React Native OCR - v0.11.0
    Preparing search index...

    Type Alias OcrModel<L>

    Model configuration for the OCR pipeline: one fused detect/recognize PTE plus its options, with optional layout (group text into regions) and documentModels (orientation, dewarp, table) stages. Omit both for whole-page OCR.

    type OcrModel<L = never> = {
        documentModels?: DocumentModelsConfig;
        layout?: ObjectDetectorModel<"xyxy", L>;
        modelOpts: OcrModelOptions;
        modelPath: string;
        tableLabel?: string;
        visualLabels?: readonly string[];
    }

    Type Parameters

    • L = never

      The layout region class-label type.

    Index
    documentModels? layout? modelOpts modelPath tableLabel? visualLabels?
    documentModels?: DocumentModelsConfig

    The document-helper model for orientation, dewarp and table structure.

    layout?: ObjectDetectorModel<"xyxy", L>

    A layout detector; when set, text is grouped into its regions.

    modelOpts: OcrModelOptions

    Recognizer charset, detector strategy and the other model-level knobs.

    modelPath: string

    Local path or remote URL of the fused detect/recognize .pte.

    tableLabel?: string

    The layout label treated as a table (structure recognition + cell fill). Default 'table'.

    visualLabels?: readonly string[]

    Layout labels carrying no text; OCR skips these regions. Default ['image', 'chart', 'seal'].

    Settings
    Member Visibility

    On This Page
    Properties