Desktop OCR Data Comparison Tool

I need a Windows-based desktop application that lets me drag-and-drop or browse for one or more JPG files, runs OCR on each image, and pulls out both the text and any embedded graphics it can recognise. Once the content is captured, the program should lay it out in a clean, column-driven .xlsx file.

The next step is automatic reconciliation against a master Excel workbook I will supply at run-time. After the comparison, the tool must generate a single downloadable workbook that contains:

• Sheet 1 – “Matched” rows copied from the master and the new extract, clearly highlighted.
• Sheet 2 – “Unmatched” rows that exist only in one source or the other, also highlighted so they are obvious at a glance.

A simple progress bar or status panel is enough for feedback while the OCR and comparison run. Saving the workbook to a location of the user’s choice completes the workflow.

Acceptance criteria
• Runs natively on Windows 10/11 without additional licensing costs.
• Accurately captures text & images from standard-resolution JPGs.
• Produces a structured .xlsx file with no data loss.
• Correctly flags every match / mismatch against the provided master sheet.
• Exports the final workbook in under two minutes for a batch of 20 images on a mid-range PC.

I am open to popular OCR libraries such as Tesseract, Windows.Media.Ocr, or a C# Computer Vision API—whatever you are most efficient with—as long as the installer is lightweight and the source code is handed over.