Eyeglass Prescription OCR for Pattern-Based Refractive Value Extraction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing technologies struggle to efficiently and reliably extract refractive values from standard and non-standard spectacle lens prescription documents, which are often drafted in various formats, hindering automated processing for producing spectacle lenses.

Innovation Solution

A processing device and system that utilize pattern recognition algorithms, combined with optical character recognition, to identify and validate refractive values from electronic representations of spectacle lens prescription documents, employing machine learning and neural networks for improved accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual data entry is used for extracting refractive values from prescription documents, then data extraction can be performed, but processing time is excessive and productivity is low

Engineering Contradiction:
Improveprocessing speedVSAvoiddata entry time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical data entry process with an automated optical character recognition (OCR) system combined with pattern recognition algorithms. The system captures images of prescription documents, automatically recognizes text and numerical data, extracts refractive values through pattern matching, and transfers data to the lens manufacturing system without human intervention, thereby dramatically improving processing speed and productivity.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by allowing the prescription document itself to serve as the data source. The OCR system automatically reads and interprets the document content, the pattern recognition algorithms independently identify refractive value patterns, and the system autonomously validates and processes the extracted data, eliminating the need for manual data transcription by opticians.

Inventive Principle:
Principle #25Self-service

2Productivity

If automated OCR systems are used for data extraction, then processing speed improves, but accuracy of extracting refractive values deteriorates due to various document formats

Engineering Contradiction:
Improveprocessing speedVSAvoiddata extraction accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The patent applies segmentation by dividing the prescription document into distinct regions and elements. The system identifies and segments different sections such as patient information, prescription data, and optical parameters. By segmenting the document structure and recognizing patterns within each segment, the system can accurately locate and extract refractive values even from varied document formats, maintaining high precision while achieving automated processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system employs parameter changes by adjusting recognition thresholds, pattern matching criteria, and validation parameters based on the detected document type and format. The pattern recognition algorithms dynamically modify extraction parameters to accommodate different prescription document styles, ensuring accurate identification of refractive values across various formats while maintaining automated processing speed.

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If multiple document formats are supported, then system versatility improves, but device complexity increases

Engineering Contradiction:
Improvedocument format compatibilityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements universality by designing a single integrated system that can process multiple document formats through a common architecture. The OCR engine, pattern recognition module, and validation system work together as a unified multi-functional platform that automatically adapts to different prescription document types. This universal approach allows the system to handle various formats without requiring separate specialized systems, managing complexity through consolidation rather than multiplication of components.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system applies dynamics by incorporating adaptive algorithms that dynamically adjust to different document formats during processing. The pattern recognition system dynamically identifies document structure, adapts extraction patterns based on detected format characteristics, and modifies validation criteria accordingly. This dynamic adaptation enables versatile format support while keeping the core system architecture relatively simple and manageable.

Inventive Principle:
Principle #15Dynamics

4Measurement precision

If manual validation of extracted data is performed, then data accuracy improves, but processing time increases and productivity decreases

Engineering Contradiction:
Improvedata validation accuracyVSAvoidprocessing throughput
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent implements feedback through an automated validation system that checks extracted refractive values against predefined criteria and logical constraints. The system provides immediate feedback on data validity, automatically correcting obvious errors or flagging anomalies for review. This automated feedback loop maintains high data accuracy without requiring manual validation, thereby preserving processing throughput and productivity while ensuring data quality.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4295263B1Devices and methods for processing eyeglass prescriptions
Publication Date: 2025.09.03 CARL ZEISS VISION INTERNATIONAL GMBH
  • EP4295263B1 patent drawingFigure 1A
  • EP4295263B1 patent drawingFigure 1B~1C
  • EP4295263B1 patent drawingFigure 1D

AI summary

The present invention relates to a processing device (214), a computer-implemented method (320) and a related computer program for determining data related to refractive values (112) from at least one spectacle lens prescription document (110), a system (210), a computer-implemented method and a related computer program for processing the at least one spectacle lens prescription document (110), and a method for producing at least one spectacle lens based on the data related to refractive values (112) as determined from the at least one spectacle lens prescription document (110). The processing device (214) for determining data related to refractive values (112) from at least one spectacle lens prescription document (110), wherein the processing device (214) is configured to: determine at least one pattern (114, 114',...) comprising the data related to the refractive values (112) from an electronic representation of the at least one spectacle lens prescription document (110); and determine text comprising the data related to the refractive values (112) from the electronic representation of the at least one spectacle lens prescription document (110), wherein the processing device (214) is configured to determine the data related to the refractive values (112) by identifying and using a surface distribution of the at least one pattern (114, 114',...) comprising the data related to the refractive values (112) over the electronic representation of the at least one spectacle lens prescription document (110). The present invention allows processing standard or non-standard spectacle lens prescription documents (110) drafted in a plurality of fashions and provided by eye care professionals, such as opticians, or by end consumers in one of a plurality of languages to determine the data related to the refractive values (112) in a reliable, scalable and automated fashion.