Custom OCR Java

The JAVA OCR SDK supports custom-built API from the API Builder.
If your document isn't covered by one of Mindee's Off-the-Shelf APIs,
you can create your own API using the API Builder.

For the following examples, we are using our own W9s custom API
created with the API Builder.

Quick Start

String path = "/path/to/the/file.ext";
DocumentToParse documentToParse = new DocumentToParse(new File(path));
CustomEndpoint myEndpoint = new CustomEndpoint(
    "wnine",
    "john",
    "1.0" // optional
);

Document<CustomV1Inference> customDocument = mindeeClient
    .parse(documentToParse, myEndpoint);

If the version argument is set, you'll be required to update it every time a new model is trained.
This is probably not needed for development but essential for production use.

Parsing Documents

Use the ParseAsync method to call the API prediction on your custom document.
The response class and document type must be specified when calling this method.

You have two different ways to parse a custom document.

1. Use the default one (named CustomPrediction):

String path = "/path/to/the/file.ext";
DocumentToParse documentToParse = new DocumentToParse(new File(path));
CustomEndpoint myEndpoint = new CustomEndpoint(
    "wnine",
    "john",
    "1.0" // optional
);

Document<CustomV1Inference> customDocument = mindeeClient.parse(documentToParse, myEndpoint);

2. You can also use your own class which will represent the required fields. For example:

// The CustomEndpointInfo annotation is required when using your own model.
// It will be used to know which Mindee API to call.

public class WNineV1DocumentPrediction {
  @JsonProperty("name")
  private StringField name;

  @JsonProperty("employerId")
  private StringField employerId;
  
  (...)
}

@EndpointInfo(endpointName = "wnine", accountName = "john" version = "1")
public class WNineV1Inference
  extends Inference Inference<WNineV1DocumentPrediction, WNineV1DocumentPrediction> {
}

String path = "/path/to/the/file.ext";
DocumentToParse documentToParse = new DocumentToParse(new File(path));

Document<WNineV1Inference> myCustomDocument = mindeeClient
    .parse(WNineV1Inference.class, documentToParse);

CustomV1Inference object

All the fields which are present in the API builder
are available (the fields are defined when creating your custom API).

CustomV1Inference is an object which contains a document prediction and pages prediction result.

CustomV1PagePrediction

Which is a HashMap<String, ListField> with the key as a string for the name of the field, and a ListField as a value.

CustomV1DocumentPrediction

Which contains 2 properties : classificationFields and fields.
Both are a Map and the key is a string for the name of the field and for the value :

• classificationFields have a ClassificationField object as value. Each ClassificationField contains a value.
• fields have a ListField object as value. Each ListField contains a list of all values extracted for this field.

📘

Info

Both document level and page level objects work in the same way.

Fields property

A Map with the following structure:

• confidence: a double
• values: a list of ListFieldValue which containing a list of all values found for the field.

In the examples below we'll use the employer_id field.

String path = "/path/to/the/file.ext";
DocumentToParse documentToParse = new DocumentToParse(new File(path));

Document<WNineV1Inference> myCustomDocument = mindeeClient
    .parse(WNineV1Inference.class, documentToParse);

ListField employerId = document.getInference().getDocumentPrediction().get("employer_id");

Questions?

Join our Slack