File size: 5,129 Bytes
3f7e152
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
import os
from pathlib import Path
from docling.datamodel.document import ConversionResult
from huggingface_hub import snapshot_download

from docling.datamodel.base_models import InputFormat
from docling.datamodel.pipeline_options import EasyOcrOptions, OcrMacOptions, PdfPipeline, PdfPipelineOptions, PipelineOptions, RapidOcrOptions, TesseractOcrOptions
from docling.document_converter import DocumentConverter, ImageFormatOption, PdfFormatOption
from docling.backend.pypdfium2_backend import PyPdfiumDocumentBackend, PyPdfiumPageBackend
from docling.pipeline.standard_pdf_pipeline import StandardPdfPipeline
from docling.pipeline.simple_pipeline import SimplePipeline


# TODO: REFACTOR LOAD OCR MODEL TO JUST EITHER USE SERVER MODELS OR MOBILE MODELS
def load_rapid_ocr_model(det_model: str, rec_model: str, cls_model: str) -> DocumentConverter:
    """
    Load the RapidOCR model from Hugging Face Hub.
    Args:
        det_model (str): Path to the detection model.
        rec_model (str): Path to the recognition model.
        cls_model (str): Path to the classification model.
    Returns:
        DocumentConverter: The loaded RapidOCR model.
    """
    print("Downloading RapidOCR models")
    download_path = snapshot_download(repo_id="SWHL/RapidOCR")

    det_model_path = os.path.join(
        download_path, det_model
    )
    rec_model_path = os.path.join(
        download_path, rec_model
    )
    cls_model_path = os.path.join(
        download_path, cls_model
    )

    ocr_options = RapidOcrOptions(
        det_model_path=det_model_path,
        rec_model_path=rec_model_path,
        cls_model_path=cls_model_path
    )

    pipeline_options = PdfPipelineOptions(
        ocr_options=ocr_options
    )

    doc_converter = DocumentConverter(
        format_options={
            InputFormat.IMAGE: ImageFormatOption(
                pipeline_options=pipeline_options
            )
        }
    )

    return doc_converter

def load_ocr_mac_model() -> DocumentConverter:
    """
    Load the OCR Mac model.
    Returns:
        DocumentConverter: The loaded OCR Mac model.
    """
    ocr_options = OcrMacOptions(
        framework='vision'
    )

    pipeline_options = PdfPipelineOptions(
        ocr_options=ocr_options
    )

    doc_converter = DocumentConverter(
        allowed_formats=[
            InputFormat.PDF,
            InputFormat.IMAGE,
        ],
        format_options={
            InputFormat.PDF: PdfFormatOption(
                pipeline_cls=StandardPdfPipeline, backend=PyPdfiumDocumentBackend, pipeline_options=pipeline_options
            ),
            InputFormat.IMAGE: PdfFormatOption(
                pipeline_cls=StandardPdfPipeline, backend=PyPdfiumDocumentBackend, pipeline_options=pipeline_options
            )
        }
    )
    
    return doc_converter

def load_tesseract_model(tessdata_path: str) -> DocumentConverter:
    """
    Load the Tesseract OCR model. 
    Args:
        tessdata_path (str): Path to the Tesseract data directory.
    Returns:
        DocumentConverter: The loaded Tesseract OCR model.
    """
    os.environ["TESSDATA_PREFIX"] = tessdata_path

    ocr_options = TesseractOcrOptions()

    pipeline_options = PdfPipelineOptions(
        ocr_options=ocr_options
    )

    doc_converter = DocumentConverter(
        allowed_formats=[
            InputFormat.PDF,
            InputFormat.IMAGE
        ],
        format_options={
            InputFormat.PDF: PdfFormatOption(
                pipeline_cls=StandardPdfPipeline, backend=PyPdfiumDocumentBackend, pipeline_options=pipeline_options
            ),
            InputFormat.IMAGE: PdfFormatOption(
                pipeline_cls=StandardPdfPipeline, backend=PyPdfiumDocumentBackend, pipeline_options=pipeline_options
            )
        }
    )

    return doc_converter

def load_easy_ocr_model() -> DocumentConverter:
    """
    Load the EasyOCR model.
    Returns:
        DocumentConverter: The loaded EasyOCR model.
    """
    ocr_options = EasyOcrOptions()

    pipeline_options = PdfPipelineOptions(
        ocr_options=ocr_options
    )

    doc_converter = DocumentConverter(
        allowed_formats=[
            InputFormat.PDF,
            InputFormat.IMAGE
        ],
        format_options={
            InputFormat.PDF: PdfFormatOption(
                pipeline_cls=StandardPdfPipeline, backend=PyPdfiumDocumentBackend, pipeline_options=pipeline_options
            ),
            InputFormat.IMAGE: PdfFormatOption(
                pipeline_cls=StandardPdfPipeline, backend=PyPdfiumDocumentBackend, pipeline_options=pipeline_options
            )
        }
    )
    return doc_converter

def image_to_text(document_converter: DocumentConverter, file_path: Path) -> ConversionResult:
    """
    Convert an image to text using the specified document converter.
    Args:
        document_converter (DocumentConverter): The document converter to use.
        file_path (Path): Path to the image file.
    Returns:
        ConversionResult: The result of the conversion.
    """
    conv_results = document_converter.convert(file_path)
    return conv_results