OneDrive & SharePoint: Extract Text from Any Image

Key insights

• Copilot Integration: Microsoft has introduced a feature in OneDrive and SharePoint that allows users to ask Copilot to explain what is in an image and extract text from it, making images more interactive and useful.

• Optical Character Recognition (OCR): When you upload an image, OCR automatically scans the picture for any text. The detected text is stored in a special metadata column called "Extracted Text", making it easy to find through search.

• Searchable Content: Extracted text from images becomes indexable and searchable. Users can search for images using keywords found within the pictures, not just by file name or tags.

• No Manual Tagging Needed: The system uses AI to add image tags, so users do not need to manually describe or tag images. This saves time and ensures consistent tagging across all files.

• Differentiation Between Services: Extraction happens faster in SharePoint, usually within a minute, while OneDrive may take up to 24 hours. Only new uploads are processed, and embedded images inside Office files are not supported.

• Business Automation Support: Integration with tools like Power Automate, AI Builder, and Azure AI Document Intelligence enables advanced workflows, allowing organizations to automate processes based on extracted image text.

Introduction to Microsoft 365's New Image Text Extraction Feature

Microsoft has introduced a significant update to its OneDrive and SharePoint platforms, enhancing their ability to process images with the help of advanced artificial intelligence. According to a recent you_tube_video by Ami Diamond [MVP], users can now not only view and store images but also ask Microsoft's Copilot what is depicted in a picture and extract any text present in the image. This innovative feature leverages Optical Character Recognition (OCR) technology and integrates seamlessly with Microsoft's AI solutions, such as AI Builder and Azure AI Document Intelligence.

With this enhancement, users gain new ways to interact with their image files, making it easier to search, organize, and utilize visual information stored across OneDrive and SharePoint. The feature is especially useful for professionals who work with large volumes of documents and images, as it simplifies the process of turning visual data into actionable content.

How Text Extraction Works in OneDrive and SharePoint

When an image is uploaded to OneDrive or SharePoint, the system automatically activates OCR in the background. This process scans the image and identifies any text embedded within it. The extracted text is then stored in a dedicated metadata column called "Extracted Text". This enables the content to be indexed and made searchable within the Microsoft 365 environment.

Moreover, the technology attempts to generate additional image tags using AI recognition, further improving the relevance of search results. The system supports images up to 50 MB in size and resolutions ranging from 50 x 50 pixels to a substantial 16,000 x 16,000 pixels. However, it's important to note that OCR processing applies only to new images uploaded after the feature is enabled, and embedded images within Office documents are not supported. Additionally, for PDF and TIFF files, while the extracted text is indexed for search, it does not appear in the "Extracted Text" column.

Practical Applications and User Experience

This technology can be described as a smart document indexing system that not only reads text within images but also tags them internally for improved searchability. For example, if someone uploads a photo of a receipt, details such as vendor names, totals, dates, or addresses become discoverable through Microsoft Search. Thus, images are transformed from static files into dynamic sources of information.

In practical terms, users can request Copilot to explain the contents of a picture or extract the text for use in Word or PowerPoint documents. This streamlined integration means that manual tagging and description entry are no longer required, saving time and reducing the risk of human error. However, differences exist in processing times; SharePoint typically extracts and indexes text within a minute, while OneDrive may take up to 24 hours, presenting a tradeoff between platform speed and user convenience.

Benefits and Tradeoffs of AI-Powered Image Processing

The main advantage of this feature is improved discoverability. Users can now search for images based on their textual content, which is vital in environments where large numbers of files are stored. Automatic metadata enrichment further enhances organizational efficiency, as extracted text is incorporated into broader Microsoft 365 searches, spanning emails, documents, and more.

On the other hand, the challenge lies in balancing speed and accuracy. While AI-driven extraction provides significant automation, it is not infallible, and occasional errors in text recognition may occur. Additionally, the difference in processing speed between SharePoint and OneDrive could affect how quickly users can access newly extracted data, especially in time-sensitive situations.

Integration with Business Automation and AI Advancements

Microsoft's integration of AI Builder and Azure AI Document Intelligence allows organizations to train and extend text extraction models beyond basic OCR, offering greater flexibility and accuracy. Furthermore, the connection with Power Automate enables users to create automated workflows that act on extracted text, supporting business processes such as document indexing, compliance, and data analysis.

A standout aspect of this update is the inclusion of the "Extracted Text" metadata column in SharePoint libraries. This provides administrators and users with a transparent way to verify and review the content that has been captured by the system. Meanwhile, improved search integration ensures that images with embedded text are easily discoverable under the “Images” tab in Microsoft Search, greatly enhancing findability across the organization.

Conclusion: Transforming Image Management in Microsoft 365

In summary, Microsoft's new image text extraction capabilities in OneDrive and SharePoint represent a meaningful step forward in workplace productivity. By combining OCR, AI-driven tagging, and seamless integration with Microsoft 365 services, users can unlock more value from their visual content. While there are tradeoffs in processing speed and occasional accuracy issues, the benefits of automation, discoverability, and business process support are clear.

As organizations continue to digitize their workflows, features like these will become increasingly essential for managing and making sense of vast amounts of visual data. With ongoing advancements in AI, the potential for even smarter and more accurate image management within Microsoft 365 looks promising.

Keywords

Explain picture OneDrive SharePoint extract text image OCR Microsoft 365 image recognition