Smartphone Document Imaging Shadow Removal via Multi-Angle Brightness Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for reading document images using smartphones struggle with capturing high-quality images without shadows, as they lack a stable light source and are prone to light source instability and shadow generation.
Innovation Solution
An image reading apparatus comprising an imaging unit, an image data analysis unit, and an image combination unit that captures document images multiple times from different angles, analyzes and compares brightness levels to select the brightest regions, and generates a composite image by combining selected data images, thereby minimizing shadow presence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a smartphone is used to capture document images, then portability and ease of operation are improved, but image quality deteriorates due to shadow generation and light source instability
Solution Approach 1:
The patent divides the image correction process into multiple segments: capturing multiple images from different angles, segmenting the images into regions, comparing brightness in each region, and selectively combining regions. This segmentation approach allows the system to handle shadow removal and quality improvement without requiring complex hardware modifications to the smartphone.
Solution Approach 2:
The patent introduces an intermediary processing system that acts as a mediator between the smartphone camera and the final document image. This intermediary system captures multiple angles, performs brightness comparison, and synthesizes the final image, thereby compensating for the smartphone's inherent limitations in providing stable illumination.
2Reliability
If multiple images are captured from different angles, then shadow coverage is improved, but processing complexity increases
Solution Approach 1:
The patent segments the image processing task by dividing captured images into multiple regions and independently analyzing brightness in each region. This segmentation simplifies the overall processing complexity by breaking down the complex task of shadow removal into manageable regional comparisons, while still achieving comprehensive shadow coverage through multi-angle capture.
Solution Approach 2:
The patent changes the parameter of image capture by acquiring images from multiple different angles. This parameter change ensures that shadows cast from one angle are not present in images from other angles, thereby improving shadow coverage. The subsequent processing uses brightness comparison to selectively combine regions, managing the complexity introduced by multiple captures.
3Manufacturing precision
If brightness comparison is performed on a per-region basis, then image quality is improved by selecting brightest regions, but processing time increases
Solution Approach 1:
The patent performs brightness comparison on a per-region basis rather than processing entire images at once. This segmentation approach improves image quality by enabling selective combination of the brightest regions from multiple captures, while reducing processing time by limiting the comparison scope to individual regions rather than complete images.
Solution Approach 2:
The patent applies partial action by performing brightness comparison only on specific regions of interest rather than analyzing every pixel in the entire image. This partial processing approach maintains high image quality in critical regions while reducing overall processing time and computational resources required.
Data Source
AI summary
An image reading apparatus includes an imaging unit, an image data analysis unit, and an image combination unit. The imaging unit images the document image multiple times from mutually differing angles to generate a plurality of data images each representing the document image. The image data analysis unit performs matching on the plurality of data images by matching the plurality of data images, on a per-region basis in each of the plurality of images of the document represented by the plurality of data images, so as to obtain per-region brightnesses for the plurality of data images, and comparing the obtained brightnesses among the plurality of data images to select from among the plurality of data images each data image whose region is comparatively brighter. The image compositing unit uses the data images selected on the per-region basis as comparatively bright to generate a composite data image.