Document Scanner Boundary Detection Using Dual-Light Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing document scanners struggle to automatically crop the boundaries of documents, especially when using dark backgrounds or when the document is thick, requiring manual intervention or additional framing.
Innovation Solution
A document scanner that uses an external environment light source in conjunction with a light-emitting and non-emitting mode to capture two images, analyzing pixel channel values to automatically determine document boundaries for cropping.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a dark background element is used in the upper cover to achieve automatic boundary cropping, then the boundary detection capability is improved, but dark cards or documents cannot be correctly cropped
Solution Approach 1:
Instead of using a dark background to detect document boundaries, the patent inverts the approach by using a light background and detecting the absence of reflected light (dark areas) to identify document boundaries. This allows dark documents to be correctly detected against the light background, resolving the contradiction between boundary detection capability and cropping accuracy for dark documents.
Solution Approach 2:
The patent changes the background luminance parameter from dark to light, and adjusts the detection threshold parameters to identify document boundaries based on reflected light intensity. This parameter change enables reliable boundary detection for both light and dark documents, eliminating the limitation of dark document cropping.
2Measurement precision
If an upper cover is closed to press the document for boundary detection, then the flatness and boundary contrast are improved, but thick books cannot be scanned
Solution Approach 1:
The patent changes the background color parameter to light and adjusts the light detection parameters to identify document boundaries through reflected light intensity differences. This allows the system to detect boundaries of thick books without requiring physical compression from an upper cover, thereby maintaining adaptability to various document thicknesses while achieving boundary detection.
Solution Approach 2:
The patent replaces the mechanical compression system (upper cover pressing the document) with an optical detection system that identifies boundaries through light reflection characteristics. This substitution eliminates the need for physical contact and compression, enabling the scanning of thick books that cannot be compressed.
3Adaptability or versatility
If manual boundary cropping or custom framing is used for thick books, then the scanning of thick books is enabled, but the operation complexity and time consumption increase
Solution Approach 1:
The patent implements an automatic boundary detection and cropping function that operates without user intervention. The system automatically identifies document boundaries by analyzing reflected light patterns and performs cropping autonomously, eliminating the need for manual boundary cropping or custom framing operations. This self-service capability maintains ease of operation while enabling the scanning of thick books.
Solution Approach 2:
The patent replaces manual boundary cropping operations with an automated optical detection and image processing system. The system uses light reflection analysis to automatically identify and crop document boundaries, substituting manual mechanical operations with automated electronic processes that reduce operation complexity and time consumption.
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
Enables automatic boundary cropping without the need for a closed upper cover, effectively handling dark backgrounds and varying document thicknesses, enhancing scanning convenience and precision.
Implementation Method 1
the processor controls the light source to emit light to the document, and controls the image sensor to receive light and obtain a first image
Implementation Method 2
the image sensor to receive light and obtain a first image having first pixels. Each of the first pixels has a first visible light channel value
Data Source
AI summary
In a document scanner, a housing has a transparent platen for supporting a document, and a driving mechanism drives a light source and an image sensor to move relatively to the transparent platen. In a light-emitting mode, a processor controls the light source to emit light to the document, and controls the image sensor to receive light and obtain a first image having first pixels. In a non-emitting mode, the processor controls the light source not to emit light, and controls the image sensor to receive light and obtain a second image having second pixels, wherein the second pixels respectively correspond to the first pixels. The processor judges whether a difference between a first visible light channel value of the first pixel and a second visible light channel value of the second pixel corresponding to the first pixel is greater than a threshold value to judge the first pixel as being corresponding to a document part or a non-document part. A document scanning method is also disclosed.