Live Camera Long Document Detection for Multi-Image OCR

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image capture technologies struggle to identify the length of documents accurately, leading to poor-quality images that hinder optical character recognition (OCR) performance and increase processing time, especially when documents are too long to fit within a single image frame.

Innovation Solution

A system and method for detecting long documents using a long document detector and an out-of-bounds detector to analyze factors like font size, bounding information, and dimensions/aspect ratio, alerting users to capture multiple images if the document is too long, and stitching them together for OCR processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a single image capture is used, then the capture process is simple and fast, but the document length cannot be accurately identified and OCR accuracy deteriorates for long documents

Engineering Contradiction:
Improvedocument length identification accuracyVSAvoidimage capture and processing system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the document capture process into multiple segments by detecting document boundaries and capturing separate images of different document portions. The system identifies the first and second portions of the document that extend beyond a single image frame, captures them separately, and processes them independently to achieve accurate full-document recognition.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary document boundary detection and length assessment before completing the full capture process. By analyzing the initial image to determine if the document extends beyond the frame, the system can proactively plan multiple captures and alert the user, preventing incomplete captures and ensuring accurate document length identification from the outset.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If multiple images are captured for long documents, then OCR accuracy improves, but processing time increases

Engineering Contradiction:
ImproveOCR accuracyVSAvoiddocument processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary detection of document boundaries and determines whether the document is too long for a single capture before proceeding with multiple image acquisitions. This advance planning allows the system to efficiently organize multiple captures and process them in parallel where possible, reducing overall processing time while maintaining high OCR accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system automatically detects when multiple captures are needed and manages the multi-image capture and processing workflow without requiring user intervention. The patent implements self-service by having the system autonomously determine capture requirements, execute multiple captures, and process all images through OCR, eliminating manual coordination overhead.

Inventive Principle:
Principle #25Self-service

3Reliability

If documents extending beyond image frame are processed, then complete document capture is attempted, but resource waste occurs on incomplete images

Engineering Contradiction:
Improvecomplete document capture reliabilityVSAvoidprocessing resource waste
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent performs preliminary analysis of the initial image to detect document boundaries and determine whether the document extends beyond the image frame. Based on this preliminary assessment, the system proactively alerts the user to reposition the document or prepares for multiple captures, preventing wasted processing resources on inherently incomplete single-image captures.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements preliminary anti-action by detecting the problem of documents extending beyond the frame and taking preventive measures before full processing occurs. The system alerts the user to reposition the document or initiates a multi-capture sequence, thereby preventing the harmful effect of processing incomplete images and wasting computational resources.

Inventive Principle:
Principle #9Preliminary anti-action

Data Source

PatentEP3639200B1Detecting long documents in a live camera feed
Publication Date: 2025.08.06 INTUIT INC
  • EP3639200B1 patent drawingFigure 1
  • EP3639200B1 patent drawingFigure 2
  • EP3639200B1 patent drawingFigure 3

AI summary

Aspects of the present disclosure provide methods and apparatuses for processing a digital image of a document, for example, to determine whether the document is a long document. An exemplary method generally includes obtaining a plurality of digital images of the document, determining a type of the document, loading one or more pre-defined metrics associated with the document based on the determined type of the document, determining one or more characteristics of the document based on one or more analyses performed on the plurality of digital images of the document, comparing the one or more characteristics of the document with the one or more pre-defined metrics, and determining the document to be a long document based, at least in part, on the comparison.