Image Processing Apparatus for Accurate Code Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for reading code images, such as barcodes, on sheets printed by multiple sensors can result in pixel shifts during image combination, leading to incorrect reading and failure in extracting embedded information.
Innovation Solution
An image processing apparatus that acquires positional information of the code image, combines read images from multiple sensors, and determines which image includes the entire code image, allowing for accurate extraction of the information by aligning and combining the images without pixel shifts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple read sensors are used to read a sheet wider than the maximum sheet width, then the read range is extended, but pixel shifts occur at the boundary of combined images causing code reading failures
Solution Approach 1:
The patent divides the code image into multiple portions, with each read sensor responsible for reading a specific portion. By segmenting the code image and assigning each segment to a dedicated sensor, the system avoids combining multiple images and thus prevents pixel shifts at boundaries while maintaining the ability to read codes across the entire sheet width.
2Area of stationary object
If images from multiple read sensors are combined to cover the entire sheet width, then the coverage area is increased, but pixel shifts occur at combination boundaries leading to information loss
Solution Approach 1:
Instead of combining images from multiple sensors, the patent segments the code image and assigns each segment to a specific sensor. This eliminates the image combination step that causes pixel shifts and information loss, while still achieving full sheet width coverage through coordinated reading of multiple segments.
3Device complexity
If a single read sensor is used, then the system complexity is reduced, but the read range is limited and cannot cover sheets wider than the sensor's maximum width
Solution Approach 1:
The patent uses multiple read sensors, each reading a specific portion of the code image, rather than using a single sensor with extended range. This segmentation approach allows the system to cover sheets wider than any individual sensor's maximum width while keeping each sensor's function simple and well-defined.
Solution Approach 2:
The patent merges the reading results from multiple sensors to reconstruct the complete code image. By combining the segmented code portions read by different sensors in a coordinated manner, the system achieves extended read range while maintaining manageable device complexity through modular sensor integration.
Data Source
AI summary
An image processing apparatus for inspection of a sheet on which a code image obtained by encoding predetermined information is printed includes: a first acquisition unit which acquires positional information indicating a position and a size of the code image in the sheet; a second acquisition unit which acquires a plurality of read images obtained by a plurality of reading units reading different portions of the sheet, respectively, wherein the read images become an image indicating the entire sheet by being combined; a determination unit which determines whether any one of the read images includes the entire code image based on the positional information; and an extraction unit which extracts the predetermined information by, in a case where there is a read image including the entire code image, reading the code image in that read image.