This package is extremely useful to gain insights into the wealth of knowledge hiding in your images by assigning labels to images and quickly classify them into millions of predefined categories. Detect objects and faces, read printed and handwritten text, and build valuable metadata into your image catalog. This package comprises an array of activities, such as Scope, Label Detection, Face Detection, Text Detection, Logo Detection, Web Detection, Handwriting Detection, Landmark Detection and Safe Search. The output returns bounding positions, confidence score along with apt descriptions.

Introducing Google Cognitive: This package is extremely useful to gain insights into the wealth of knowledge hiding in your images.                                                                                                                       

• Google Vision Scope: Sets up connection and authentication for Google Vision activities. The Google Vision activities work with the account specified inside the Application Scope. Multiple authentication types are available. The authentication type that will be used is the Service account Key with the key type being Json file
• Label Detection:  Generates description labels for the specified image and will return all labels found in an image in a JSON format
• Face Detection: Identifies and retrieves a face from an image. This activity will return a JSON output, which holds the bounding rectangle locations of detected faces in the given image, along with confidence score
• Text Detection: Returns the detected text from a given image along with its bounding positions and confidence percentage
• Logo Detection: Returns all logos found in the given image along with description, bounding positions and confidence score
• Web Detection: Searches for images on the website that matches the specified image. This activity will detect and return related web information such as URLs for fully matching images, partial matching images, pages with matching images along with confidence score
• Handwriting Detection: Identifies and reads handwritten text in an image.  The identified text will be returned as String
• Landmark Detection: Detects landmarks in the specified image along with descriptions and the confidence score for each identified landmark
• Safe Search: Returns the adult and racy content if present in the given image, along with medical and violent content.

To read about all the activities in this Feature group, use the left-hand menu and find it under Activities section.