build your image catalog or supplement your own metadata. The Cloud Vision API features we implemented for LUNA include:

Label Detection: detecting broad sets of categories and keywords from images.

Landmark Detection: detecting natural or man-made structures and geographic locations in images.

Optical Character Recognition: detecting and transcribing text within an image, with support for a broad range of languages.

Now when you load content into LUNA, you can choose to make the information detected within images (labels, landmarks and transcripts) display as custom data fields to be used for search and display in the LUNA Viewer. You can also edit this information within the LUNA Library for greater control and flexibility.

The LUNA Viewer can display custom data fields created from information detected through the Google Cloud Vision API.  In this example the Vision API analyzed the image above and detected the following labels: "nature," "atmospheric phenomenon," "body of water," "water," and "morning."

In the example above the Vision API was able to transcribe the text, identify the landmark, and assign a number of labels to the image, including "street sign" and "commemorative plaque."