Multi-Crop Image Processing Edge Detection via Dual White Reference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing document scanning technologies face challenges in accurately extracting the image area of documents, especially for irregularly sized documents like receipts, due to difficulties in distinguishing the document edge from the scanner table cover, leading to user discomfort and complex procedures.
Innovation Solution
An image processing system that performs crop processing by detecting document edges in a first image with a higher white reference value and cutting out the image area using a second image with a lower white reference value, allowing for accurate extraction without requiring users to perform complex procedures or experience discomfort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the document table cover is closed during scanning, then the document is properly pressed and scanned, but the white portions of the cover and document have the same signal value making edge extraction difficult
Solution Approach 1:
The patent segments the scanning process into two distinct phases: a first scan with the document table cover closed to obtain the document image, and a second scan with the cover open to obtain a reference image where the cover area is differentiated. This segmentation allows edge extraction without requiring users to perform unusual operations.
Solution Approach 2:
The patent changes the scanning parameter (document table cover state) between two scans. The first scan is performed with the cover closed to capture the document, while the second scan is performed with the cover open to create a reference image where the cover area has different signal characteristics, enabling accurate edge extraction.
2Measurement precision
If the document table cover is left open during scanning, then document edge extraction becomes easier, but the scan light enters the user's eyes causing discomfort
Solution Approach 1:
The patent divides the scanning operation into two separate scans performed at different times. The first scan with the cover closed protects the user from light, while the second scan with the cover open is used only for reference image acquisition, minimizing light exposure to the user.
Solution Approach 2:
The patent performs the first scan with the document table cover closed before performing the second scan with the cover open. This preliminary action captures the document image under safe conditions, and the subsequent reference scan with the cover open is used only for edge detection, reducing overall light exposure to the user.
3Measurement precision
If a black sheet is used to cover the document table, then document edge extraction becomes easier, but the user must perform complicated procedures
Solution Approach 1:
The patent uses the document table cover itself as the reference object for edge extraction, eliminating the need for additional black sheets or external reference materials. The cover's own structure is utilized to create the reference image, simplifying the overall procedure.
Solution Approach 2:
The patent makes the document table cover serve dual purposes: it functions as both the pressing mechanism for the document during the first scan and as the reference object for edge extraction during the second scan. This multi-functionality eliminates the need for separate reference materials.
4Measurement precision
If normal white reference value is used for scanning, then the image quality is good, but the white portions of the cover and document cannot be distinguished
Solution Approach 1:
The patent uses different white reference values for two separate scans: a first white reference value for the document scan and a second white reference value for the reference image scan. This segmentation allows the system to optimize each scan for its specific purpose while maintaining overall image quality.
Solution Approach 2:
The patent changes the white reference value parameter between scans. By using a first white reference value for the document scan and a second white reference value for the reference scan, the system creates distinguishable signal values for white portions in different contexts, enabling accurate edge extraction while preserving image quality.
Data Source
AI summary
To make it possible to cut out an image area corresponding to a document with a high accuracy from an image obtained by scanning a document without forcing a user to perform complicated work or without giving a feeling of discomfort at the time of performing multi-crop processing. To this end, first, position information on the document is acquired by detecting an edge component from a first image obtained by setting high the first white reference value. Then, an image area corresponding to the document is cut out based on the acquired position information from a second image obtained by setting the second white reference value lower than the first white reference value.