Form Orientation Detection via Template Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing imaging readers require cumbersome configuration and orientation detection for reading forms, often necessitating manual rotation and OCR processing, which is slow and inefficient, especially when forms lack orientation marks or asymmetrical margins.
Innovation Solution
An electro-optical reader system with a controller that automatically identifies forms and their correct orientation by matching captured image fields with stored templates, using a solid-state imager and processing techniques like OCR, OMR, and ICR, without the need for initial rectification or orientation marks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual configuration and orientation detection are used for reading forms, then the reader can accurately identify forms, but the process becomes cumbersome and time-consuming
Solution Approach 1:
The imaging reader automatically performs form identification and orientation detection without requiring manual configuration. The system uses image capture and processing to self-identify form fields and determine correct orientation, eliminating the need for user intervention in configuration tasks.
Solution Approach 2:
The system pre-processes images by capturing and analyzing form fields before final reading operations. Template matching and orientation detection are performed in advance to prepare the form data for subsequent processing, streamlining the overall reading workflow.
2Measurement precision
If OCR processing with rectification is used for form reading, then form fields can be read, but the process becomes slow and inefficient
Solution Approach 1:
The system performs template matching and orientation detection as preliminary actions before OCR processing. By pre-identifying form fields and determining correct orientation through image capture and analysis, the subsequent OCR process operates more efficiently on already-oriented and located fields, reducing overall processing time.
Solution Approach 2:
The patent replaces traditional mechanical orientation detection methods (requiring physical rotation or manual alignment) with an electro-optical system that uses image capture, template matching, and automated orientation determination. This substitution eliminates time-consuming manual operations and accelerates the reading process.
3Difficulty of detecting and measuring
If orientation marks or asymmetrical margins are required for form detection, then orientation can be determined, but many forms lack these features
Solution Approach 1:
The imaging reader employs a universal template matching approach that can identify form fields and determine orientation across various form types without requiring specific orientation marks or asymmetrical margins. The system adapts to different form layouts by capturing images and matching them against stored templates, making it versatile for reading diverse form types.
Solution Approach 2:
The system replaces dependency on physical orientation marks with an electro-optical detection method using image capture and digital template matching. This allows the reader to determine orientation through image analysis rather than relying on physical form features, enabling compatibility with forms that lack traditional orientation indicators.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Streamlines the reading process by quickly and accurately identifying forms and their orientation, eliminating the need for manual configuration and OCR rectification, allowing for efficient processing of forms in any perspective without requiring orientation marks or margin detection.
Implementation Method 1
a solid-state imager (or image sensor) with a sensor array of photocells or light sensors (also known as pixels), and an imaging lens assembly for capturing return light scattered and/or reflected from the target being imaged
Data Source
AI summary
An arrangement for, and a method of, electro-optically reading forms, each form having a plurality of form fields arranged at locations relative to one another, by image capture, includes storing form templates, each template having a plurality of template fields arranged at locations relative to one another, and capturing images over a field of view. A form and a correct orientation of the form, whose image is being captured, are automatically identified by matching the locations of the form fields in the captured image of the form with the locations of the stored template fields. The form fields on the identified form in the correct orientation are thereupon processed.