Dual-Imaging Medium Conveyance for Shape Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Medium conveying apparatuses face challenges in accurately distinguishing between a plastic card and multi-feed conditions due to similar ultrasonic wave reception levels, leading to potential misclassification and incorrect image processing.
Innovation Solution
The apparatus employs dual imaging devices to capture images of both surfaces of the medium, detects edge pixels, and compares gradation values to select the appropriate surface for shape estimation, allowing for accurate medium shape determination and image cutting based on the estimated shape.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ultrasonic wave reception is used to detect multi-feed, then multi-feed detection function is provided, but plastic cards are mistakenly identified as multi-feed due to similar reception levels
Solution Approach 1:
The patent introduces an imaging device as an intermediary to capture images of the medium, which then serves as a basis for shape estimation. This intermediary approach allows the system to differentiate between plastic cards and multi-feed conditions by analyzing edge pixels and gradation values in the captured images, thereby improving detection accuracy without adding complex specialized sensors.
Solution Approach 2:
The patent replaces the ultrasonic wave-based detection system with an optical imaging-based system. Instead of using ultrasonic transmitters and receivers to detect medium characteristics, the system uses imaging devices to capture visual information, processes edge pixels and gradation values, and estimates medium shape. This substitution eliminates the misclassification issue while maintaining cost-effectiveness by using standard imaging components.
2Measurement precision
If imaging devices are added to capture both surfaces of the medium, then shape estimation accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments the imaging task into two parts: capturing images of both surfaces of the medium. By obtaining images from both surfaces, the system can compare edge pixels and gradation values to more accurately estimate the medium shape, particularly for curved or warped media. This segmentation approach improves measurement precision without requiring overly complex single-surface imaging systems.
Solution Approach 2:
The imaging devices serve multiple functions: they capture images for shape estimation, provide data for multi-feed detection, and enable medium surface characterization. This multi-functionality reduces the need for separate specialized sensors and processing systems, thereby limiting the increase in device complexity while achieving improved shape estimation accuracy.
3Measurement precision
If edge pixel detection and gradation value comparison are performed, then medium shape estimation is improved, but processing time increases
Solution Approach 1:
The patent extracts only the critical features from the captured images: edge pixels and gradation values. Instead of processing the entire image data, the system identifies and analyzes specifically the edge pixels and their gradation values, which are sufficient for shape estimation. This extraction approach maintains high measurement precision while significantly reducing processing time compared to full-image analysis.
Data Source
AI summary
A medium conveying apparatus includes a processor to detect edge pixel from a first line image and a second line image, select a first surface or a second surface of the medium, based on the number of the detected edge pixels, a comparison between a gradation value of each of the detected edge pixels and a gradation value of a peripheral pixel of each of the detected edge pixels, or a variation of gradation values of pixels within a predetermined distance from each of the detected edge pixels or the peripheral pixel of each of the detected edge pixels, estimate a shape of the medium based on a line image acquired by imaging the selected surface, and control a conveying roller or cut out a medium image from the first line image and the second line image and output the medium image, based on the estimated shape.


