Sometimes it can be difficult to avoid sharing a scanned PDF or photo with your students. For instance, you might want them to read an out-of-print manuscript that’s not available in any other format. The unfortunate down side, however, is that these types of files are not accessible to students who rely on screen readers or other technology to assist them when completing assigned readings.
Luckily, as UIS faculty, you have access to free tools that allow you to quickly and accurately convert PDFs and photo files to text, which is essential in making your course content accessible to all students. This can be accomplished through optical character recognition in Google Drive. (All UIS staff, faculty, and students have access to Google products with their existing UIS NetID and password through our educational license with Google; for more information, visit go.uis.edu/google.)
To get started, follow these steps:
- Go to drive.google.com.
- Sign in with your UIS NetID and password.
- Upload your PDF or image file. (instructions)
- Hover over the file in Google Drive, right-click, and select Open with > Google Docs.
The image file will then be converted to a Google Doc, and any text that was recognized in the image will appear below the original image within the Google Doc. (For multi-page PDFs, each page will appear as a separate image, with the text underneath.)
Once doing this, it is important to scan the new text for accuracy, and correct any errors that you find. Most of the time, formatting (bold, italics, etc.) will be retained.
Please also keep this in mind:
- Only these file types are currently supported: JPEG, PNG, GIF, and PDF
- Text should be right-side up. If your image or PDF is facing the wrong way, you must rotate it before uploading it to Google Drive.
- The maximum file size supported by Google at this time for optical character recognition is 2MB. This means that larger, multi-page PDFs may need to be separated into several different files before starting.
- Tables, columns, footnotes, etc. are not likely to be detected.
- For the best accuracy:
- Text should be at least 10 pixels in height.
- Common fonts like Arial or Times New Roman work best.
- Try to avoid uneven lighting or blurry photos.