Azure AI Document Intelligence
An Azure service that turns documents into usable data. Previously known as Azure Form Recognizer.
How to extract images from a document using Azure Document Intelligence?
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(262.40000000000003, 53%, 35%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EHM%3C/text%3E%3C/svg%3E)
Haritha Maturi
5
Reputation points
As per the documentation for prebuilt-layout model they are saying
But I couldn't understand how exactly I can access the cropped images when I trying to do it in python
like below:
result = documentintelligenceclient.begin_analyze_document( "prebuilt-layout",AnalyzeDocumentRequest(bytes_source=document),output=['figures'])
If I use UI I can see the download option to download the images
But I want the images accessible while doing things in python either the image itself or the bytes of image would also help.
Please help me.
Thanks Haritha
Azure AI Document Intelligence
{count} vote
-
romungi-MSFT 49,056 Reputation points Microsoft Employee Moderator2024-09-18T12:04:51.68+00:00 @Haritha Maturi While using the studio, i have observed that the download option is not available when a document with figure is analyzed with the layout model.
It only provide the bounding regions of the figure if i hover the Figures tab. I think the response in the sample from github repo is also providing options to get the figures object from the response with page number and bounding regions.
if result.figures: for figures_idx,figures in enumerate(result.figures): print(f"Figure # {figures_idx} has the following spans:{figures.spans}") for region in figures.bounding_regions: print(f"Figure # {figures_idx} location on page:{region.page_number} is within bounding polygon '{region.polygon}'")If there is no caption i see something like this with no download option.
-
romungi-MSFT 49,056 Reputation points Microsoft Employee Moderator
2024-09-20T10:30:20.4433333+00:00 @Haritha Maturi Did you get a chance to look at the above and confirm if you are seeing similar behavior?
-
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(35.2, 85%, 13%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ELS%3C/text%3E%3C/svg%3E)
Logeshwaran S 0 Reputation points2024-09-29T14:18:29.59+00:00 I have the similar question. How to get the image using Document Intelligence through Python. While there's a method called get_analyze_result_figure for the class DocumentIntelligentClient, I'm not sure how to render image out of it's output. Kindly assist on the same. Thanks Logesh S
-
romungi-MSFT 49,056 Reputation points Microsoft Employee Moderator
2024-10-01T15:54:36.1733333+00:00 @Logeshwaran S See this thread where a user posted how they are using the method get_analyze_result_figure to download images locally through SDK.
Sign in to comment
Sign in to answer