Operation op_image_analysis_openai for intelligent image analysis

To analyze images with Artificial Intelligence you can use the following operation:

op_image_analysis_openai 

This operation saves the result of the analysis in a document feature with the following name:

openai.image.analysis

The image analysis is very flexible and will depend on the context that has been specified. The context is defined directly in the transaction.

For example, in the case of an insurance company that needs to analyze car images to determine damage, we can define a promt as follows:

Analyze the possible damages that the car shown in the image may have (if any) from among the following:

Scratch (internal value “scratch”): marks or scratches on the surface.
Dent (internal value “dent”): Surface depression caused by a blow
Crack/Crack (internal value crack_crack): crack or opening in which the car is not completely divided
Perforation (internal value perforation): Hole produced on surface, probably by blow or impact against a blunt object
Damaged glass (internal value glass_damaged): glass or windshield of the car with any kind of damage affecting the driver's vision
Worn paint (internal value worn_paint): car's exterior paintwork worn and/or affected by an accident.

Return me as answer only and exclusively the internal value of the type of damage that it presents (it can only be one, the clearest), in case of not presenting it, return “NA”.

We will write this promt in the operation:

So, for the next image...

After the analysis, the operation will return the corresponding damage in a feature as shown below.

The feature can be exploited as you wish, for example, you can transfer its value to a field with a dynamic expression.