yujunjun/unstructured
1
0
Fork 0
mirror of https://github.com/Unstructured-IO/unstructured.git synced 2025-08-18 13:45:45 +00:00
Code Issues Packages Projects Releases Wiki Activity

docs: add information about MIME type of extracted images (#2515)

Browse Source 
Include information about what mime type is expected when extracting
images.

Co-authored-by: Filip Knefel <filip@unstructured.io>
This commit is contained in:
Filip Knefel 2024-02-07 09:40:24 +01:00 committed by GitHub
parent be71633415
commit 5defe79bf2
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docs/source/core/partition.rst
View File

@ -634,7 +634,7 @@ The available strategies for PDFs are ``"auto"``, ``"hi_res"``, ``"ocr_only"``,
* The ``"fast"`` strategy will extract the text using ``pdfminer`` and process the raw text with ``partition_text``. If the PDF text is not extractable, ``partition_pdf`` will fall back to ``"ocr_only"``. We recommend using the ``"fast"`` strategy in most cases where the PDF has extractable text. * The ``"fast"`` strategy will extract the text using ``pdfminer`` and process the raw text with ``partition_text``. If the PDF text is not extractable, ``partition_pdf`` will fall back to ``"ocr_only"``. We recommend using the ``"fast"`` strategy in most cases where the PDF has extractable text.
To extract images and elements as image blocks from a PDF, it is mandatory to set ``strategy="hi_res"`` when setting ``extract_images_in_pdf=True``. With this configuration, detected images are saved in a specified directory or encoded within the file. However, keep in mind that ``extract_images_in_pdf`` is being phased out in favor of ``extract_image_block_types``. This option allows you to specify types of images or elements, like "Image" or "Table". If some extracted images have content clipped, you can adjust the padding by specifying two environment variables "EXTRACT_IMAGE_BLOCK_CROP_HORIZONTAL_PAD" and "EXTRACT_IMAGE_BLOCK_CROP_VERTICAL_PAD" (for example, EXTRACT_IMAGE_BLOCK_CROP_HORIZONTAL_PAD = 20, EXTRACT_IMAGE_BLOCK_CROP_VERTICAL_PAD = 10). For integrating these images directly into web applications or APIs, ``extract_image_block_to_payload`` can be used to convert them into ``base64`` format, including details about the image type. Lastly, the ``extract_image_block_output_dir`` can be used to specify the filesystem path for saving the extracted images when not embedding them in payloads. To extract images and elements as image blocks from a PDF, it is mandatory to set ``strategy="hi_res"`` when setting ``extract_images_in_pdf=True``. With this configuration, detected images are saved in a specified directory or encoded within the file. However, keep in mind that ``extract_images_in_pdf`` is being phased out in favor of ``extract_image_block_types``. This option allows you to specify types of images or elements, like "Image" or "Table". If some extracted images have content clipped, you can adjust the padding by specifying two environment variables "EXTRACT_IMAGE_BLOCK_CROP_HORIZONTAL_PAD" and "EXTRACT_IMAGE_BLOCK_CROP_VERTICAL_PAD" (for example, EXTRACT_IMAGE_BLOCK_CROP_HORIZONTAL_PAD = 20, EXTRACT_IMAGE_BLOCK_CROP_VERTICAL_PAD = 10). For integrating these images directly into web applications or APIs, ``extract_image_block_to_payload`` can be used to convert them into ``base64`` format, including details about the image type, currently it's always ``image/jpeg``. Lastly, the ``extract_image_block_output_dir`` can be used to specify the filesystem path for saving the extracted images when not embedding them in payloads.
Examples: Examples: