Document Reading Unit Page Number Extraction for Overlapped Feeding Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image-forming apparatuses face issues with determining whether document reading has ended properly and detecting overlapped feeding, especially when the number of documents is unknown, leading to incomplete reading and difficulty in identifying the location of overlapped feeding.
Innovation Solution
The apparatus includes an extraction-area-specifying unit, a number-image-extracting unit, and a number-image-summarizing unit that automatically specify and extract number images from the top and bottom margins of document image data, generating a number-verification page to summarize these images, allowing for verification of proper reading completion and identification of overlapped feeding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the document reading unit reads documents fed by the document feeding unit, then document image data is obtained, but overlapped feeding may occur causing documents to be fed while being stacked, leading to incomplete reading
Solution Approach 1:
The patent extracts the page number information from the document images and separates it for verification purposes. The verification image generation unit creates a separate verification image that contains only the page numbers, allowing independent verification of reading completeness without re-examining the entire document content. This extraction approach resolves the contradiction by providing a simple verification mechanism that doesn't require complex analysis of the full document set.
Solution Approach 2:
The patent implements a feedback mechanism where the verification image is compared against the expected page number sequence. The comparison result provides feedback about whether overlapped feeding occurred, enabling automatic detection and correction. This feedback loop resolves the technical contradiction by providing a reliable method to detect reading errors without requiring complex real-time monitoring during the feeding process.
2Adaptability or versatility
If the number of documents is unknown, then flexibility in processing is improved, but the ability to determine proper reading completion deteriorates
Solution Approach 1:
The patent enables the document set itself to provide verification information through its page numbers. Each document contains its own identification (page number) that can be used to verify reading completion. This self-service approach resolves the contradiction by allowing the system to verify reading completeness using information inherently present in the documents, without requiring external documentation of the total document count.
Solution Approach 2:
The patent replaces manual counting or mechanical tracking of document numbers with automated image recognition and comparison. The verification image generation unit automatically extracts page numbers from images and compares them against the expected sequence, substituting mechanical verification methods with automated image processing. This resolves the contradiction by providing precise verification capability that works regardless of whether the total document count is known in advance.
3Reliability
If traditional methods are used to detect overlapped feeding, then detection capability is maintained, but ease of identifying the location of overlapped feeding deteriorates
Solution Approach 1:
The patent segments the verification process into distinct components: page number extraction, sequence comparison, and error location identification. The verification image clearly shows which page numbers are missing or out of sequence, making it easy to identify the specific location of overlapped feeding. This segmentation resolves the contradiction by providing both reliable detection and easy location identification through structured presentation of verification data.
Solution Approach 2:
The patent uses visual differentiation in the verification image to highlight anomalies. Page numbers that are missing or out of sequence can be visually identified through their absence or positional anomalies in the verification image sequence. This visual approach resolves the contradiction by making overlapped feeding locations immediately apparent through the structured verification output, eliminating the need for complex analysis.
Data Source
AI summary
Provided is an image-forming apparatus that is able to determine whether or not reading of a document ends properly without knowing the number of sheets of the document, and when overlapped feeding occurs, is able to estimate the location where overlapped feeding occurs. The image-forming apparatus has a document feeding unit that conveys a set document one sheet at a time, and a document reading unit that reads the conveyed document and obtains document image data. The image-forming apparatus includes an extraction-area-specifying unit that specifies part or all of the top and bottom margins of document image data as extraction areas, a number-image-extracting unit that respectively extracts the images in the extraction areas specified by the extraction-area-specifying unit as number images, and a number-image-summarizing unit that generates a number-verification page on which plural number images are summarized.