OCR Region Extraction for Automated Scanned Form Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems fail to automatically output the results of optical character recognition (OCR) processing on predetermined regions of image data from scanned forms and perform image processing in response to user selection of identification information, necessitating manual data extraction from digitized documents.
Innovation Solution
A data output system that includes a communication unit to receive image data, a region identification unit to detect OCR regions based on predetermined information, and an output unit to perform OCR processing on these regions, as well as an image processing system that performs image processing on selected regions in response to user input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If forms are scanned and digitized into image data, then the forms can be stored and processed electronically, but manual data extraction from fixed regions is still required
Solution Approach 1:
The system performs OCR processing on predetermined regions in advance, before the data needs to be extracted. By pre-processing the image data to identify and process specific regions (such as date fields, expense amounts, etc.), the system eliminates the need for manual extraction later, fully automating the data extraction process and resolving the contradiction between automation extent and time loss.
2Productivity
If image processing is performed on all regions of image data, then complete data extraction is achieved, but processing time and computational resources increase
Solution Approach 1:
The system extracts only the necessary predetermined regions from the entire image data for OCR processing. Instead of processing the whole document, it identifies and processes only specific regions where data is located (such as tables, forms, or highlighted sections), thereby improving processing efficiency and reducing time loss while maintaining complete data extraction capability.
Solution Approach 2:
The system segments the image data into multiple predetermined regions based on the form structure or user-defined areas. By dividing the large image into manageable segments (e.g., date region, amount region, description region), the system can process each segment independently and efficiently, improving overall productivity while minimizing processing time through selective processing of only relevant segments.
3Loss of information
If OCR processing is performed on the entire image data, then all data is extracted, but the system becomes complex and resource-intensive
Solution Approach 1:
The system applies different processing qualities to different regions. Instead of uniformly processing the entire image, it identifies predetermined regions that contain data and applies OCR processing only to those regions. This local quality approach ensures complete data extraction from relevant areas while simplifying the overall system by avoiding unnecessary processing complexity in empty or irrelevant regions.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
An apparatus, system, and method, each of which receives image data from the device; identifies an OCR region of the image data, based on a region in which predetermined information is detected, and outputs data, the output data being obtained by performing OCR processing on the identified OCR region of the image data. An apparatus, system, and method, each of which associates a region subjected to image processing with identification information, receives, in response to a device accepting selection from a user, the identification information and image data from the device, and outputs a result of the image processing performed on the region subjected to image processing associated with the identification information. Carrier means carrying computer program code for controlling a computer system to carry out any of the above-described method is provided.