Tabular Data Extraction

with cloud vision

Posted on March 6, 2018

After some initial attempts around tabular OCR, I wanted to explore further the capabilities of Google’s Cloud Vision API for OCR processing jobs.

More Samples

The following are some additional tabular data tables in PNG format, with results returned from the Cloud Vision OCR engine below.

Sample 1

The first sample is a fairly clear inventory table with readable text.

Input

Result

Result

Sample 2

The second sample is a less clear extract from an old book.

Input

Result

Result

Conclusion

Overall, Google Cloud Vision seems like a very capable approach for quickly solving OCR tasks. Definitely worth keeping an eye on during project evaluation stages.

Next Steps

Next, I’m going to give Tesseract’s new LSTM support another try. Hopefully, they’ve made some progress towards a final release in the past few months.

More in this series…