import { LiteParseConfig, PageComplexityStats } from '@llamaindex/liteparse';
export interface DetectPdfNeedsOcrOptions {
    liteParseOptions?: Partial<LiteParseConfig>;
}
export interface PdfOcrAssessment {
    /** True when any page needs OCR or other advanced parsing. */
    needsOcr: boolean;
    /** Per-page complexity signals, one entry per page. */
    pages: PageComplexityStats[];
    /**
     * Every distinct reason found across all pages, e.g. `"scanned"`,
     * `"no-text"`, `"sparse-text"`, `"embedded-images"`, `"garbled"`,
     * `"vector-text"`. Empty when `needsOcr` is false.
     */
    reasons: string[];
}
/**
 * Determines whether a PDF (URL or ArrayBuffer) needs OCR to extract usable
 * text, without running a full parse.
 * @param pdfURLOrBuffer - URL to a PDF file or buffer from fs.readFile
 * @category Extract
 */
export declare function detectPdfNeedsOcr(pdfURLOrBuffer: any, options?: DetectPdfNeedsOcrOptions): Promise<PdfOcrAssessment>;