Optical Prescription Image Parsing With Field-Based Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The non-standardized and unclear format of handwritten optical prescriptions, coupled with the inefficiency of conventional OCR technologies, leads to inaccurate and cumbersome manual classification and correction processes, making it difficult for users to understand and monitor their visual health.
Innovation Solution
A method and system for detecting, extracting, classifying, and correcting optical prescription data into predefined categories, using machine learning algorithms to enhance accuracy and efficiency in reading optical prescriptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If conventional OCR technology is used to convert handwritten optical prescriptions, then the conversion process is automated, but the accuracy of the converted text deteriorates due to the non-standardized format and the resulting incoherent string of characters
Solution Approach 1:
The patent segments the optical prescription text into distinct fields (sphere, cylinder, axis, add, PD) based on predefined categories and positional relationships. This segmentation allows the system to process each field independently with specific parsing rules, improving accuracy while maintaining automation. The segmentation transforms the incoherent OCR output into structured, meaningful data.
Solution Approach 2:
The patent introduces an intermediary processing layer between OCR conversion and final interpretation. This intermediary system applies classification rules, error detection algorithms, and correction mechanisms to bridge the gap between raw OCR output and accurate prescription data, resolving the accuracy issue while preserving automation.
2Measurement precision
If manual classification and correction of OCR converted text is performed, then the accuracy of the optical prescription data is improved, but the time consumption and operational complexity deteriorate
Solution Approach 1:
The patent implements self-service through automated error detection and correction algorithms that operate without human intervention. The system automatically identifies inconsistencies in OCR output, applies correction rules based on optical prescription standards, and validates the final data, eliminating the need for manual classification and correction while maintaining high accuracy.
Solution Approach 2:
The patent incorporates feedback mechanisms where the system continuously validates extracted data against predefined optical prescription formats and ranges. When errors are detected, the system automatically adjusts and re-processes the data, providing real-time feedback that ensures accuracy without requiring manual intervention and reducing time loss.
3Reliability
If manual classification of OCR converted text is performed, then the optical prescription data can be categorized correctly, but the operational complexity and skill requirements deteriorate
Solution Approach 1:
The patent applies parameter changes by transforming the classification problem into a set of predefined parameter extractions (sphere, cylinder, axis, add, PD). Each parameter has specific extraction rules and validation criteria, simplifying the classification process from a complex manual task to a structured automated procedure with clear parameters and decision logic.
4Productivity
If automated error correction is implemented, then the efficiency of the prescription processing is improved, but the risk of introducing new errors or incorrect corrections may worsen
Solution Approach 1:
The patent implements beforehand cushioning by incorporating multiple validation checks and error prevention mechanisms before final correction is applied. The system validates data ranges, checks for consistency across parameters, and uses confidence scoring to determine when correction is appropriate, cushioning against the risk of introducing new errors while maintaining processing efficiency.
Data Source
AI summary
A method for reading an optical prescription on an optical prescription image. The method includes detecting a region comprising the optical prescription on the optical prescription image; extracting the optical prescription and converting the optical prescription into machine-encoded optical prescription data; classifying a portion of the optical prescription data into one or more predetermined categories, to generate an optical prescription value associated with a respective one of the one or more predetermined categories; and determining whether the optical prescription value associated with the respective one of the one or more predetermined categories contains an error, and, if the optical prescription value contains the error, correcting the error within the optical prescription value, to generate a corrected optical prescription value associated with the respective one of the one or more predetermined categories. A system for reading an optical prescription on an optical prescription is also disclosed.