docs/source/templates/use-cases/ocr_ed.md
Optical Character Recognition (OCR) labeled data is vital for AI applications in educational assessment, enabling models to accurately extract and interpret written text from scanned documents. This capability facilitates tasks such as scoring exams, analyzing student submissions, and automating administrative processes, which can ultimately enhance the quality and efficiency of educational evaluation.
However, the data labeling process for OCR presents significant challenges: it is often time-intensive due to the volume of documents that require attention, can suffer from inconsistency due to varied handwriting or text formatting, and necessitates domain expertise to ensure accurate interpretation of educational content. Label Studio effectively addresses these challenges through its hybrid AI-assisted pre-labeling, which accelerates initial labeling efforts while maintaining high accuracy rates. The platform's expert validation feature ensures that labels meet the required standards, empowering teams to collaborate seamlessly with customizable templates tailored specifically for educational assessment tasks. This approach not only streamlines the labeling process but also enhances model performance and scalability, allowing organizations to process large volumes of data efficiently and effectively.
<a href="https://app.humansignal.com/b/NTQx" target="_blank" rel="noopener" aria-label="Open in Label Studio" style="all:unset;cursor:pointer;display:inline-flex;align-items:center;justify-content:center;border-radius:4px;border:1px solid rgb(109,135,241);padding:8px 12px;background:rgb(87 108 193);color:white;font-weight:500;font-family:sans-serif;gap:6px;transition:background 0.2s ease;" onmouseover="this.style.background='rgb(97 122 218)'" onmouseout="this.style.background='rgb(87 108 193)'"> <svg style="width:20px;height:20px" viewBox="0 0 26 26" fill="none"><path fill="#FFBAAA" d="M3.5 4.5h19v18h-19z"/><path fill-rule="evenodd" clip-rule="evenodd" d="M25.7 7.503h-7.087V5.147H7.588V2.792h11.025V.436H25.7v7.067Zm-18.112 0H5.225v10.994H2.863V7.503H.5V.436h7.088v7.067Zm0 18.061v-7.067H.5v7.067h7.088ZM25.7 18.497v7.067h-7.088v-2.356H7.588v-2.355h11.025v-2.356H25.7Zm-2.363 0V7.503h-2.363v10.994h2.363Z" fill="#FF7557"/></svg> <span style="font-size:14px">Open in Label Studio</span> <svg style="width:16px;height:16px" viewBox="0 0 24 24"><path d="M14,3V5H17.59L7.76,14.83L9.17,16.24L19,6.41V10H21V3M19,19H5V5H12V3H5C3.89,3 3,3.9 3,5V19A2,2 0 0,0 5,21H19A2,2 0 0,0 21,19V12H19V19Z" fill="white"/></svg> </a>
<View>
<Image name="image" value="$ocr"/>
<Labels name="label" toName="image">
<Label value="Multiple Choice" background="green"/>
<Label value="Essay" background="blue"/>
<Label value="True/False" background="orange"/>
<Label value="Matching" background="purple"/>
</Labels>
<Rectangle name="bbox" toName="image" strokeWidth="3"/>
<Polygon name="poly" toName="image" strokeWidth="3"/>
<TextArea name="transcription" toName="image"
editable="true"
perRegion="true"
required="true"
maxSubmissions="1"
rows="5"
placeholder="Transcribed Question or Answer Text"
displayMode="region-list"
/>
</View>
This labeling configuration lets you perform educational assessment annotation tasks by identifying regions on the image with shapes and transcribing the corresponding question or answer text. You can select the label type (e.g., "Multiple Choice") and then highlight relevant text regions to associate transcriptions with each region.
All labeling configurations must be wrapped in <a href="https://labelstud.io/tags/view">View</a> tags.
Use the <a href="https://labelstud.io/tags/image">Image</a> object tag to specify the image to label:
<Image name="image" value="$ocr"/>
Use the <a href="https://labelstud.io/tags/labels">Labels</a> control tag to specify which labels are available to apply to the different shapes added to the image:
<Labels name="label" toName="image">
<Label value="Multiple Choice" background="green"/>
<Label value="Essay" background="blue"/>
<Label value="True/False" background="orange"/>
<Label value="Matching" background="purple"/>
</Labels>
You can change the <code>value</code> of each <code>Label</code> to assign different assessment item types to regions in the educational content image.
Use the <a href="https://labelstud.io/tags/rectangle">Rectangle</a> control tag to add unlabeled rectangles:
<Rectangle name="bbox" toName="image" strokeWidth="3"/>
Choosing the Rectangle tag instead of <a href="https://labelstud.io/tags/rectanglelabels">RectangleLabels</a> means annotators can first create regions highlighting questions or answers, then assign labels, then transcribe the text for each region. This workflow facilitates stepwise annotation in educational assessment.
Use the <a href="https://labelstud.io/tags/polygon">Polygon</a> control tag to add unlabeled polygons:
<Polygon name="poly" toName="image" strokeWidth="3"/>
The <code>strokeWidth</code> argument controls the thickness of the polygon outline.
Use the <a href="https://labelstud.io/tags/textarea">TextArea</a> control tag to add transcriptions for each region drawn on the image:
<TextArea name="transcription" toName="image"
editable="true"
perRegion="true"
required="true"
maxSubmissions="1"
rows="5"
placeholder="Transcribed Question or Answer Text"
displayMode="region-list"
/>
The <code>editable="true"</code> setting allows annotators to revise transcription after submitting, and <code>displayMode="region-list"</code> shows the text boxes in the region list linked to each shape, easing updates. The <code>perRegion="true"</code> attribute means each transcription belongs to a specific region, and <code>required="true"</code> enforces transcription input before submissions. The <code>placeholder</code> text guides annotators before editing.
Related tags: