Skip to content
All projects

2025 · Case study

InvoiceWise AI — Shop Bill OCR & Zoho Integration

Extracting structured ledger data from physical purchase receipts with OCR and AI.

A billing data extraction utility that processes scanned images of shop purchase bills. Utilizes the Tesseract OCR engine to read raw receipt text, structures it into formatted records, compiles it into downloadable Excel spreadsheets, and automatically pushes the extracted details to Zoho Books via API. Synced items and processing histories are backed up securely using Firestore database events.

  • TypeScript
  • Tesseract OCR
  • Excel Generation
  • Zoho API
  • Firebase Auth
  • Firestore DB
  • Node.js
  • Image Processing

01

The problem

Retailers and shop owners spend hours manually keying purchase details from supplier papers into accounting systems like Zoho.

Manual data entry is prone to typographical errors, mistyping line items, and missing expense records, which distorts accounting balances.

The goal was an automated pipeline where a user uploads a photo of a bill, and structured rows are instantly created in Zoho and Excel.

02

Architecture

The system ingests receipt photos, running them through Tesseract OCR to perform local optical character recognition and text extraction.

A serverless parser sanitizes the OCR output, parsing items, quantities, and totals into unified objects matching Zoho's API schemas.

Data is stored in Firestore for audits, synchronized directly with Zoho books, and converted into downloadable spreadsheet files.

OCR Extraction & Multi-Destination Sync Pipeline
Input
  • Bill Image Upload (Receipt/Invoice JPG/PNG)
Extraction
  • Tesseract OCR Engine (local character recognition)
Storage & Backend
  • Firebase Cloud Functions
  • Firestore DB logs
Sync & Export
  • Excel JS Spreadsheet Compiler
  • Zoho API Connector

Raw image text is segmented by OCR, structured via serverless handlers, then dispatched to Zoho and generated Excel sheets.

03

Tech stack & why

  • OCR Processing

    Tesseract OCR Engine

    Provides robust offline text extraction from receipt scans without proprietary cloud dependencies.

  • External Sync

    Zoho REST API integration

    Automatically populates inventory and expense ledgers directly from processed receipts.

  • Spreadsheet Output

    Excel JS Compiler

    Allows shop owners to download structured files for local backups and tax filings.

  • Database & Events

    Firebase & Firestore

    Ensures transaction logging, processing status tracking, and secure database syncing.

04

Challenges & solutions

  • 1

    Parsing Tabular OCR Text

    Receipt scans often align poorly, causing OCR columns to shift. I designed regex pattern classifiers to group item names with corresponding price columns based on line heights.

  • 2

    Rate-Limiting on External APIs

    Pushing bulk receipt items to Zoho can trigger rate limit errors. I built a queue handler in Firebase Cloud Functions that throttles API calls and handles retries gracefully.

05

What I'd build next

  • Integrate LLM-based receipt correction to fix character mistakes made by OCR on low-quality camera captures.
  • Add automated email inbox scanning to process PDF bills directly as they arrive from vendors.

Next case study

AI POS System

Read it