VisualRecogClassify

Classifies image by using IBM Watson Visual Recognition API.

Syntax

bool VisualRecogClassify (string ClassifierName)

Parameters

string ClassifierName - Name of the Classifier to be used

Returns

True, if the action succeeds. Otherwise, False

Level

Page level.

Details

This action identifies a page by using the Watson™ Recognition technology. This technology analyzes the image-based full page and attempts to find match within the classes that have been defined for the selected classifier. If a match is found, the Page type is populated with the ID of the category that was matched.

If a match is not found, the page type is set to "Other".

When classification is complete, a list of matches and their confidence values are stored in the "MatchingCategoryX" and "MatchingCategoryConfX" variables. The number of matches is stored in the variable "MatchingCategoriesCount".

To run classification without updating the page type, set the variable "UpdateDCOType" to "0" before calling this action. In this case, classification does not update the page type, but the variables that are mentioned above is populated.

Because this action supports image files (.jpg, .jpeg or .png) only. ConvertToJPEG must be called before using the Classify action for other types of files.

The ClassifierName can be a smart parameter.

Example
VisualRecogSetURL("@APPVAR(values/gen/url)") 
VisualRecogSetCredentials("@APPVAR(values/adv/VRAPIKey)") 
VisualRecogClassify("@APPVAR(values/gen/VRClassifierName)") 
Note: VisualRecogSetURL action is not included if your IBM Visual Recognition token is created before May 2018.