Flatbed Cutter Fiducial Recognition for Distorted Sheet Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing CNC flatbed cutting machines face challenges in quickly and accurately identifying the orientation and position of graphics sheets with arbitrary locations and orientations on the work plane, especially when sheets are distorted, leading to errors in determining the correct cutting curve.
Innovation Solution
The implementation of a method and apparatus using optically readable two-dimensional codes, such as QR codes, on the graphics sheets, combined with a digital camera system and computer analysis to identify the fiducial markers, allowing for precise determination of the sheet's orientation and position, and adjustment of the cutting curve to account for distortions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If crosses or geometrical characteristics are used for recognition, then the cutting path can be determined, but the risk for error increases especially when graphics are slightly distorted
Solution Approach 1:
The patent changes the recognition parameter from geometric shape matching to fiducial marker detection. Fiducial markers have specific geometric patterns (squares, circles, or other distinct shapes) with known positions and orientations, allowing the system to determine sheet position and orientation through marker detection rather than complex graphic recognition, thereby maintaining accuracy even when graphics are distorted
Solution Approach 2:
The patent introduces fiducial markers as intermediary reference elements between the sheet and the recognition system. These markers serve as stable, easily detectable intermediaries that bridge the gap between the physical sheet and the digital cutting path determination, providing reliable position and orientation information without being affected by graphic distortion
2Measurement precision
If a double camera system is used, then position and orientation can be determined, but the computing capacity required is substantial
Solution Approach 1:
The patent extracts the position and orientation information from the fiducial markers themselves, which contain embedded directional indicators (such as asymmetric patterns or oriented features). This allows the system to determine orientation directly from the marker geometry without requiring complex image processing or database matching, significantly reducing computing capacity requirements
Solution Approach 2:
The fiducial markers are pre-designed with known geometric characteristics and orientations. The system performs preliminary encoding of position and orientation information into the marker structures themselves, so that during operation, the camera only needs to detect and decode these pre-encoded features rather than performing complex analysis, reducing real-time computing requirements
3Reliability
If QR codes are used for identification, then the graphics can be identified, but the orientation and position on the work plane cannot be determined
Solution Approach 1:
The patent merges the identification function (QR code) with the orientation and position determination function (fiducial marker) into a single integrated element. The fiducial marker incorporates both the QR code for graphic identification and geometric features for position and orientation determination, eliminating the need for separate markers and enabling simultaneous extraction of all required information from a single detection process
Data Source
Figure 1~2a
Figure 2b~3b
Figure 3c~4
AI summary
An apparatus comprising a flatbed cutting machine with a work plane (2) with an upper surface (3) for receiving the printed sheets (4); the machine further comprising an op- erating group (5) mobile along the work plane (2) and comprising a cutting member (6) for cutting the sheets (4). The apparatus comprises a first camera (9) above the work plane (2) for imaging the upper surface and a computer system (8) functionally connected to the first camera (9) and configured for receiving digital images from the first camera (9). The computer system (8) analyses received images with respect to image data for a fiducial (15); the fiducial (15) comprising an optically readable two- dimensional code for a numeric or alphanumeric sequence. The computer system is decoding the two-dimensional code to extract an ID code that identifies the graphics on the sheet (4).On the basis of stored data in the database, related to the ID code, a cutting curve is determined and corresponding computer instructions submitted to the machine for moving the operating group (5) with the cutting element (6) along the work plane (2) for cutting the sheet (4) along the cutting curve.