Finger Occlusion Marker for Accurate Book Scanning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing book scanning systems face challenges in accurately capturing images of book pages without manual intervention, as manual page turning and pressing cause finger inclusion, leading to false detection and missed detection due to varying book contents and light conditions.
Innovation Solution
A marker with a two-sided continuous pattern of geometric primitives, such as equal-length straight-line segments and quarter circles, is used to occlude foreign objects like fingers, facilitating algorithm recognition by enhancing contrast and edge detection, and a method for image preprocessing and edge detection is employed to accurately position and recognize the marker.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual page turning and pressing is used to scan thick books or maintain page flatness, then the book page can be scanned, but finger images are inevitably captured causing false detection and missed detection
Solution Approach 1:
The detection process is segmented into multiple stages: first detecting the marker pattern on the finger, then using the marker position to guide the removal of finger regions. This segmentation allows the system to handle finger detection as a separate, controlled step rather than attempting to detect all foreign objects in one pass, thereby improving overall detection accuracy while maintaining scanning speed.
Solution Approach 2:
A marker is introduced as an intermediary element attached to the finger. This marker serves as a detectable signal that mediates between the finger (which causes detection errors) and the detection algorithm. The marker's distinct geometric pattern allows the system to identify and locate fingers reliably, enabling subsequent removal of finger images without affecting the detection of actual foreign objects on the book page.
2Device complexity
If color characteristics alone are used to recognize finger regions, then the recognition process is simple, but false detection and missed detection occur due to varying book contents and light conditions
Solution Approach 1:
The marker uses a specific color scheme (yellow geometric patterns on a blue background) that creates strong contrast with typical book page colors. This color design ensures the marker remains detectable under varying light conditions and against diverse book contents, improving recognition reliability without requiring complex algorithms.
Solution Approach 2:
The recognition approach changes from relying solely on skin color parameters to detecting geometric pattern parameters (shape, orientation, arrangement of line segments and circles). This parameter change makes the detection more robust to lighting variations and book content, as geometric patterns maintain their structural characteristics regardless of illumination conditions.
3Measurement precision
If a marker with complex geometric patterns is used to occlude foreign objects, then recognition accuracy improves, but the marker design and manufacturing becomes more complex
Solution Approach 1:
The marker pattern is segmented into simple, repeatable geometric elements (line segments and circles) that can be easily manufactured. Each element is basic in form but their specific arrangement and orientation create a unique, easily recognizable pattern. This segmentation allows for simple manufacturing while maintaining high recognition accuracy.
Solution Approach 2:
The marker employs asymmetric arrangement of geometric elements (specific orientation of line segments, non-uniform distribution of circles) that creates a distinctive pattern easy to recognize. The asymmetry ensures the marker can be reliably distinguished from symmetric patterns that might naturally occur on book pages, improving recognition accuracy without requiring complex manufacturing.
Data Source
AI summary
A marker for occluding a foreign matter in an acquired image contains a mark part whose surface is provided with a two-side continuous pattern formed by combining at least one or multiple primitives; and a fixing part which fixes a marker to a surface of foreign matter in the acquired image with the mark part to facilitate algorithm recognition and marking. The method for recognizing a foreign matter marker in an image includes the steps of performing edge detection on a planar image to acquire an edge map in the planar image; and extracting all contours in the edge map. A certain number of alternative straight-line segments are determined using an algorithm; a region is determined according to the position of each alternative straight-line segment; and finally the region of the approximate area above or below the marker is used.