Sheet Image Acquisition Using Thickness Data for Transparent Edge Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sheet image acquisition systems struggle to accurately extract the shape of sheets with transparent portions, leading to incorrect recognition and processing issues, such as assimilation of transparent regions into the background and miscomputation of edge points, which can result in rejection and miscalculation of sheet size and orientation.
Innovation Solution
A sheet image acquisition device comprising a thickness detection sensor and an optical line sensor, with edge detectors and a computed point corrector, where the thickness sensor collects data in both main and sub-scanning directions, and the optical line sensor collects image data, allowing for precise edge detection and correction of edge points to accurately extract the sheet shape.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an optical line sensor is used to capture sheet images, then the sheet can be processed for recognition, but transparent portions cause light transmission that leads to incorrect shape extraction and edge detection
Solution Approach 1:
The patent segments the shape extraction process into two independent parts: optical line sensor data for overall sheet capture and thickness detection sensor data for edge boundary determination. This segmentation allows each sensor to contribute its strength while avoiding the weakness of the other, particularly for transparent portions where the optical sensor fails but the thickness sensor succeeds.
Solution Approach 2:
The thickness detection sensor acts as an intermediary that provides complementary information about sheet edges. By introducing this additional detection mechanism, the system overcomes the limitation of the optical line sensor in detecting transparent portions, as the thickness sensor can detect physical presence regardless of optical transparency.
2Reliability
If multiple transport path sensors are used to detect sheet passage, then transport monitoring is improved, but transparent portions cause false detection of abnormal transport
Solution Approach 1:
The patent merges the detection functions of multiple sensor types (optical line sensor and thickness detection sensor) to achieve reliable transport monitoring. By combining their outputs, the system can distinguish between actual transport events and false signals caused by transparent portions, improving both reliability and precision simultaneously.
3Device complexity
If only optical line sensor data is used for shape extraction, then the processing is simple, but transparent regions are assimilated into the background causing incorrect edge computation
Solution Approach 1:
The patent introduces a new dimension of detection by adding thickness measurement capability to the existing optical imaging system. This additional dimensional information (physical thickness vs. optical reflectance) enables the system to correctly identify sheet boundaries even when optical properties are misleading, without significantly increasing overall system complexity.
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
This solution improves the precision of extracting the shape of sheets with transparent portions, reducing rejection rates and miscalculations by correctly identifying the sheet's outline and size, even in the presence of transparent regions.
Implementation Method 1
an optical line sensor configured to collect image data of the entire surface of the sheet
Implementation Method 2
The light, such as infrared light, emitted from a light source of the optical line sensor passes through a transparent portion of a sheet
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present invention provides a sheet image acquisition device, a sheet recognition device, a sheet handling device, and a sheet image acquisition method capable of improving the precision of extracting the shape of a sheet having a transparent portion. The sheet image acquisition device of the present invention for acquiring an image of a sheet includes a thickness detection sensor configured to collect thickness data of the sheet and an optical line sensor configured to collect image data of the entire surface of the sheet. The sheet image acquisition device is configured to correct outline information of the sheet derived from the image data based on outline information of the sheet derived from the thickness data.